GENES, CHROMOSOMES & CANCER 00:00–00 (2014)
RESEARCH ARTICLE
Different Exosome Cargo from Plasma/ Bronchoalveolar Lavage in Non-Small-Cell Lung Cancer Marta Rodrıguez,1† Javier Silva,1† Ana L opez-Alfonso,2 Marıa Belen L opez-Mu~ niz,3 Cristina Pe~ na,1 1 1 4 1 Gemma Domınguez, Jose Miguel Garcıa, Ana L opez-G onzalez, Miriam M endez, Mariano Provencio,1 Vanesa Garcıa,1* and F elix Bonilla1* 1
Department of Medical Oncology,University Hospital Puerta de Hierro Majadahonda,Madrid, Spain Department of Medical Oncology,University Hospital Infanta Leonor,Madrid, Spain 3 Department of Respiratory Medicine,University Hospital Infanta Leonor,Madrid, Spain 4 Department of Medical Oncology,University Hospital Leon,Le on, Spain 2
Tumor-derived exosomes mediate tumorigenesis by facilitating tumor growth, metastasis, development of drug resistance, and immunosuppression. However, little is known about the exosomes isolated from bronchoalveolar lavage (BAL) in patients with lung neoplasm. Exosomes isolated in plasma and BAL from 30 and 75 patients with tumor and nontumor pathology were quantified by acetylcholinesterase activity and characterized by Western Blot, Electron Microscopy, and Nanoparticle Tracking Analysis. Differences in exosome cargo were analyzed by miRNA quantitative PCR in pooled samples and validated in a second series of patients. More exosomes were detected in plasma than in BAL in both groups (P < 0.001). The most miRNAs evaluated by PCR array were detected in tumor plasma, tumor BAL, and nontumor BAL pools, but only 56% were detected in the nontumor plasma pool. Comparing the top miRNAs with the highest levels detected in each pool, we found close homology only between the BAL samples of the two pathologies. In tumor plasma, we found a higher percentage of miRNAs with increased levels than in tumor BAL or in nontumor plasma. The data reveal C 2014 Wiley Periodicals, Inc. V differences between BAL and plasma exosome amount and miRNA content.
INTRODUCTION
Analysis of bronchoalveolar lavage (BAL; Reynolds and Newball, 1974) facilitates the diagnosis of infectious, neoplastic, inflammatory, and immunological diseases of the alveolar area and distal airways. It is especially valuable in alveolar proteinosis, type-X histiocytosis, pulmonary infiltrates with eosinophilia, neoplastic pulmonary infiltration, pulmonary hemosiderosis, fat embolism, and lipoid pneumonia (Danel et al., 1992; Rennard, 1992). Exosomes are cell-derived vesicles with a diameter between 30 and 100 nm, which are protected by lipid-rich bound membranes. They are released from both normal and tumor cells, and they can transmit cargo from one cell to another. Their protein content and genetic material, including miRNA, depend on the cells from which they originate. This genetic material may have an epigenetic function of reprogramming in recipient cells, or evasion of the host response to tumors (Azmi et al., 2013; Raposo and Stoorvogel, 2013; Yuana et al., 2013). A wide range of cells release exosomes. They have been detected in physiological fluids such as C 2014 Wiley Periodicals, Inc. V
plasma (Caby et al., 2005), nasal washing liquid (L€asser et al., 2011), saliva (Palanisamy et al., 2010), BAL (Admyre et al., 2003), urine (Mitchell et al., 2009), maternal milk (Admyre et al., 2007), pleural effusions and ascites (Bard et al., 2004), cerebrospinal fluid (Street et al., 2012), and semen (Raimondo et al., 2011). The presence of exosomes in BAL in various clinical situations has also been reported. More BAL exosomes were found in granulomatosis and sarcoidosis patients than in healthy controls (Qazi et al., 2010); and Type II *Correspondence to: V. Garcıa, Department of Medical Oncology, Puerta de Hierro Majadahonda University Hospital, C/Joaquın Rodrigo, 2, E-28222 Majadahonda, Madrid, Spain. E-mail: vanesa. [email protected] or F. Bonilla, Department of Medical Oncology, Puerta de Hierro Majadahonda University Hospital, C/Joaquın Rodrigo, 2, E-28222 Majadahonda; Madrid, Spain. E-mail: [email protected] Supported by: Ministerio de Economıa y Competitividad SAF2010-20750; RD12/0036/0041; Fundaci on Banco Santander; Fundaci on AECC; Comunidad de Madrid CM: S/BMD-234; Fundaci on AECC fellowship (V. Garcıa). † Marta Rodrıguez and Javier Silva contributed equally to this study. Received 3 July 2013; Accepted 8 April 2014 DOI 10.1002/gcc.22181 Published online 00 Month 2014 in Wiley Online Library (wileyonlinelibrary.com).
