RAB27A, RAB27B and VPS36 are downregulated in advanced prostate cancer and show functional relevance in prostate cancer cells

  • Authors:
    • Thomas Stefan Worst
    • Yannic Meyer
    • Maria Gottschalt
    • Cleo-Aron Weis
    • Jost von Hardenberg
    • Christine Frank
    • Annette Steidler
    • Maurice Stephan Michel
    • Philipp Erben
  • View Affiliations

  • Published online on: February 10, 2017     https://doi.org/10.3892/ijo.2017.3872
  • Pages: 920-932
Metrics: Total Views: 0 (Spandidos Publications: | PMC Statistics: )
Total PDF Downloads: 0 (Spandidos Publications: | PMC Statistics: )


Abstract

Paracrine and long-range signaling via extracellular vesicles, such as exosomes and microvesicles, is deemed crucial for tumorigenesis, invasion and spread of solid tumors. The ESCRT machinery (endosomal sorting complexes required for transport) and Rab-proteins act as key players in vesicular trafficking and secretion. Yet, their role in prostate cancer (PCa) is unknown. Therefore, this study aimed to elucidate the relevance of these components in PCa. In silico reanalysis of genes with known involvement in vesicular trafficking and secretion in an existing microarray dataset revealed low expression of RAB27A, RAB27B and VPS36 to be predictive for reduced BCR-free survival in patients with localized PCa (p=0.033, 0.025 and 0.005). In the same microarray dataset underexpression of RAB27A, RAB27B and VPS36 was seen in distant metastases (p<0.001; p=0.003; p<0.001). This was consistent in two further microarray datasets. qRT-PCR-validation in two independent cohorts of PCa specimens (n=90) showed low expression of VPS36 in PCa tissue (p=0.023), especially in castration-resistant tumors (p=0.002). In all five datasets there were significant correlations between the expression of at least two of the candidates. Upon knockdown of VPS36 an increase of RAB27A and RAB27B expression, but not vice versa, was observed in both prostate and breast cancer cells (PC3, MDA-MB‑231). In PC3 cell knockdown of RAB27B and VPS36 dramatically reduced colony formation (-52.2%, p<0.001; -71.1%, p<0.001) and, controversial to reports in other tumor entities, increased the release of extracellular particles (+25.3%, p=0.014; +45.6%, p<0.001). Taken together RAB27A, RAB27B and VPS36 are frequently underexpressed in advanced PCa and are inversely correlated with PCa outcome. There seems to be a close relationship in the expression of RAB27A, RAB27B and VPS36, with RAB27A and RAB27B being dependent on VPS36. Changes in colony formation and particle release upon RNAi indicate an involvement in paracrine cell-cell communication.

Introduction

Prostate cancer (PCa) is the most common solid cancer entity among men in industrialized countries (1). Annually more than 250,000 men die from metastasized PCa worldwide. However, many aspects in PCa development and progression are still unknown.

In many different cancer entities signaling via extracellular particles has impact on different tumor-related processes (2,3). Especially extracellular vesicles are in the focus of current research. They play a role in modulation of tumor microenvironment (4), neo-angiogenesis (5), metastatic niche formation (6) and systemic inflammation (7). PCa extracellular vesicles are involved in crosstalk between tumor cells and the surrounding stroma (8,9).

Exosomes are the best-studied subpopulation of extracellular vesicles. They have a typical size of 30–100 nm and, unlike the larger microvesicles which directly originate from the plasma membrane, are generated from the membranes of endosomes. These endosomes are termed as multi-vesicular-bodies (MVB). Exosome generation is mainly mediated by the components of the ESCRT machinery (endosomal sorting complexes required for transport), with tetraspanin-rich microdomains and lipids like sphingosin-1-phosphate also being involved in these processes (1012). MVBs release exosomes into the extracellular space by fusion with the plasma membrane (12). Based on current knowledge, fusion with the plasma membrane is mainly mediated by the GTPases RAB27A and RAB27B (13). Further proteins such as Plectin and Rab11 are also involved in this process (12). Extracellular vesicles are rich in proteins involved in membrane transportation and membrane fusion (GTPases, annexins, flotillins), tetraspanins like CD9, CD63 and CD81, heat shock proteins HSP70 and HSP90 and proteins involved in MVB formation such as ALIX and TSG101 (14,15). Compared to microvesicles, the composition of exosomes is more stringently controlled by specific processes (1618).

Many studies deal with the function of extracellular vesicles and their potential use as biomarkers. However, not much is known about the role of genes involved in vesicular trafficking and secretion, in cancer. Therefore, it was the aim of this study to elucidate their relevance in PCa. Their expression was re-analyzed in existing microarray data with regard to PCa patient outcome prediction. qRT-PCR profiling of PCa tissue samples at various stages served as validation. In vitro studies were done to functionally examine the impact of the genes associated with patient outcome on proliferation, colony formation and release of extracellular particles.

Materials and methods

Candidate list generation and in silico analyses

Based on recent literature a list of genes with known functions in vesicular trafficking and secretion, like the ESCRT components TSG101, MVB12, VPS36 and PDCD6IP (ALIX), and of typical extracellular vesicle markers (e.g., flotillins and the tetraspanins CD9, CD63 and CD81) was assembled. Furthermore, genes typically associated with PCa, such as KLK3 (PSA) or PTEN were added. The full list of genes is given in Table I. RNA expression data and clinical and pathological parameters of a PCa dataset (19), comprising of 131 primary (mean age, 58.03±6.99 years; mean PSA, 8.48±12.28 ng/ml; T2, 85; T3, 40; T4, 6) and 19 metastatic tumor samples (mean age, 60.51±7.38 years; mean PSA, 72.99±135.59 ng/ml) were derived from cBioPortal (20). Gene expression in tumor samples is compared to healthy controls and adjacent tumor-free tissue.

