Background
Exosomes or extracellular vesicles (EV) are microvesicles with a diameter of 40–150 nm that are secreted from various cells [
1]. Numerous proteins, miRNAs, RNAs and DNAs are contained in exosomes and their molecular signature largely reflects that of the cells from which they are originated. Exosomes exist in the body fluids such as blood and urine and thus are expected to be a new marker for various diseases including cancer. Yoshioka et al. demonstrated that CD147 embedded in cancer-linked EV in blood can be used for detection of colorectal cancer [
2]. Melo et al. recently reported that exosomes expressing glypican-1 in blood can differentiate patients with pancreatic cancer from healthy subjects and those with benign pancreatic disease [
3].
Prostate cancer (PC), one of the most common male cancer, is the second-leading cause of cancer death among men in the United States [
4]. PC well responds to androgen deprivation therapy, but 10 to 20% of patients develop castration-resistant prostate cancer (CRPC) [
5]. In patients with advanced CRPC, bone metastasis is commonly found. Docetaxel, a microtubule-stabilizing taxane, has been used as the first-line chemotherapy for CRPC, but there is a finite amount of time before acquiring resistance [
6,
7]. The recent introduction of cabazitaxel, enzalutamide and abiraterone has expanded treatment options for metastatic CRPC patients [
8].
Prostate-specific antigen (PSA) has been commonly used as a marker for PC, but it cannot differentiate PC from benign prostatic hyperplasia (BPH) unless PC is advanced and shows much higher serum PSA levels than BPH [
9,
10]. In conjunction with measuring PSA levels, imaging modalities such as CT, MRI and bone scan are recommended to monitor the status of patients. There are numerous reports that identified potential markers to diagnose PC, to diagnose progression or aggressiveness of CRPC and to predict prognosis of PC [
11]. Since serial prostate biopsy is not usually performed due to its invasiveness and inaccuracy, it would be of great benefit if PC could be diagnosed and monitored by exosomes in the body fluids. We and others have demonstrated that prostate-specific membrane antigen (PSMA) and P-glycoprotein (P-gp) encoded by multi-drug resistance protein 1 (MDR1) expressed on the surface of blood exosomes could be a marker for PC and taxane-resistant CRPC, respectively [
12‐
15]. We have also recently reported the potential of integrin β4 and vinculin in exosomes as markers for progression and aggressiveness of CRPC [
16].
In the present study, we aimed to identify novel exosomal markers for PC especially those for castration-resistance and bone metastasis by analyzing exosomes secreted from PC cell lines including androgen-dependent LNCaP cell line and its sublines of partially androgen-independent C4, androgen-independent C4–2 and bone metastatic C4–2B [
17,
18]. Among proteins identified by proteomic analysis, we focused on gamma-glutamyltransferase 1 (GGT1), a cell-surface enzyme that regulates the catabolism of extracellular L-gamma-glutamyl-L-cysteinylglycine (glutathione; GSH). Since GGT activity correlated with GGT1 expression in serum exosomes isolated by differential centrifugation, we measured GGT activity in patients. Contrary to our expectation, we found that serum exosomal GGT activity was significantly higher in PC patients than in BPH patients, which was supported by the finding that GGT1 expression was increased in PC tissues compared with BPH tissues. Altogether, we have identified serum exosomal GGT activity as a novel marker to diagnose PC or to distinguish PC from BPH.
Methods
Cell culture
Human prostate cancer LNCaP cell line and its sublines of C4, C4–2 and C4–2B cell lines were obtained from the MD Anderson Cancer Center (Houston, TX, USA) and cultured in DMEM/Ham’s F12 (4:1) medium supplemented with 10% fetal bovine serum, 5 μg/mL insulin, 13.65 pg/mL triiodo-thyronine, 4.4 μg/mL apo-transferrin, 0.244 μg/mL d-biotin and 12.5 μg/mL adenine in a humidified atmosphere containing 5% CO2.
Isolation of exosomes by differential centrifugation
Cells (3.5 × 10
6) seeded on 150-mm dish were cultured for 72 h in DMEM/Ham’s F12 (4:1) medium containing 10% exosome-deprived fetal bovine serum and other supplements described above. Exosomes were isolated from the conditioned medium as previously described [
19]. Briefly, the medium was centrifuged at 2000 xg for 10 min to eliminate cells. Second, the supernatant was centrifuged at 12000 xg for 30 min to remove debris. Third, the supernatant was filtered through 0.22 μm polyvinylidene difluoride (PVDF) filter. Finally, exosomes were pelleted by ultracentrifugation at 110,000 xg for 70 min, resuspended in PBS and stored at −80 °C until use.