RODRIGUEZ ET AL.
2
major histocompatibility complex molecules have been identified in exosomes extracted from BAL from healthy subjects (Admyre et al., 2003). It has been suggested that BAL exosomes might contribute to subclinical inflammation in asthmatics by increasing cytokine and leukotriene generation in the airway epithelium (Torregrosa et al., 2012). However, little is known about exosomes isolated from BAL in patients with lung neoplasm. Thus, as described in allergic processes in the bronchial tree (Admyre et al., 2008), we examined whether tumor cells release exosomes to the airspace. Our hypothesis is based on the ability of tumor cells to release exosomes and on the specific features of exosomes isolated from plasma and BAL in patients with non-small-cell lung cancer (NSCLC; Silva et al., 2011). We hypothesized that the content of exosomes released to plasma by tumor cells differs from that of exosomes released to the airway by the same tumor cells. This study aimed to identify differences in miRNA content between exosomes from plasma and those isolated from BAL in two groups: patients with lung cancer and patients without tumor. MATERIAL AND METHODS Patient Samples
Between January 2010 and May 2011, we recruited a consecutive series of blood and BAL samples from 30 NSCLC patients and 75 patients with nontumor pathology. Bronchoscopy was clinically indicated for complement the diagnosis. The tumors were classified as follows: 16 (54%) squamous cell carcinomas and 14 (46%) adenocarcinomas; patients’ median age was 69 years (45–83). Twenty-three (77%) were men and seven (23%) were women, with 26 (87%) smokers and 4 (13%) nonsmokers. The most frequent pathologies in patients without tumor were: inflammatoryinterstitial diseases 24 (32%) patients, infections 18 (24%), nontumor lung nodules 14 (18%), hemoptysis 11 (15%), and other diseases 8 (11%). The median age was 62 years (18–87) with 29 (39%) women and 46 (61%) men, with 52 (69%) smokers and 23 (31%) nonsmokers. Informed consent was obtained from all participants, as approved by the research ethics board of the Infanta Leonor University Hospital. Isolation of Exosomes
Plasma was prepared by centrifuging peripheral blood at 5003g for 25 min. Plasma and BAL exoGenes, Chromosomes & Cancer DOI 10.1002/gcc
somes were isolated by centrifugation: 5003g for 10 min to eliminate cells; supernatant was centrifuged at 17,0003g for 20 min, followed by passage through a 0.22 lm PVDF filter (Millipore, Ireland) and ultracentrifugation (OptimaTM MAX-XP, Beckman Coulter, Fullerton, CA) at 120,0003g for 90 min. The exosome pellet was washed and resuspended in 1mL of phosphate-buffered saline. Characterization and Quantification of Exosomes
Exosome protein extraction and Western Blot analyses were performed according to previously described protocols (Silva et al., 2012). Exosome proteins were extracted by RIPA Buffer (Sigma, St Louis, MO) and Halt Protease and Phosphatase Single-use Inhibitor Cocktail (Thermo, Rockford, IL). Proteins were denatured in 23 SDS buffer at 95 C, separated by 10% SDS-polyacrylamide gel electrophoresis and transferred by the iBlot System (Invitrogen, Paisley, UK). Membranes were blocked and incubated with primary and secondary antibodies. We used mouse monoclonal antibody for CD63 (1:500 dilution, ab59479, abcam, Cambridge, UK); and rabbit polyclonal antibody for Calnexin as loading control (1:200 dilution, sc11397, Santa Cruz Biotechnology, Inc., Santa Cruz, CA). Anti-mouse and anti-rabbit IRDye 800 CW infrared polyclonal secondary antibodies were used at a dilution of 1:10,000 (Rocklan, Gilbertsville, PA) to detect the target protein in the membrane. Proteins were detected and quantified by band densitometry, using the Odyssey Infrared Imaging System (LI-COR Biosciences, Cambridge, UK). For identification by transmission electron microscopy (model JEOL Jem1010, 100 kV), exosomes were fixed in 2% PFA (w/v) in 200 mM phosphate buffer (pH 7.4). Fixed exosomes were dropped onto a formvar/carbon-coated grid and left to dry at room temperature for 20 min. After washing in PBS, the exosomes were fixed in 1% glutaraldehyde for 5 min, washed in water, and stained with saturated aqueous uranyl oxalate for 5 min. Samples were then embedded in 0.4% (w/v) uranyl acetate and 1.8% (w/v) methylcellulose and incubated on ice for 10 min. The excess liquid was then removed. The grid was dried at room temperature for 10 min and viewed at 380,000 and 3120,000. The LM10 nanoparticle characterization system (NanoSight, Amesbury, UK) was used for realtime characterization and quantification of the vesicles.