Table I

A list of genes identified from the literature with known function in exosome biogenesis and secretion and known PCa biomarkers was assembled for further in silico analyses.a

Table I

A list of genes identified from the literature with known function in exosome biogenesis and secretion and known PCa biomarkers was assembled for further in silico analyses.a

GroupGene symbolGene name
Known PCa markersAMACRα-methylacyl-CoA racemase
ARAndrogen receptor
EZH2Enhancer of zeste 2 polycomb repressive complex 2 subunit
FOLH1 (PSMA)Folate hydroxylase 1 (prostate specific membrane antigen)
KLK3 (PSA)Kallikrein 3 (prostate specific antigen)
PCA3Prostate cancer antigen 3
PTENPhosphatase and tensin homolog
ESCRT-0STAM1Signal transducing adaptor molecule 1
STAM2Signal transducing adaptor molecule 2
VPS27Vacuolar protein sorting 27 homolog
ESCRT-IMVB12Multivesicular body subunit 12
PDCD6IPProgrammed cell death 6 interacting protein
TSG101Tumor susceptibility gene 101
UBAP1Ubiquitin associated protein 1
VPS28VPS28, ESCRT-I subunit
VPS37VPS37, ESCRT-I subunit
ESCR-IISNF8 (EAP30)SNF8, ESCRT-II complex subunit
VPS25 (EAP20)Vacuolar protein sorting 25 homolog
VPS36 (EAP45)Vacuolar protein sorting 36 homolog
ESCRT-IIICHMP6Charged multivesicular body protein 6
IST1IST1, ESCRT-III associated factor
VPS4AVacuolar protein sorting 4 homolog A
VTA1 (LIP5)Vesicle trafficking 1
Rab-kinasesRAB5aRAB5A, member RAS oncogene family
Rab11RAB11, member RAS oncogene family
RAB27ARAB27A, member RAS oncogene family
RAB27BRAB27B, member RAS oncogene family
Rab35RAB35, member RAS oncogene family
RAB36RAB36, member RAS oncogene family
TBC1D10A (EPI64)TBC1 domain family member 10A
Exosome-associated genesCD9CD9 molecule
CD63CD63 molecule
CD81CD81 molecule
CHMP4CCharged multivesicular body protein 4C
DCT (TYRP2)Dopachrome tautomerase
EPCAMEpithelial cell adhesion molecule
FLOT1Flotillin 1
FLOT2Flotillin 2
HSP4AHeat shock protein family A (Hsp70) member 4
HSP90AA1Heat shock protein 90 α family class A member 1
MMP2Matrix metallopeptidase 2
MMP9Matrix metallopeptidase 9
MMP14Matrix metallopeptidase 14
STEAP3STEAP3 metalloreductase

a Genes printed in non-bold letters were omitted from further analyses as they were not available in the analyzed dataset.

The expression z-scores in patients with localized PCa, obtained from the original data, were used to define expression cut-off values (high vs. low expression) for the event of biochemical recurrence (BCR). Calculations were done using the partition test in SAS JMP 11 (SAS Institute, Cary, NC, USA). These cut-off values were applied to the cohort and correlated with BCR-free survival using log-rank test. Only candidates with >20 patients in the high and low expression groups were followed-up. BCR-free survival was displayed using Kaplan-Meier graphs. In candidates reaching significance for prediction of BCR, expression values were stratified according to T-stage and Gleason grade. Correlation with clinical and pathological parameters was additionally done using two further microarray datasets (Tomlins et al and Yu et al) (21,22). In all three datasets correlations between the expression of the different candidate genes were done using Spearman correlation.

Cohorts and patient samples

A cDNA array (Origene, Rockville, MD, USA) consisting of 40 localized PCa and 8 benign control samples was used as a first step to independently re-test the expression of RAB27A, RAB27B and VPS36. Patient characteristics of this cohort are shown in Table II. Expression of all three candidates was correlated with T-stage and Gleason grade.

Table II

Patient characteristics in the Origene cDNA array.

Table II

Patient characteristics in the Origene cDNA array.

PatientsLocalized PCa (n=40)BPH (n=8)
Mean age (SD)62.75±8.1564.15±10.9
T1
T222
T312
T4
Not specified6
Gleason 52
Gleason 68
Gleason 7a14
Gleason 7b8
Gleason 83
Gleason 94
Not specified1

In a second step a cohort of 41 patients who underwent transurethral resection of PCa for palliation in the Department of Urology of the Mannheim Medical Center between May 2005 and September 2015 was retrospectively analyzed. Nine patients were treated twice with transurethral resection, resulting in a total of 50 FFPE tumor samples. FFPE prostate tissue specimen of ten patients who underwent transurethral resection due to benign prostatic hyperplasia (BPH), histologically negative for PCa, served as controls. Patient data of this cohort is given in Table III. All experiments were in accordance with the institutional ethics review board (Medical Ethics Committee II of the Medical Faculty Mannheim, ethics approval 2013-845R-MA). Expression data were correlated with tumor stage and Gleason grade in both datasets. In tumor samples obtained from transurethral resection further correlations with castration resistance were done.

Table III

Patient characteristics in the TUR cohort.

Table III

Patient characteristics in the TUR cohort.

Patients
Samplesa
PCa (n=41)
PCa (n=50)
BPH (n=10)
BPH (n=10)
Mean age (SD)75.48±6.8666.9±13.56
 Hormone-sensitive74.43±6.13
 Castration-resistant75.89±7.15
Hormone-sensitive14
 Gleason 5:2
 Gleason 63
 Gleason 74
 Gleason 84
 Gleason 91
Castration-resistant36

a Age and tumor characteristics are given with regard to the number of samples.

RNA-extraction and cDNA-synthesis

Tumor-bearing and tumor-free FFPE prostate tissue specimen of patients from our institution were sectioned and stained with hematoxylin and eosin and reviewed microscopically. Subsequent unstained 10-μm sections were placed on glass slides. Tissue sections of control patients and those of PCa patients bearing ≥80% of tumor tissue were used completely. On sections containing <80% of tumor, non-tumorous areas were scratched off prior to RNA extraction using a sterile scalpel. Sections were deparaffinized three times in 2 ml of Neo Clear (Merck Millipore, Billerica, MA, USA). The deparaffinized tissue was then scraped off the glass slides into a reaction tube containing 150 μl of lysis buffer. RNA extraction was conducted with the XTRACT FFPE kit (Stratifyer, Cologne, Germany) as suggested by the manufacturer (23,24). In brief the tubes were incubated 30 min at 80°C with shaking. After cooling to 65°C 50 μl of Proteinase K (Promega, Fitchburg, WI, USA) were added and incubated for 30 min with shaking. Subsequently 800 μl of MagiX-RNA buffer and 40 μl of MagiX-RNA ferromagnetic beads were added and incubated 15 min at room temperature with shaking. The tubes were placed on a magnetic rack and the supernatant was removed. Three washing steps were subsequently conducted, after each washing, supernatant was removed with the tubes on the magnetic rack. The RNA was eluted by adding 100 μl of elution buffer which was incubated for 15 min at 70°C. With the tube on the magnetic rack the supernatant, containing the RNA, was transferred to a new reaction tube and either used immediately or stored at −80°C. RNA-concentration was determined spectrophotometrically using the NanoDrop 1000 (Thermo Fisher Scientific, Waltham, MA, USA).

cDNA synthesis was performed using the SuperScript III reverse transcription kit (Thermo Fisher Scientific) with sequence specific primers. In brief 2.5 μl of forward primers of RAB27A, RAB27B, VPS36, androgen receptor (AR) or the house keeping gene calmodulin 2 (CALM2), 1 μl of 10 mM dNTP Mix and 3.5 μl of nuclease-free H2O were added to 5 μl of RNA template. This mixture was incubated for 10 min at 65°C and was subsequently cooled on ice. For cDNA synthesis 4 μl of First Strand synthesis buffer, 1 μl of 0.1 M DDT, 1 μl of RNAseOut (Thermo Fisher Scientific) and 2 μl of SuperSript III reverse transcriptase were added and incubated for 120 min at 70°C. cDNA was immediately used for qRT-PCR or stored at −20°C.

qRT-PCR analyses in patient samples

The expression of RAB27A, RAB27B and VPS36 was determined in all samples of the cDNA array in relation to the housekeeping gene CALM2. For all assays intron spanning primer pairs were designed using Primer-BLAST, based on the primer3 algorithm (25).