Isolation of exosomes by immunocapture
Mouse monoclonal anti-CD9 antibody (BioLegend, San Diego, CA, USA) and anti-PSMA antibody (MBL, Nagoya, Japan) were conjugated with Dynabeads M-270 epoxy magnetic beads (Life Technologies, Eugene, OR, USA) according to the manufacturer’s protocol. The conditioned medium was centrifuged at 2000 xg for 10 min and the supernatant was centrifuged at 12000 xg for 30 min. The supernatant was filtered through 0.22 μm PVDF filter and 30 mL of the filtrate were incubated with 1 mg of the antibody-conjugated beads at 4 °C for 90 min with rotation. The beads were washed 3 times with PBS and resuspended in sample buffer. After separation from magnetic beads, samples were boiled and stored at −20 °C until use.
Isolation of exosomes by size exclusion chromatography
A commercially available size exclusion chromatography column, EVSecond (GL Science, Tokyo, Japan), was used for isolation of exosomes. After washing with PBS, 500 μL serum was loaded onto the column and eluted with PBS. The first 1 mL of eluate was discarded and thereafter the eluate was collected in 24 fractions of 0.1 mL each.
Quantitative proteomic analysis
Proteomic analysis was performed as previously described [
16]. In brief, exosomes were labeled with iTRAQ reagents using the iTRAQ multiplex kit (AB Sciex, Foster City, CA, USA). Labeled samples were separated and automatically spotted onto a MALDI plate using the direct nanoLC and MALDI fraction system DiNa-MaP (KYA Technologies, Tokyo Japan). Mass spectra were acquired using the AB Sciex TOF/TOF 5800 system operated on the TOF/TOF Series Explorer software version 4.1 (AB Sciex). All MS/MS data were submitted to the ProteinPilot software version 4.5 (AB Sciex). Protein identification was considered to be correct based on the following selection criteria: protein having at least 2 peptides with an ion score above 95% confidence; and protein with protein score (ProtScore) > 1.3 (unused,
p < 0.05, 95% confidence).
Western blot analysis
Whole cell lysates were prepared in ice-cold lysis buffer (1% Igepal CA-630, 1% sodium deoxycholate, 0.1% SDS, 150 mM NaCl, 25 mM Tris-HCl [pH 7.6]) containing protease inhibitor cocktail. Cell lysates and exosomes were subjected to electrophoresis on SDS-polyacrylamide gels and transferred to PVDF membranes. After blocking in 5% skim milk, membranes were hybridized with a primary antibody and then with a horseradish peroxidase-linked secondary antibody. After washing, bound proteins were visualized using the ECL Prime Western blotting detection system (GE Healthcare, Little Chalfont, UK) or Immunostar LD (Wako Pure Chemical Industries, Osaka, Japan). Anti-CD9, -PSMA and -β-actin antibodies were obtained from Cell Signaling Technology (Danvers, MA, USA). Antibodies recognizing GGT1 small subunit was purchased from Abnova (Taipei, Taiwan). Anti-GGT1 large subunit and -Alix antibody were from Santa Cruz Biotechnology (Santa Cruz, CA, USA).
Measurement of CD9 level
The CD9 level in exosomes was determined by a sandwich ELISA. A MaxiSorp micro titer plate (Thermo Fisher Scientific, MA, USA) was coated with 5 μg/mL anti-CD9 antibody (Ancell Corporation, Bayport, MN, USA) in carbonate buffer (pH 9.6) at 4 °C overnight. After washing 3 times with PBS, 200 μL of 1% BSA/PBS was added and incubated at room temperature for 1 h with shaking. After washing, sample was added in a final volume of 100 μL and incubated at room temperature for 2 h. After washing, 0.5 μg/mL biotinylated anti-CD9 antibody (Ancell Corporation) in 1% BSA/PBS was added in a final volume of 100 μL and incubated at room temperature for 1 h. After washing, 1:5000 diluted streptavidin-AP (Roche, Basel, Switzerland) in 1% BSA/PBS was added in a final volume of 100 μL and incubated at room temperature for 1 h. After washing 6 times with PBS, CDP-Star substrate with Emerald II Enhancer (Thermo Fisher Scientific) was added and chemiluminescence was recorded by the EnVision Multilabel Reader (PerkinElmer, MA, USA).