EXOSOME CARGO FROM PLASMA AND BAL IN NSCLC
To quantify the exosomes, we measured the activity of acetylcholinesterase (AChE), an enzyme that is present in these vesicles. This method has been successful in the analysis of exosomes isolated from human lung adenocarcinoma, human mucoepidermoid bronchial carcinoma and colon and pancreas carcinoma cell lines (Rieu et al., 2000; Savina et al., 2002; Gastpar et al., 2005; Cantin et al., 2008; Chalmin et al., 2010; Merendino et al., 2010). Briefly, 40 lL of the exosome fraction was added to individual wells on a 96-well flat-bottomed microplate. 1.25 mM acetylthiocholine and 0.1 mM 5,50 -dithiobis(2-nitrobenzoic acid) were added to exosome fractions in a final volume of 300 lL. Changes in absorbance at 412 nm were monitored for 30 min. RNA Extraction and Retrotranscription
Total RNA was extracted from exosomes by the mirVanaTM miRNA Isolation Kit (Ambion Inc., TX) and retro-transcribed, using miScript II RT Kit (Qiagen, Valencia, CA) for miRNA quantitative PCR Array and TaqMan microRNA Reverse Transcription Kit (Applied Biosystem, Foster City, CA) for miRNA quantification in a validation set. cDNAs were quantified by NanoDrop ND1000 (Thermo Scientific, Wilmington, DE); and concentrations were equalized to 100 lg/lL to standardize per cDNA amount. miRNA Quantitative PCR Array and Real-Time PCR for Independent Series Validation
miRNA expression profiles were analyzed in R Green PCR each sample by the miScript SYBRV Kit (Qiagen), according to the manufacturer’s instructions. This system accurately analyzes the levels of 84 miRNA sequences in the same time as a real-time PCR instrument (Roche Diagnostics, Mannheim, Germany). We then compared the data from four pooled samples using relative quantifications. Ratios were calculated using 2Dct. For validation set analysis, relative quantificaR Microtion was performed using the TaqManV RNA Assay specific to each miRNA (Applied Biosystems), in line with the manufacturer’s instructions. The relative concentrations of each miRNA were calculated by interpolation using a standard curve for each respective miRNA generated by the serial dilution of cDNA from cell lines. Samples were validated independently for each miRNA, and were normalized using the mean value of miRNA evaluated.
3
miRNA Pathway Analysis and Statistical Analysis
miRNA-targeted pathway analysis was performed using the DNA Intelligent Analysis (DIANA)-miRPath v2.0 database. miRNA and pathway-related information for this database was obtained from miRBase 18 and the Kyoto Encyclopedia of Genes and Genomes (KEGG) v58.1. In silico target prediction was performed by DIANA-microT-CDS. The Union of Pathways mode was used to identify all the pathways significantly targeted by the selected microRNAs. We performed the False Discovery Rate method as a correction for multiple hypothesis testing and we selected Union of Pathways analysis. The server combines the previously calculated significance levels and provides a merged P-value for each pathway, by applying Fisher’s method (Vlachos et al., 2012). Concentrations of exosomes measured by NanoSight were contrasted by the paired Student’s t-test. AChE levels in patients were contrasted by the Mann–Whitney test. miRNA levels in the validation set were analyzed by the Mann– Whitney test. Results were considered significant where P < 0.05.
RESULTS Characterization and Quantification of Exosome Levels in Plasma and BAL Samples from Patients with NSCLC and Nontumor Pathologies
Initially, exosomes of plasma and BAL samples from 30 patients with NSCLC and 75 patients with nontumor pathology were isolated. Next, to confirm specific isolation of exosomes, four pools (each one mixing 15 cases) of exosomes isolated from plasma and BAL of patients with lung cancer (Tpl and Tbal pools, respectively), and from plasma and BAL of patients with nontumor pathology (NTpl and NTbal pools, respectively), were analyzed by a variety of methods. These pools were balanced for age, gender, and smoker/ nonsmoker. The identity of exosomes from the four pools was confirmed by electron microscopy (Fig. 1A). The presence of CD63, a specific exosome marker, with Calnexin as negative control, was confirmed by Western blot. CD63 protein was detected in plasma samples from both groups of patients and, to a lesser extent, in BAL samples, whereas all samples were negative for Calnexin, an endoplasmic reticulum marker (Fig. 1B). In addition, size distribution and levels of exosomes in the four pools were examined by NanoSight Genes, Chromosomes & Cancer DOI 10.1002/gcc
4
RODRIGUEZ ET AL.