In brief, 10 μl of TaqMan Fast Universal PCR Mastermix (Life Technologies, Darmstadt, Germany), 0.75 μl of forward and reverse primer, each (300 nM), 0.5 μl of PCR probe (200 nM) (both MWG Eurofins, Ebersberg, Germany) and 6 μl of nuclease-free H2O were added to 2 μl of cDNA template. Subsequently 50 cycles of PCR amplification with 3 sec of 95°C and 30 sec of 60°C were conducted in a StepOnePlus™ System (Thermo Fisher Scientific). Relative candidate gene expression normalized to the expression of CALM2 was calculated in reference to the average expression in non-tumorous samples using the 2−ΔΔCT-method (26). Expression of each gene was correlated with clinical data. Correlations between the expression of different genes were done using Spearman correlation.

Cell lines, siRNA knockdown and qRT-PCR of in vitro samples

Human PC3 metastatic PCa cells and MDA-MB-231 breast cancer cells, which is a well-studied model in case of extracellular vesicle release, were obtained from ATCC (Wesel, Germany) and grown under standard conditions either in DMEM (PC3) or RPMI (MDA-MB-231) (each from Life Technologies) supplemented with 10% FCS (Sigma-Aldrich, St. Louis, MO, USA) and 2 nM glutamine (Life Technologies).

Cell lines were transfected with siGenome pooled and individual siRNAs against RAB27A, RAB27B and VPS36 at a concentration of 30 nmol using Dharmafect I transfection reagent (all from Dharmacon, Lafayette, USA). Dharmacon non-targeting siRNA served as negative control. Transfection was conducted as recommended by the manufacturer with minor modifications. While cells were detached, harvested, spun down and diluted to the desired concentration, siRNA and Dharmafect I were separately incubated in pure RPMI (Life Technologies) for 10 min at room temperature and where then mixed 1:1 for subsequent 30 min incubation at room temperature. Hereafter cell suspension was added to the transfection mix 3:1 and incubated at 37°C. The supernatant was replaced with fresh medium after 24 h.

RNA-extraction was performed after further 48 h using RNeasy Mini kit (Qiagen, Hilden, Germany) according to the recommendations of the manufacturer. For cNDA-synthesis 40 μl of diluted RNA were mixed with 4 μl of 5 mg/ml pdN6 random primers, 4 μl of 10 mM dNTP Mix, 16 μl of 5X M-MLV buffer, 8 μl of 0.1 M RNase inhibitor, 4 μl of 0.1 M DTT and 4 μl of M-MLV reverse transcriptase (all from Roche Diagnostics). After an incubation for 2 h at 37°C and a deactivation step of 5 min at 65°C, cDNA was directly used for qRT-PCR or stored at −20°C. qRT-PCR was performed and analyzed analogously to patient samples.

qRT-PCR was conducted to validate knockdown efficiency and to monitor the impact of the knockdown of one candidated gene on the other two. PCR experiments and functional assays were conducted in three biological replicates each.

Proliferation assay

PC3 cells were seeded and transfected following the protocol described above in 96-well plates with a total volume of 100 μl/well. After 24 h the supernatant was replaced by 100 μl of fresh growth medium. After further 24, 48 and 72 h of incubation, 10 μl of MTT-reagent (Promega) was added per well and incubated for 3 h at 37°C. Absorption measurement at 570 nm was performed using an Infinite M1000 Pro plate reader (Tecan, Männerdorf, Switzerland).

Scratch assay

Using the same transfection protocol, PC3 cells were seeded in 24-well plates with a volume of 1 ml per well. The medium was changed 24 h after transfection. Again 24 h later a defined scratch was introduced in the center of the well, using a sterile 200 μl pipette tip. The medium was changed again. At this point of time and after further 24, 48 and 72 h the scratch was photographed at a ×20 magnification. The calculation of the cell-free space in the scratch area was performed with the open source software tscratch (ETH Zürich, Switzerland) (27). The open area 24 h after scratch was compared to the initial scratch size.

Colony formation assay

In total, 500 PC3 cells were seeded and transfected per well in 6-well plates. Medium was changed after 24 h. After 14 days of cultivation, the supernatant was removed and the cells were stained using 0.05% crystal violet for 15 min. Wells were photographed at ×20 and the number of colonies was counted.

Nanoparticle tracking analysis of supernatant of transfected and inhibitor-treated cells

PC3 cells were seeded and transfected in 24-well plates as described above. Forty-eight hours after transfection the supernatant was removed and the cells were washed with sterile PBS two times. Afterwards cells were again incubated for 48 h in serum-free RPMI medium. Subsequently the supernatant was recovered and sequentially centrifuged at 300, 2,000 and 12,000 g to remove cellular debris, larger particles and microvesicles. Each sample of cleared supernatant was measured five times for 45 sec using nanoparticle tracking analysis (NTA) on an LM10 (Malvern Instruments, Malvern, UK) at dilutions between 1:20 and 1:50. Particle concentration and size were determined. The same was done for the supernatant of PC3 cells treated with the sphingomyelinase inhibitor GW4869, known to block exosome release in vitro. For GW4869 a concentration of 5 ng/μl in 1.7% DMSO was chosen. The same concentration of DMSO without inhibitor served as control condition.

Statistics

Statistical calculations and graph design were performed using Prism 6 (GraphPad Software, Inc., La Jolla, CA, USA) or JMP 11 (SAS Institute GmbH, Heidelberg, Germany). Testing for normality showed non-normal data distribution in all tested datasets thus non-parametric tests were used. Calculation of inter-group gene expression changes, for both in silico cohorts and patient samples analyzed with qRT-PCR, were performed using Mann-Whitney test. Correlation of candidate gene expression was done using Spearman correlation. Differences in BCR were calculated using log-rank test. For in vitro assays parametric t-test was used. P-values ≤0.05 were considered statistically significant.