Measurement of GGT activity
GGT activity was measured using a fluorescent probe, γ-glutamyl hydroxymethyl rhodamine green (gGlu-HMRG), which is commercially called ProteoGREEN-gGlu (Goryo Chemical, Hokkaido, Japan) [
20]. Twenty microliter of sample was reacted with 180 μL of 1.11 μM ProteoGREEN-gGlu in PBS in each well of 96-well black plates (Corning, NY, USA). The plate was incubated at room temperature for 1 h and fluorescence intensity (Ex/Em 490/520 nm) was measured using the EnVision Multilabel Reader (PerkinElmer).
OptiPrep density gradient centrifugation
Five hundred microliter of serum was centrifuged at 12000 xg for 30 min and the supernatant was filtered through 0.22 μm PVDF filter. The filtered sample was diluted with 11 mL of PBS and centrifuged at 110,000 xg for 70 min. The pellet was resuspended in 500 μL of PBS. A stock solution of OptiPrep (60% w/v iodixanol) (Axis-Shield, Dundee, Scotland) was diluted with 0.25 M sucrose, 10 mM Tris-HCl (pH 7.6) to generate 40%, 20%, 10% and 5% w/v iodixanol solutions. A discontinuous density gradient was generated by sequential layering of 3 mL each of 40, 20 and 10% (w/v) iodixanol solutions, followed by 2.5 mL of 5% iodixanol solution in ultracentrifuge tubes. Sample was overlaid on the discontinuous iodixanol gradient followed by centrifugation at 110,000 xg for 16 h. One milliliter fractions were collected from the top of the gradient. Each sample was diluted with 11 mL of PBS and centrifuged at 110,000 xg for 70 min. The pellet was resuspended in PBS and stored at 4 °C until use.
Collection of blood from patients and isolation of exosomes
This study was approved by the Bioethics Committees of Gifu University and Tokyo Metropolitan Institute of Gerontology and a written informed consent was obtained from all patients. Thirty-nine patients suspicious of PC due to either abnormal MRI findings or elevated PSA levels were recruited. Blood was corrected from patients prior to biopsy. After biopsy, 31 patients and 8 patients were pathologically diagnosed as PC and BPH, respectively. Serum was separated from whole blood by centrifugation at 1800 xg and stored at −80 °C until use. For exosome isolation, 210 μL of serum was centrifuged at 12000 xg for 30 min and the supernatant was filtered through 0.22 μm PVDF filter. The 200 μL of filtered sample diluted with 800 μL of PBS was centrifuged at 100,000 xg for 75 min. The pellet was washed in PBS and centrifuged at 100,000 xg for 75 min. The final pellet was resuspended in PBS and stored at 4 °C until use.
Immunohistochemical analysis of GGT1
This study was approved by the Bioethics Committees of Tokyo Metropolitan Institute of Gerontology. Formalin-fixed paraffin-embedded biopsies and surgically resected tissue specimens from PC (n = 50) and BPH (n = 50) patients were stained for GGT1. The tissue sections (3 μm) were subjected to immunostaining using anti-GGT1 antibody raised against the small subunit (Abnova). After deparaffinization, the sections were preheated in Heat processor solution (pH 6.0, Nichirei, Tokyo, Japan) at 100 °C for 30 min. The sections were then incubated with the anti-GGT1 antibody (1:800 in dilution) at 4 °C overnight. Bound antibodies were detected with the Envision kit (Dako Denmark A/S, Glostrup, Denmark) using diaminobenzidine tetrahydrochloride as a substrate. The sections were then counterstained with Mayer’s hematoxylin. Negative control tissue sections were prepared by omitting the primary antibody. In order to evaluate GGT1 expression, the intensity (1, 0+; 2, 1+; 3, 2+; 4, 3+) and percentage (1, 0–25%; 2, 26–50%; 3, 51–75%; 4, 76–100%) of membranous and cytoplasmic GGT1 staining were scored. GGT1 expression in the prostatic glands and prostatic cancer cells was evaluated under ×200 magnification. The score of intensity multiplied by that of percentage was used as the final score for GGT1 expression. Two independent pathologists blinded to the clinical and pathological information performed scoring.
Statistical analysis
Statistical differences were determined by one-way ANOVA with Tukey’s multiple comparison tests (for comparison among cell lysates and exosomes isolated from cultured cells), Welch’s t-test (for comparison among serum PSA concentration, serum GGT activity and serum exosomal GGT activity), Brunner-Munzel test (for comparison between BPH and PC in immunohistochemical analysis) or paired Student’s t-test (for comparison between cancerous and non-cancerous lesions in immunohistochemical analysis). Spearman’s rank correlation coefficient was used to evaluate the correlation between GGT activity and GGT1 expression. p < 0.05 was considered statistically significant.