Figure 1. Characterization and quantification of exosome levels in plasma and BAL samples from patients with NSCLC and nontumor pathologies. (A) Electron Microscopy showing isolated exosomes in four pools, NTpl, NTbal, Tpl, and Tbal. (B) Western Blot of exosome-specific
protein, CD63, and Calnexin loading control in the four pools. (C) Nanoparticle Tracking Analysis showing size and concentration of isolated exosomes in the four pools (***P < 0.001). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
(Fig. 1C). Size distribution was consistent with previously described in other studies. Exosomes isolated from plasma and BAL samples of 30 patients with NSCLC and 75 patients with nontumor pathology were quantified by AChE activity. In patients with lung cancer, exosomes were detected in 97 and 93% of plasma and BAL samples, respectively. Similarly, in patients with nontumor pathology, exosomes were detected in 92 and 89% of plasma and BAL samples, respectively. Plasma samples contained significantly higher concentrations of exosomes than BAL samples in both groups of patients (Fig. 2, A and B panels; mean rank shown in Table 1, P < 0.001 in both cases). Moreover, exosome levels in BAL samples were significantly higher in tumor patients than in nontumor, while no such difference was observed in plasma samples. (Fig. 2, C and D panels; mean rank shown in Table 1, P 5 0.441 and P 5 0.040, respectively). Therefore, on the basis of the levels of exosome proteins and AChE activity, more exosomes were released into
plasma than pathology.
Genes, Chromosomes & Cancer DOI 10.1002/gcc
into
BAL,
regardless
of
the
miRNA Profiles in Exosomes
The miScript miRNA quantitative PCR arrays were performed in the four pools: Tpl, Tbal, NTpl, and NTbal. As extraction from a single sample did not provide enough material, pooling was adopted as an alternative to microRNA amplification. Moreover, examination of pooled samples, rather than individual samples, reduces intersubject variability and furnishes a stable reference condition (Churchill, 2002; Gold et al., 2004; Han et al., 2004). Of the 84 miRNAs analyzed in this PCR array, 87% were detected in Tpl and Tbal pools, 56% in NTpl, and 88% in NTbal (Fig. 3). When population type was analyzed, one miRNA was detected specifically in tumor samples (miR122), but none was specific to nontumor samples. When sample type was studied, two miRNAs were specific to plasma (miR-126 and miR-144) and two others were specific to BAL (miR-302a and miR-
EXOSOME CARGO FROM PLASMA AND BAL IN NSCLC
5
Figure 2. Box plots showing the levels of exosomes measured by AChE and their distribution in the subpopulations of patients and sample types studied.
302c). Moreover, miR-143 was detected only in Tbal, and miR-128 was specific to NTpl. However, no specific miRNA was identified in the Tpl or NTbal groups. The functions of these specific miRNAs are listed in Table 2. The top 10 enriched miRNAs with the highest levels detected in each pool are shown in Table 3. Thirty percentage of the miRNAs were detected in equal amounts in NTpl and NTbal, 40% were equal in Tpl and Tbal, 40% in Tpl and NTpl, and 80% in Tbal and NTbal. These results suggest that the miRNAs released in exosomes to plasma are different from those released into the airway, regardless of the pathology. A high percentage of miRNAs were common to BAL samples from tumor and nontumor patients, but this coincidence
was not observed in plasma samples. This contrast could be explained by a selective release in tumor pathology of miRNAs from exosomes to plasma, but not to BAL. Differences in the miRNA Levels in Plasma and BAL of Patients with NSCLC and Nontumor Pathology
We compared the concentrations of miRNAs in exosomes in the following pools: (a) plasma versus BAL in tumor patients, (b) plasma versus BAL in nontumor patients, (c) plasma of tumor versus nontumor patients, and (d) BAL of tumor versus nontumor patients. Table 4 shows fold-changes in expression of the miRNAs. In tumor patients, Genes, Chromosomes & Cancer DOI 10.1002/gcc
RODRIGUEZ ET AL.