Results

Several candidate genes are associated with BCR in localized PCa in silico

Thirty-seven genes either involved in vesicular trafficking and secretion or typically present on extracellular vesicles were identified from recent literature. These and five genes associated with PCa were tested in the microarray dataset of localized PCa by Taylor et al (19) for their prediction of BCR. Gene expression cut-off levels, defined by partition test, divided the dataset into two groups of patients for each candidate, one group with candidate gene high expression and the other with candidate gene low expression. Table IV summarizes cut-off values, group sizes, number of BCR-events and p-values of log-rank test for BCR. For 22 genes a significant correlation between gene expression and BCR was seen. Yet, for many genes partition test had resulted in very imbalanced group sizes. With a minimum group size requirement set to 20 patients, six genes remained. Kaplan-Meier graphs with BCR-free survival as outcome parameter for these genes are displayed in Fig. 1. Beside genes known to be associated with PCa outcome, such as EZH2 and PTEN, there were also the two RABs RAB27A and RAB27B and the ESCRT-II component VPS36 associated with BCR. In case of the latter three candidates a lower expression was predictive for a worse outcome.

Table IV

Cut-off values, group sizes, number of BCR-events and p-values of the log-rank test for time to BCR for 37 genes tested in the Taylor et al dataset (19) are displayed.a

Table IV

Cut-off values, group sizes, number of BCR-events and p-values of the log-rank test for time to BCR for 37 genes tested in the Taylor et al dataset (19) are displayed.a

GeneCut-off
(z-score)
High expression
(patients/events)
Low expression
(patients/events)
p-value
(BCR)
AMACR3.39072 (18)59 (9)0.213
AR1.11332 (8)99 (19)0.424
EZH23.88330 (10)101 (17)0.027
FOLH11.95711 (9)120 (18) <0.001
KLK30.68331 (4)99 (23)0.255
PCA32.45285 (21)46 (6)0.147
PTEN−1.300108 (17)23 (10) <0.001
PDCD6IP−1,909126 (24)5 (3)0.001
TSG101−1.672124 (23)7 (4)0.002
UBAP1−1.771115 (19)16 (8) <0.001
VPS280.84227 (3)104 (24)0.912
SNF81.39710 (0)121 (27)0.186
VPS25−1.651114 (21)17 (6)0.206
VPS36−0.60479 (10)52 (17)0.005
ST1−2.274121 (20)10 (7) <0.001
VPS4A−2.468126 (23)5 (4) <0.001
VTA1−1.277120 (22)11 (5) <0.001
RAB5A−1.506121 (22)10 (5) <0.001
RAB27A0.60752 (6)79 (21)0.033
RAB27B−0.30196 (15)35 (12)0.025
RAB35−0.80279 (12)52 (15)0.077
TBC1D10A−0.78761 (9)70 (18)0.064
CD9−0.9227102 (16)29 (11)0.007
CD631.8145 (3)126 (24)0.014
CD810.5318 (4)123 (23)0.014
CHMP4C3.2667 (4)124 (23)0.004
DCT−1.809124 (24)7 (3)0.083
EPCAM1.926568 (18)63 (9)0.111
HSPA4−1.489120 (22)11 (5) <0.001
HSP90AA1−1.352124 (22)7 (5) <0.001
MMP21.5837 (3)124 (24)0.043
MMP9−0.669125 (27)6 (0)0.233
MMP141.7405 (3)126 (24) <0.001
STEAP3−2.119120 (22)11 (5)0.007

a Log-rank test revealed candidate genes to be associated with patient outcome after determination of gene expression cut-off values.

RAB27A, RAB27B and VPS36 are overexpressed in metastatic PCa in silico

The dataset by Taylor et al (19) was used to stratify gene expression of RAB27A, RAB27B and VPS36 for tumor stage. Both for RAB27A and RAB27B a significantly decreased expression in distant metastatic lesions (RAB27A: p<0.001; RAB27B: p<0.001) but not in primary tumors and lymph node metastases (Fig. 2A and B) was seen.

The expression of VPS36 was significantly reduced in locally advanced PCa (p=0.041), as well as in lymph node (p=0.003) and distant metastases (p<0.001) (Fig. 2C). These findings were re-assessed in two other publicly available microarray datasets. In the Tomlins et al dataset (21) the three candidate genes were significantly reduced in tumors with Gleason score ≥7b (RAB27A, p=0.007; RAB27B, p=0.003; VPS36, p=0.001) and metastatic lesions (RAB27A, p<0.001; RAB27B, p<0.001; VPS36, p<0.001) (Fig. 2D–F). In the Yu et al dataset (22) all three candidates were underexpressed in metastases (RAB27A, p=0.0389; RAB27B, p<0.001; VPS36, p=0.001). RAB27B was also higher expressed in stage 2 tumors (p=0.014) (Fig. 2G–H).

Expression of candidate genes correlates among each other in silico

Analysis of microarray datasets furthermore revealed the expression of RAB27A, RAB27B and VPS36 to be positively correlated with each other. A low expression of RAB27A was associated with a low expression of RAB27B in all three datasets. In the Taylor et al dataset (19) expression of both genes was also positively correlated with the expression of VPS36. This could partly be confirmed in the other two datasets. Bivariate correlation plots are shown in Fig. 3A–C for the Taylor et al dataset (19) exemplarily. For the two other analyzed datasets correlation coefficients and p-values are shown as interaction graphs in Fig. 3D and E.

VPS36 expression is decreased in advanced primary PCa

To validate these findings in independent patient samples with a different technique, two cohorts of PCa patients and benign prostatic tissue controls were analyzed using qRT-PCR. In the cDNA array consisting of 40 patients with localized PCa and 8 patients with BPH no altered expression of all three candidates, compared to benign controls, was seen when stratifying for tumor stage and Gleason score (data not shown). Consistent with the in silico analyses, RAB27A and RAB27 were positively correlated in this dataset and also a positive correlation between RAB27B and VPS36 was found (Fig. 4).

Since in silico a downregulation of RAB27A, RAB27B and VPS36 was seen in metastatic PCa, but not in organ confined PCa, a second dataset consisting of patients undergoing transurethral resection of PCa for palliation, with many of them having metastatic and/or castration-resistant disease and therefore reflecting an advanced disease state, was analyzed.

In this cohort a significant reduction of VPS36 expression in tumor samples (p=0.022) was found, whilst RAB27A and RAB27B were not reduced in tumor samples, instead showing a slight but not significant increase in expression (Fig. 5A–C). When stratifying the tumor samples into androgen-dependent (AD) and castration-resistant (CR) tumors according to the clinical data, VPS36, but not RAB27A and RAB27B, was lower expressed in CR tumors, compared to both controls (p=0.002) and AD tumors (p=0.026) (Fig. 5E–G).