Discussion
Based on proteomic analysis of exosomes isolated from PC cell lines by differential centrifugation, we identified GGT1 as a potential exosomal marker for PC. GGT also known as gamma-glutamyl transpeptidase is an enzyme that transfers a gamma-glutamyl group from GSH and other γ-glutamyl compounds to amino acids or dipeptides. GSH is abundant in the cells and plays important roles in protection from oxidative stress and maintenance of the redox status [
26]. GGT initiates the degradation of extracellular GSH, resulting in production of cysteinylglycine and glutamate. Cysteinylglycine is then hydrolyzed by cell surface dipeptidase to generate glycine and cysteine. The degraded amino acids are used for de novo synthesis of GSH. In normal human tissues, strong GGT immunoreactivity was observed on the surface of renal proximal tubule cells, hepatic bile canaliculi and capillary endothelial cells within the nervous system [
27]. Secretory or absorptive cells in sweat glands, prostate, salivary gland ducts, bile ducts, pancreatic acini, intestinal crypts and testicular tubules were also GGT-positive. Among a family of GGT genes in the human genome [
28], GGT1, which is generally referred to as GGT, is shown to be involved in GSH metabolism [
29].
Elevation of GGT expression has been reported for a number of cancers including colon, ovary and liver cancer, astrocytic glioma, soft tissue sarcoma, melanoma and leukemia [
22]. A comprehensive analysis of GGT expression showed that most tumors derived from tissues expressing GGT were positive for GGT and that lung and ovary cancer derived from GGT-negative epithelia also expressed GGT [
22]. GGT expression was linked to unfavorable prognostic signs in breast cancer, but no correlation between GGT expression and standard clinical pathological parameters has been found in prostatic, colorectal and breast cancer [
22].
Upregulation of GGT expression in cancer has been considered to protect cancer cells against oxidative stress by increasing the intracellular GSH level and thereby support their growth and survival [
30]. However, it was also demonstrated that the metabolism of GSH by GGT can exert pro-oxidant effects [
31]. Upregulation of GGT may impose an increased oxidative burden on the cell, resulting in GSH consumption and a decrease of cellular GSH stores. The persistent production of ROS caused by increased GGT expression may contribute to genetic instability and tumor progression [
32].
Serum GGT activity is commonly used as a marker for liver, gallbladder and biliary tract diseases especially alcoholic liver disease because it is particularly sensitive to alcohol consumption [
33]. On the other hand, a positive association of serum GGT activity with the risk of cancer [
34,
35] as well as cardiovascular diseases and metabolic syndrome [
36] has been reported. Furthermore, serum GGT levels were found to be higher in hepatocellular carcinoma patients with poorly differentiated tumors as compared to those with well and moderately differentiated tumors [
37]. In renal cell carcinoma, serum GGT activity was reported to be increased in most of patients with metastasis, while it was normal in majority of patients with localized tumor [
38].
Franzini et al. performed gel filtration chromatography followed by postcolumn reaction with a fluorescent GGT substrate, gamma-glutamyl-7-amido-4-methylcoumarin (γGluAMC) and identified four GGT fractions in serum: big-GGT, medium-GGT, small-GGT and free-GGT fractions of different molecular weight (molecular masses >2000 kDa, 940 kDa, 140 kDa and 70 kDa, respectively) [
24]. The authors demonstrated that b-GGT increased in non-alcoholic fatty liver disease (NAFLD) but not in chronic hepatitis C (CHC) and that b-GGT/s-GGT ratio showed the highest diagnostic accuracy for distinguishing NAFLD and CHC [
39]. They also showed that the big-GGT fraction corresponds to serum exosomal GGT [
25].
In order to determine GGT activity on exosomes, we used a newly reported fluorescence probe, gGlu-HMRG, which is activated by rapid one-step cleavage of glutamate with GGT [
20]. This probe was developed to detect cancers cells during surgical and endoscopic procedures, taking advantage of its activation by GGT that is present on the cell surface. In vivo imaging of superficial head and neck squamous cell carcinoma and beast, lung and colorectal cancer using gGlu-HMRG has been reported [
40‐
43]. In vitro activation of gGlu-HMRG was also shown in human ovarian cancer cell lines [
20].