6
TABLE 1. Associations Between Exosome Levels in Plasma and BAL Samples of NSCLC and Nontumor Patients
Tumor Nontumor Plasma BAL
Plasma BAL Plasma BAL Tumor Nontumor Tumor Nontumor
N
Mean rank
P
30 30 75 75 30 75 30 75
44 17 107.1 43.9 56.6 51.6 62.6 49.2
RESEARCH ARTICLE
Different Exosome Cargo from Plasma/ Bronchoalveolar Lavage in Non-Small-Cell Lung Cancer Marta Rodrıguez,1† Javier Silva,1† Ana L opez-Alfonso,2 Marıa Belen L opez-Mu~ niz,3 Cristina Pe~ na,1 1 1 4 1 Gemma Domınguez, Jose Miguel Garcıa, Ana L opez-G onzalez, Miriam M endez, Mariano Provencio,1 Vanesa Garcıa,1* and F elix Bonilla1* 1
Department of Medical Oncology,University Hospital Puerta de Hierro Majadahonda,Madrid, Spain Department of Medical Oncology,University Hospital Infanta Leonor,Madrid, Spain 3 Department of Respiratory Medicine,University Hospital Infanta Leonor,Madrid, Spain 4 Department of Medical Oncology,University Hospital Leon,Le on, Spain 2
Tumor-derived exosomes mediate tumorigenesis by facilitating tumor growth, metastasis, development of drug resistance, and immunosuppression. However, little is known about the exosomes isolated from bronchoalveolar lavage (BAL) in patients with lung neoplasm. Exosomes isolated in plasma and BAL from 30 and 75 patients with tumor and nontumor pathology were quantified by acetylcholinesterase activity and characterized by Western Blot, Electron Microscopy, and Nanoparticle Tracking Analysis. Differences in exosome cargo were analyzed by miRNA quantitative PCR in pooled samples and validated in a second series of patients. More exosomes were detected in plasma than in BAL in both groups (P < 0.001). The most miRNAs evaluated by PCR array were detected in tumor plasma, tumor BAL, and nontumor BAL pools, but only 56% were detected in the nontumor plasma pool. Comparing the top miRNAs with the highest levels detected in each pool, we found close homology only between the BAL samples of the two pathologies. In tumor plasma, we found a higher percentage of miRNAs with increased levels than in tumor BAL or in nontumor plasma. The data reveal C 2014 Wiley Periodicals, Inc. V differences between BAL and plasma exosome amount and miRNA content.
INTRODUCTION
Analysis of bronchoalveolar lavage (BAL; Reynolds and Newball, 1974) facilitates the diagnosis of infectious, neoplastic, inflammatory, and immunological diseases of the alveolar area and distal airways. It is especially valuable in alveolar proteinosis, type-X histiocytosis, pulmonary infiltrates with eosinophilia, neoplastic pulmonary infiltration, pulmonary hemosiderosis, fat embolism, and lipoid pneumonia (Danel et al., 1992; Rennard, 1992). Exosomes are cell-derived vesicles with a diameter between 30 and 100 nm, which are protected by lipid-rich bound membranes. They are released from both normal and tumor cells, and they can transmit cargo from one cell to another. Their protein content and genetic material, including miRNA, depend on the cells from which they originate. This genetic material may have an epigenetic function of reprogramming in recipient cells, or evasion of the host response to tumors (Azmi et al., 2013; Raposo and Stoorvogel, 2013; Yuana et al., 2013). A wide range of cells release exosomes. They have been detected in physiological fluids such as C 2014 Wiley Periodicals, Inc. V
plasma (Caby et al., 2005), nasal washing liquid (L€asser et al., 2011), saliva (Palanisamy et al., 2010), BAL (Admyre et al., 2003), urine (Mitchell et al., 2009), maternal milk (Admyre et al., 2007), pleural effusions and ascites (Bard et al., 2004), cerebrospinal fluid (Street et al., 2012), and semen (Raimondo et al., 2011). The presence of exosomes in BAL in various clinical situations has also been reported. More BAL exosomes were found in granulomatosis and sarcoidosis patients than in healthy controls (Qazi et al., 2010); and Type II *Correspondence to: V. Garcıa, Department of Medical Oncology, Puerta de Hierro Majadahonda University Hospital, C/Joaquın Rodrigo, 2, E-28222 Majadahonda, Madrid, Spain. E-mail: vanesa. [email protected] or F. Bonilla, Department of Medical Oncology, Puerta de Hierro Majadahonda University Hospital, C/Joaquın Rodrigo, 2, E-28222 Majadahonda; Madrid, Spain. E-mail: [email protected] Supported by: Ministerio de Economıa y Competitividad SAF2010-20750; RD12/0036/0041; Fundaci on Banco Santander; Fundaci on AECC; Comunidad de Madrid CM: S/BMD-234; Fundaci on AECC fellowship (V. Garcıa). † Marta Rodrıguez and Javier Silva contributed equally to this study. Received 3 July 2013; Accepted 8 April 2014 DOI 10.1002/gcc.22181 Published online 00 Month 2014 in Wiley Online Library (wileyonlinelibrary.com).