Furthermore, the expression of AR, typically overexpressed in CR tumors as a sign of aberrant androgen signaling, was analyzed. As expected a strong overexpression of AR in CR tumors compared to controls (p<0.001) and to AD tumors (p<0.001) was seen (Fig. 5D and H).

The expression of the three candidate genes positively correlated with each other (Fig. 5I). The expression of VPS36 and RAB27A did not show a significant correlation with AR (Fig. 5J), whilst for RAB27B a positive correlation with AR was seen (Fig. 5K).

Knockdown experiments confirm interdependency of candidate genes

To further elucidate the unknown role of RAB27A, RAB27B and VPS36 in PCa, siRNA-mediated knockdown experiments were performed in the castration-resistant PC3 cell line. Whilst knockdown of RAB27A and RAB27B resulted in no significant changes in the expression of the other genes in PC3 cells, knockdown of VPS36 resulted in a significant increase in the expression of RAB27A (p=0.001) and RAB27B (p=0.046), indicating potential crosstalk (Fig. 6A–C). Interestingly, in MDA-MB-231 breast cancer cells the same dependency with an elevated expression of RAB27A (p=0.003) and RAB27B (p=0.019) upon knockdown of VPS36 was seen (Fig. 6D–F).

Knockdown of RAB27B and VPS36 reduces colony formation and increases extracellular particle release

Knockdown of none of the candidates significantly reduced the growth of PC3 cells, measured with MTT assay under standard dense 2D cell culture conditions (Fig. 7A). Also under standard conditions, the migration of PC3 cells, determined by the scratch assay, was not significantly altered upon knockdown of the candidate genes (Fig. 7C).

Unlike in the MTT assay, when transfected PC3 cells were seeded at low density for long-term cultivation, a strong reduction of colony formation upon knockdown of RAB27B (p<0.001) and VPS36 (p<0.001), potentially indicating an impairment of paracrine cell-to-cell signaling, was seen (Fig. 7B).

Following this approach, the influence of the candidate genes on the production of extracellular particles not pelleting at 12,000 g centrifugation, being a surrogate for small extracellular vesicles, was investigated. As control PC3 cells were treated with GW4869, which is a known blocker of the release of exosomes. GW4869 treatment did not significant increase cellular growth, but reduced the number of released particles by 37.1% (p<0.001), indicating that the experimental setup could monitor modulations in extracellular vesicle release (Fig. 8A and B). Interestingly there was a significant increase in particle release in PC3 cells upon knockdown of RAB27B (p=0.014) and VPS36 (p<0.001), again pointing to an alteration in cell-to-cell signaling dependent on the candidate genes (Fig. 8D). Both for inhibitor treatment and RNAi no changes in the size of the released particles were seen (Fig. 8C and F). Fig. 8E shows the typical size distribution of the particles released from PC3 cells treated with #CTRL and #VPS36.

Discussion

Genes with known function in vesicular trafficking and secretion and genes coding for typical extracellular vesicle marker proteins were analyzed in existing microarray datasets, to investigate their role in PCa, Among these, the expression of VPS36, RAB27A and RAB27B was inversely correlated with BCR-free survival. These three genes were underexpressed in metastatic PCa tissue samples.

RAB27A and RAB27B are deemed to be key regulators of exosome secretion, since they mediate the transport of MVBs to the plasma membrane and their fusion with the plasma membrane (13). Several studies have shown their expression to be correlated with exosome release in vitro and in vivo in different cancer entities (4,28). Typically blockage of their expression reduces the number of exosomes secreted (29). Two studies suggest RAB27A to be involved in the secretion of typical PCa markers the PSA and PSAP by modulating PI3K-signaling (30,31). Furthermore, reduction of RAB27A expression results in a reduction of stroma-assisted tumor growth by reducing the number of secreted exosomes from PCa cells (32). It is known that the regulatory effects of RAB27 GTPases are cell-specific, depending on expression and function of potential effector molecules (33). In mast cells they were shown to regulate granule exocytosis, with RAB27B being more of a positive regulator and RAB27A more of a negative regulator of degranulation (34). This underlines that they can both have concordant and discordant function and are also involved in immune-modulatory processes (35).

RAB27B is known to be expressed in various epithelial tissues in adults, while RAB27B expression is low in fetal tissue (36). Yet, the expression of RAB27A and RAB27B has only been studied in some cancer entities. In breast cancer an elevated expression of RAB27B is associated with lymph node metastasis and predicts poor prognosis (37). Also in colorectal cancer RAB27B has been identified as potential predictor of metastasis and outcome (38). A higher expression of RAB27B mRNA was observed in cancer tissue and was associated with decreased overall survival. Conversely another study found a lower expression of RAB27A and RAB27B to be correlated with colorectal cancer progression (39). To date nothing is known about the expression of RAB27A and RAB27B in PCa. In this study data from in silico analysis consistently showed the expression of both genes to be lower in PCa metastases. In the Tomlins et al dataset (21) both genes were also under-expressed in localized PCa with a Gleason score of 7b and higher. Also a positive correlation between RAB27B and AR was seen. A recent study found the expression of RAB27A to be regulated by AR upon androgen deprivation therapy in PCa (40), but the underlying mechanisms are yet unknown.

Unlike for RAB27A and RAB27B, the role of VPS36 in cancer has not been investigated at all. As a member of the ESCRT-II complex it is mainly associated with MVB biogenesis, cellular abscission and viral budding (41). Deregulation of JAK/STAT signaling in tissues with modifications in the ESCRT-II complex resulted in facilitated tumorigenesis in Drosophila, indicating crosstalk with known cancer-pathways (42). VPS36, also known as EAP45, is located on chromosome 13 and transcribes for three different transcript variants. Crystal structure analyses propose it to directly interact with ubiquitinated proteins (43). It showed a negative correlation with advanced tumor stage and worse outcome in silico. This was confirmed by qRT-PCR analysis. Especially in CR PCa VPS36 expression was significantly reduced, whilst AR expression was strongly increased. Yet, no significant correlation between these two genes was seen.

Both in existing microarray datasets and in tumor samples analyzed with qRT-PCR, the expression of the candidate genes was consistently correlated among each other. This might be due to a mechanism simultaneously regulating the candidate gene expression. In vitro a significant increase of RAB27A and RAB27B expression was seen upon knockdown of VPS36 in PC3 cells, but not vice versa. This suggests an additional directional interdependency of these genes. At least with regard to exosome release VPS36 is located prior in the signaling cascade, making this relationship plausible. Re-testing in the MDA MB-231 breast cancer cell line confirmed these results and suggests this interdependency not to be PCa-specific.