In the present study, we first showed correlation of GGT1 expression with GGT activity in cell lysates and exosomes. Second, we separated human serum by SEC and demonstrated that the minor peak that was positive for CD9 contained GGT1 large and small subunits as well as GGT activity and that the major peak was presumably comprised of medium-GGT, small-GGT and free-GGT fractions other than big-GGT or exosomal GGT fraction. Third, we subjected exosomes isolated from human serum by differential centrifugation to OptiPrep density gradient centrifugation and confirmed that exosomes isolated from human serum by differential centrifugation is free of contamination with other GGT forms. Lastly, based on these findings, we measured serum exosomal GGT activity in patients. Despite the fact that GGT1 was upregulated in exosomes isolated from androgen-independent C4–2 and bone metastatic C4–2B cells, there was no difference between PC patients with and without castration-resistance. Unexpectedly, we found that serum exosomal GGT activity was significantly higher in PC patients than in BPH patients.
In support of our findings of increased serum exosomal GGT activity in PC patients, GGT1 expression was elevated in PC tissues compared with BPH tissues. A previous report showed that the majority of neoplastic cells were positive for GGT1 in most of PC [
44]. In the present study, we demonstrated that there was a significant difference in GGT1 expression between PC and BPH tissues. Furthermore, cancer cells showed stronger expression for GGT1 in the cytoplasm and membrane than background noncancerous prostatic glands. These results suggested that prostatic cancer cells may produce more exosomes expressing GGT1. The underlying mechanism that is responsible for overexpression of GGT1 in PC remains to be elucidated.
Numerous reports have proposed potential markers for PC based on pathological and clinical research [
45]. More recently identified PC markers include prostate cancer antigen 3 (PCA3) [
46], TMPRSS2-ERG fusion gene [
47] and their combined use [
48]. Although there have been a limited number of reports describing exosomal miRNA as a marker for PC [
49], we and others have reported exosomal protein markers that would be helpful to diagnose PC (PSMA), taxane-resistant CRPC (P-gp) and progression and aggressiveness of PC (integrin β4 and vinculin) [
12‐
16]. This is the first report that described serum exosomal GGT1 expression or GGT activity as a potential marker to diagnose PC.
PSA is a commonly used marker for PC, but it cannot distinguish PC from BPH when the levels are similar [
9,
10]. In the present study, we measured serum exosomal GGT activity as well as serum GGT activity and serum PSA level in two patient groups. As shown in Additional file
4: Fig. S2, the AUC of serum exosomal GGT activity was 0.714 (95% CI between 0.535 and 0.892), while that of serum GGT activity was 0.621 (95% CI between 0.396 and 0.846) and that of serum PSA concentration was 0.601 (95% CI between 0.361 and 0.841). These results suggest that serum exosomal GGT activity but not serum GGT activity could be a biomarker to differentiate PC patients from BPH patients, both of which exhibit similar serum PSA levels.
Although we have demonstrated the potential of serum exosomal GGT activity for differential diagnosis of PC and BPH, the current detection system has limitations for clinical application, because differential centrifugation is required to measure the activity. It is also worth noting that GGT1 is expressed in normal tissues and thus serum exosomes isolated by differential centrifugation may contain those derived from various tissues. We and others have recently demonstrated that exosomes derived from PC could be isolated by immunocapture with anti-PSMA antibody [
12,
13]. The development of an antibody with a higher affinity for PSMA and its use would enable us to increase the specificity and sensitivity of serum exosomal GGT activity as a marker for PC.
The usefulness of serum exosomal GGT activity as a maker to diagnose PC needs to be validated in large-scale clinical studies. Since serum GGT activity has been implicated in a variety of diseases by clinical and epidemiological studies [
34‐
36,
50], it would be of great interest to test if serum exosomal GGT activity is superior to serum GGT activity in other diseases than PC. Nevertheless, in order to conduct large-scale studies, a simple and rapid detection system remains to be established, which would make it possible to evaluate the potential of serum exosomal GGT activity as prognostic as well as diagnostic markers in prospective clinical studies. Finally, it is also of great importance to understand the properties and roles of GGT1 on exosomes in serum of patients.
Acknowledgements
We thank Mr. Tsuyoshi Maruyama at Tokyo Metropolitan Geriatric Hospital and Mr. Yasuo Hasegawa at Tokyo Metropolitan Institute of Gerontology for their technical assistance. We also thank Drs. Hiroki Tsumoto and Yuri Miura at Tokyo Metropolitan Institute of Gerontology for their assistance in proteomic analysis.