Knockdown of the candidate genes did not result in a significant alteration of growth and migration of PC3 cells under standard culture conditions. The significant reduction of colony formation upon knockdown of RAB27B and VPS36 seems to be controversial to the findings in patient samples. Yet, cellular behavior in this long-term setup is less dependent on the induced changes in the individual cells and direct cell-to-cell interaction. Instead it also reflects aspects of paracrine stimulation via secreted molecules and particles like exosomes. Again comparable data from other studies in PCa are lacking. In vivo data from lung cancer mouse models show a reduction of tumor growth only for metastatic, but not for non-metastatic cell lines treated with shRNAs targeting RAB27A (4).

To investigate the impact of candidate gene expression on extracellular vesicle secretion, NTA analyses of the supernatant of treated cells were conducted. In order to facilitate measurement of multiple samples under different treatment conditions, particles in the supernatant of cancer cells not pelleting at 12,000 g were used as a surrogate for small extracellular vesicles. Using this system, treatment of PC3 cells with the sphingomyelinase inhibitor GW4869, a known blocker of exosome release (44), resulted in a strong reduction of extracellular particles, indicating the applicability of this setup. Unexpectedly, knockdown of RAB27B and VPS36 resulted in a significant increase of extracellular particles remaining in the PC3 supernatant after centrifugation at 12,000 g. Again, there is not much comparable data in the literature. One study has used a similar experimental setup to monitor vesicle release from MDA MB-231 cells. shRNA-mediated knockout of RAB27A resulted in a significant decrease of secreted vesicles (29). As in this study, no changes in the size of secreted particles were observed. RAB27B and VPS36 were not tested in this study. Ostrowski et al (13) also found no changes in size and morphology of HeLa exosomes upon shRNA mediated knockdown of RAB27A and RAB27B. By analyzing ultracentrifugation pellets they saw a reduction of exosomes upon knockdown, which is controversial to our findings. These different observations might be attributed to the different assay and readout conditions. Also a time-dependent regulation of extracellular particle release upon modulation of RAB27A and RAB27B expression can be debated. Furthermore, the used cell lines represent different cancer entities with a potentially different regulation of RAB27A and RAB27B expression and function. Also the different cell lines originate from different microenvironmental conditions. In this context it has been shown that extracellular vesicles of malignant cells have an ambivalent role both in immune-evasion (45,46) and the activation of the immune system against cancer cells (47), adding a further potential regulatory element on extracellular vesicle biogenesis and secretion.

Our study is limited by the heterogeneous assay platforms used for microarray experiments. Furthermore, sample handling and data post-processing offer potential source of bias. The cohorts analyzed by qRT-PCR were heterogeneous with regard to tumor stage and, mainly for the patients treated with transurethral resection, also for iatrogenous factors as pretreatment and castration-resistance. Also the sample material itself, being FFPE tissue, is not optimal for downstream analyses due to a strong degradation of nucleic acids. Yet studies relying on antibody-based protein analyses, being an alternative strategy to analyze candidate expression, also have specific limitations such as degradation of epitopes, unspecific antibody binding and a more difficult quantification.

In conclusion, several genes with known function in vesicular trafficking and secretion are deregulated in PCa. Of these, RAB27A, RAB27B and VPS36 are consistently underexpressed in advanced PCa and predict worse outcome. For VPS36 underexpression could also be shown in primary tumors undergoing local palliation. Subgroup analysis indicates a stronger decrease in CR tumors, whilst no correlation was found with the expression the AR.

In the analyzed datasets the expression of RAB27A, RAB27B and VPS36 showed a strong positive correlation among each other, pointing to a common regulatory mechanism. Additionally in vitro manipulation of the gene expression indicates RAB27A and RAB27B to be dependent of VPS36. An increase of extracellular particles and a reduction of colony formation upon knockdown of RAB27B and VPS36 indicate an involvement in paracrine cell-cell communication.

Further analyses in larger cohorts are needed to validate the repressed expression of VPS36, RAB27A and RAB27B in advanced and metastatic PCa. Additionally, studies dissecting their specific regulation and function should help to elucidate their role in vesicular trafficking and secretion and paracrine signaling in PCa.

Abbreviations:

AD

androgen dependent

AR

androgen receptor

BCR

biochemical recurrence

CR

castration-resistant

FFPE

formalin-fixed paraffin-embedded tissue

PCa

prostate cancer

PSA

prostate-specific antigen

PSMA

prostate-specific membrane antigen

TUR-P

transurethral resection of the prostate

Acknowledgments

This study was supported by the Foundation on Cancer and Scarlet Research of the University of Heidelberg. T.S.W. was supported by a Ferdinand Eisenberger scholarship of the German Society of Urology.

References

1 

Jemal A, Bray F, Center MM, Ferlay J, Ward E and Forman D: Global cancer statistics. CA Cancer J Clin. 61:69–90. 2011. View Article : Google Scholar : PubMed/NCBI

2 

Vlassov AV, Magdaleno S, Setterquist R and Conrad R: Exosomes: Current knowledge of their composition, biological functions, and diagnostic and therapeutic potentials. Biochim Biophys Acta. 1820:940–948. 2012. View Article : Google Scholar : PubMed/NCBI

3 

Junker K, Heinzelmann J, Beckham C, Ochiya T and Jenster G: Extracellular vesicles and their role in urologic malignancies. Eur Urol. 70:323–331. 2016. View Article : Google Scholar : PubMed/NCBI

4 

Bobrie A, Krumeich S, Reyal F, Recchi C, Moita LF, Seabra MC, Ostrowski M and Théry C: Rab27a supports exosome-dependent and -independent mechanisms that modify the tumor microenvironment and can promote tumor progression. Cancer Res. 72:4920–4930. 2012. View Article : Google Scholar : PubMed/NCBI

5 

Beckham CJ, Olsen J, Yin P-N, Wu CH, Ting HJ, Hagen FK, Scosyrev E, Messing EM and Lee YF: Bladder cancer exosomes contain EDIL-3/Del1 and facilitate cancer progression. J Urol. 192:583–592. 2014. View Article : Google Scholar : PubMed/NCBI

6 

Alderton GK: Metastasis. Exosomes drive premetastatic niche formation. Nat Rev Cancer. 12:4472012. View Article : Google Scholar : PubMed/NCBI

7 

Peinado H, Alečković M, Lavotshkin S, Matei I, Costa-Silva B, Moreno-Bueno G, Hergueta-Redondo M, Williams C, García-Santos G, Ghajar C, et al: Melanoma exosomes educate bone marrow progenitor cells toward a pro-metastatic phenotype through MET. Nat Med. 18:883–891. 2012. View Article : Google Scholar : PubMed/NCBI

8 

Webber MM, Trakul N, Thraves PS, Bello-DeOcampo D, Chu WW, Storto PD, Huard TK, Rhim JS and Williams DE: A human prostatic stromal myofibroblast cell line WPMY-1: A model for stromal-epithelial interactions in prostatic neoplasia. Carcinogenesis. 20:1185–1192. 1999. View Article : Google Scholar : PubMed/NCBI

9 

Webber J, Steadman R, Mason MD, Tabi Z and Clayton A: Cancer exosomes trigger fibroblast to myofibroblast differentiation. Cancer Res. 70:9621–9630. 2010. View Article : Google Scholar : PubMed/NCBI

10 

Colombo M, Moita C, van Niel G, Kowal J, Vigneron J, Benaroch P, Manel N, Moita LF, Théry C and Raposo G: Analysis of ESCRT functions in exosome biogenesis, composition and secretion highlights the heterogeneity of extracellular vesicles. J Cell Sci. 126:5553–5565. 2013. View Article : Google Scholar : PubMed/NCBI

11 

Adell MA and Teis D: Assembly and disassembly of the ESCRT-III membrane scission complex. FEBS Lett. 585:3191–3196. 2011. View Article : Google Scholar : PubMed/NCBI

12 

Brinton LT, Sloane HS, Kester M and Kelly KA: Formation and role of exosomes in cancer. Cell Mol Life Sci. 72:659–671. 2015. View Article : Google Scholar

13 

Ostrowski M, Carmo NB, Krumeich S, Fanget I, Raposo G, Savina A, Moita CF, Schauer K, Hume AN, Freitas RP, et al: Rab27a and Rab27b control different steps of the exosome secretion pathway. Nat Cell Biol. 12:19–30. 2010. View Article : Google Scholar

14 

Théry C, Amigorena S, Raposo G and Clayton A: Isolation and characterization of exosomes from cell culture supernatants and biological fluids. Curr Protoc Cell Biol. Chapter 3: Unit 3.22. 2006. View Article : Google Scholar

15 

Théry C, Zitvogel L and Amigorena S: Exosomes: Composition, biogenesis and function. Nat Rev Immunol. 2:569–579. 2002.PubMed/NCBI

16 

Villarroya-Beltri C, Gutiérrez-Vázquez C, Sánchez-Cabo F, Pérez-Hernández D, Vázquez J, Martin-Cofreces N, Martinez-Herrera DJ, Pascual-Montano A, Mittelbrunn M and Sánchez-Madrid F: Sumoylated hnRNPA2B1 controls the sorting of miRNAs into exosomes through binding to specific motifs. Nat Commun. 4:29802013. View Article : Google Scholar : PubMed/NCBI

17 

Koppers-Lalic D, Hackenberg M, Bijnsdorp IV, van Eijndhoven MA, Sadek P, Sie D, Zini N, Middeldorp JM, Ylstra B, de Menezes RX, et al: Nontemplated nucleotide additions distinguish the small RNA composition in cells from exosomes. Cell Rep. 8:1649–1658. 2014. View Article : Google Scholar : PubMed/NCBI

18 

van der Pol E, Böing AN, Harrison P, Sturk A and Nieuwland R: Classification, functions, and clinical relevance of extracellular vesicles. Pharmacol Rev. 64:676–705. 2012. View Article : Google Scholar : PubMed/NCBI

19 

Taylor BS, Schultz N, Hieronymus H, Gopalan A, Xiao Y, Carver BS, Arora VK, Kaushik P, Cerami E, Reva B, et al: Integrative genomic profiling of human prostate cancer. Cancer Cell. 18:11–22. 2010. View Article : Google Scholar : PubMed/NCBI

20 

Cerami E, Gao J, Dogrusoz U, Gross BE, Sumer SO, Aksoy BA, Jacobsen A, Byrne CJ, Heuer ML, Larsson E, et al: The cBio cancer genomics portal: An open platform for exploring multidimensional cancer genomics data. Cancer Discov. 2:401–404. 2012. View Article : Google Scholar : PubMed/NCBI

21 

Tomlins SA, Mehra R, Rhodes DR, Cao X, Wang L, Dhanasekaran SM, Kalyana-Sundaram S, Wei JT, Rubin MA, Pienta KJ, et al: Integrative molecular concept modeling of prostate cancer progression. Nat Genet. 39:41–51. 2007. View Article : Google Scholar

22 

Yu YP, Landsittel D, Jing L, Nelson J, Ren B, Liu L, McDonald C, Thomas R, Dhir R, Finkelstein S, et al: Gene expression alterations in prostate cancer predicting tumor aggression and preceding development of malignancy. J Clin Oncol. 22:2790–2799. 2004. View Article : Google Scholar : PubMed/NCBI

23 

Specht E, Kaemmerer D, Sänger J, Wirtz RM, Schulz S and Lupp A: Comparison of immunoreactive score, HER2/neu score and H score for the immunohistochemical evaluation of somatostatin receptors in bronchopulmonary neuroendocrine neoplasms. Histopathology. 67:368–377. 2015. View Article : Google Scholar : PubMed/NCBI

24 

Kaemmerer D, Reimann C, Specht E, Wirtz RM, Sayeg M, Baum RP, Schulz S and Lupp A: Differential expression and prognostic value of the chemokine receptor CXCR4 in bronchopulmonary neuroendocrine neoplasms. Oncotarget. 6:3346–3358. 2015. View Article : Google Scholar : PubMed/NCBI

25 

Untergasser A, Cutcutache I, Koressaar T, Ye J, Faircloth BC, Remm M and Rozen SG: Primer3 - new capabilities and interfaces. Nucleic Acids Res. 40:e1152012. View Article : Google Scholar

26 

Livak KJ and Schmittgen TD: Analysis of relative gene expression data using real-time quantitative PCR and the 2(−∆∆C(T)) method. Methods. 25:402–408. 2001. View Article : Google Scholar

27 

Ashby WJ and Zijlstra A: Established and novel methods of interrogating two-dimensional cell migration. Integr Biol. 4:1338–1350. 2012. View Article : Google Scholar

28 

Li W, Hu Y, Jiang T, Han Y, Han G, Chen J and Li X: Rab27A regulates exosome secretion from lung adenocarcinoma cells A549: Involvement of EPI64. APMIS. 122:1080–1087. 2014.PubMed/NCBI

29 

Zheng Y, Campbell EC, Lucocq J, Riches A and Powis SJ: Monitoring the Rab27 associated exosome pathway using nanoparticle tracking analysis. Exp Cell Res. 319:1706–1713. 2013. View Article : Google Scholar

30 

Johnson JL, Ellis BA, Noack D, Seabra MC and Catz SD: The Rab27a-binding protein, JFC1, regulates androgen-dependent secretion of prostate-specific antigen and prostatic-specific acid phosphatase. Biochem J. 391:699–710. 2005. View Article : Google Scholar : PubMed/NCBI

31 

Catz SD: Characterization of Rab27a and JFC1 as constituents of the secretory machinery of prostate-specific antigen in prostate carcinoma cells. Methods Enzymol. 438:25–40. 2008. View Article : Google Scholar : PubMed/NCBI

32 

Webber JP, Spary LK, Sanders AJ, Chowdhury R, Jiang WG, Steadman R, Wymant J, Jones AT, Kynaston H, Mason MD, et al: Differentiation of tumour-promoting stromal myofibroblasts by cancer exosomes. Oncogene. 34:290–302. 2015. View Article : Google Scholar

33 

Catz SD: Regulation of vesicular trafficking and leukocyte function by Rab27 GTPases and their effectors. J Leukoc Biol. 94:613–622. 2013. View Article : Google Scholar : PubMed/NCBI

34 

Singh RK, Mizuno K, Wasmeier C, Wavre-Shapton ST, Recchi C, Catz SD, Futter C, Tolmachova T, Hume AN and Seabra MC: Distinct and opposing roles for Rab27a/Mlph/MyoVa and Rab27b/Munc13-4 in mast cell secretion. FEBS J. 280:892–903. 2013.PubMed/NCBI

35 

Elstak ED, Neeft M, Nehme NT, Voortman J, Cheung M, Goodarzifard M, Gerritsen HC, van Bergen En Henegouwen PM, Callebaut I, de Saint Basile G, et al: The munc13-4-rab27 complex is specifically required for tethering secretory lysosomes at the plasma membrane. Blood. 118:1570–1578. 2011. View Article : Google Scholar : PubMed/NCBI

36 

Hendrix A, Lambein K, Westbroek W, Seabra MC, Cocquyt V, Pauwels P, Bracke M, Gespach C and De Wever O: An immunohistochemical analysis of Rab27B distribution in fetal and adult tissue. Int J Dev Biol. 56:363–368. 2012. View Article : Google Scholar : PubMed/NCBI

37 

Zhang J-X, Huang X-X, Cai M-B, Tong ZT, Chen JW, Qian D, Liao YJ, Deng HX, Liao DZ, Huang MY, et al: Overexpression of the secretory small GTPase Rab27B in human breast cancer correlates closely with lymph node metastasis and predicts poor prognosis. J Transl Med. 10:2422012. View Article : Google Scholar : PubMed/NCBI

38 

Bao J, Ni Y, Qin H, Xu L, Ge Z, Zhan F, Zhu H, Zhao J, Zhou X, Tang X, et al: Rab27b is a potential predictor for metastasis and prognosis in colorectal cancer. Gastroenterol Res Pract. 2014:9131062014. View Article : Google Scholar

39 

Dong W, Cui J, Yang J, Li W, Wang S, Wang X, Li X, Lu Y and Xiao W: Decreased expression of Rab27A and Rab27B correlates with metastasis and poor prognosis in colorectal cancer. Discov Med. 20:357–367. 2015.

40 

Shaw GL, Whitaker H, Corcoran M, Dunning MJ, Luxton H, Kay J, Massie CE, Miller JL, Lamb AD, Ross-Adams H, et al: The early effects of rapid androgen deprivation on human prostate cancer. Eur Urol. 70:214–218. 2016. View Article : Google Scholar :

41 

Schuh AL and Audhya A: The ESCRT machinery: From the plasma membrane to endosomes and back again. Crit Rev Biochem Mol Biol. 49:242–261. 2014. View Article : Google Scholar : PubMed/NCBI

42 

Woodfield SE, Graves HK, Hernandez JA and Bergmann A: De-regulation of JNK and JAK/STAT signaling in ESCRT-II mutant tissues cooperatively contributes to neoplastic tumorigenesis. PLoS One. 8:e560212013. View Article : Google Scholar : PubMed/NCBI

43 

Alam SL, Langelier C, Whitby FG, Koirala S, Robinson H, Hill CP and Sundquist WI: Structural basis for ubiquitin recognition by the human ESCRT-II EAP45 GLUE domain. Nat Struct Mol Biol. 13:1029–1030. 2006. View Article : Google Scholar : PubMed/NCBI

44 

Chairoungdua A, Smith DL, Pochard P, Hull M and Caplan MJ: Exosome release of β-catenin: A novel mechanism that antagonizes Wnt signaling. J Cell Biol. 190:1079–1091. 2010. View Article : Google Scholar : PubMed/NCBI

45 

Liu Y, Gu Y and Cao X: The exosomes in tumor immunity. OncoImmunology. 4:e10274722015. View Article : Google Scholar : PubMed/NCBI

46 

Lundholm M, Schröder M, Nagaeva O, Baranov V, Widmark A, Mincheva-Nilsson L and Wikström P: Prostate tumor-derived exosomes downregulate NKG2D expression on natural killer cells and CD8+ T cells: Mechanism of immune evasion. PLoS One. 9:e1089252014. View Article : Google Scholar

47 

Li W, Mu D, Tian F, Hu Y, Jiang T, Han Y, Chen J, Han G and Li X: Exosomes derived from Rab27a-overexpressing tumor cells elicit efficient induction of antitumor immunity. Mol Med Rep. 8:1876–1882. 2013.PubMed/NCBI

Related Articles

Journal Cover

March 2017
Volume 50 Issue 3

Print ISSN: 1019-6439
Online ISSN:1791-2423

Sign up for eToc alerts

Recommend to Library

Copy and paste a formatted citation
APA
Worst, T.S., Meyer, Y., Gottschalt, M., Weis, C., von Hardenberg, J., Frank, C. ... Erben, P. (2017). RAB27A, RAB27B and VPS36 are downregulated in advanced prostate cancer and show functional relevance in prostate cancer cells. International Journal of Oncology, 50, 920-932. https://doi.org/10.3892/ijo.2017.3872
MLA
Worst, T. S., Meyer, Y., Gottschalt, M., Weis, C., von Hardenberg, J., Frank, C., Steidler, A., Michel, M. S., Erben, P."RAB27A, RAB27B and VPS36 are downregulated in advanced prostate cancer and show functional relevance in prostate cancer cells". International Journal of Oncology 50.3 (2017): 920-932.
Chicago
Worst, T. S., Meyer, Y., Gottschalt, M., Weis, C., von Hardenberg, J., Frank, C., Steidler, A., Michel, M. S., Erben, P."RAB27A, RAB27B and VPS36 are downregulated in advanced prostate cancer and show functional relevance in prostate cancer cells". International Journal of Oncology 50, no. 3 (2017): 920-932. https://doi.org/10.3892/ijo.2017.3872