RODRIGUEZ ET AL.
2
major histocompatibility complex molecules have been identified in exosomes extracted from BAL from healthy subjects (Admyre et al., 2003). It has been suggested that BAL exosomes might contribute to subclinical inflammation in asthmatics by increasing cytokine and leukotriene generation in the airway epithelium (Torregrosa et al., 2012). However, little is known about exosomes isolated from BAL in patients with lung neoplasm. Thus, as described in allergic processes in the bronchial tree (Admyre et al., 2008), we examined whether tumor cells release exosomes to the airspace. Our hypothesis is based on the ability of tumor cells to release exosomes and on the specific features of exosomes isolated from plasma and BAL in patients with non-small-cell lung cancer (NSCLC; Silva et al., 2011). We hypothesized that the content of exosomes released to plasma by tumor cells differs from that of exosomes released to the airway by the same tumor cells. This study aimed to identify differences in miRNA content between exosomes from plasma and those isolated from BAL in two groups: patients with lung cancer and patients without tumor. MATERIAL AND METHODS Patient Samples
Between January 2010 and May 2011, we recruited a consecutive series of blood and BAL samples from 30 NSCLC patients and 75 patients with nontumor pathology. Bronchoscopy was clinically indicated for complement the diagnosis. The tumors were classified as follows: 16 (54%) squamous cell carcinomas and 14 (46%) adenocarcinomas; patients’ median age was 69 years (45–83). Twenty-three (77%) were men and seven (23%) were women, with 26 (87%) smokers and 4 (13%) nonsmokers. The most frequent pathologies in patients without tumor were: inflammatoryinterstitial diseases 24 (32%) patients, infections 18 (24%), nontumor lung nodules 14 (18%), hemoptysis 11 (15%), and other diseases 8 (11%). The median age was 62 years (18–87) with 29 (39%) women and 46 (61%) men, with 52 (69%) smokers and 23 (31%) nonsmokers. Informed consent was obtained from all participants, as approved by the research ethics board of the Infanta Leonor University Hospital. Isolation of Exosomes
Plasma was prepared by centrifuging peripheral blood at 5003g for 25 min. Plasma and BAL exoGenes, Chromosomes & Cancer DOI 10.1002/gcc
somes were isolated by centrifugation: 5003g for 10 min to eliminate cells; supernatant was centrifuged at 17,0003g for 20 min, followed by passage through a 0.22 lm PVDF filter (Millipore, Ireland) and ultracentrifugation (OptimaTM MAX-XP, Beckman Coulter, Fullerton, CA) at 120,0003g for 90 min. The exosome pellet was washed and resuspended in 1mL of phosphate-buffered saline. Characterization and Quantification of Exosomes
Exosome protein extraction and Western Blot analyses were performed according to previously described protocols (Silva et al., 2012). Exosome proteins were extracted by RIPA Buffer (Sigma, St Louis, MO) and Halt Protease and Phosphatase Single-use Inhibitor Cocktail (Thermo, Rockford, IL). Proteins were denatured in 23 SDS buffer at 95 C, separated by 10% SDS-polyacrylamide gel electrophoresis and transferred by the iBlot System (Invitrogen, Paisley, UK). Membranes were blocked and incubated with primary and secondary antibodies. We used mouse monoclonal antibody for CD63 (1:500 dilution, ab59479, abcam, Cambridge, UK); and rabbit polyclonal antibody for Calnexin as loading control (1:200 dilution, sc11397, Santa Cruz Biotechnology, Inc., Santa Cruz, CA). Anti-mouse and anti-rabbit IRDye 800 CW infrared polyclonal secondary antibodies were used at a dilution of 1:10,000 (Rocklan, Gilbertsville, PA) to detect the target protein in the membrane. Proteins were detected and quantified by band densitometry, using the Odyssey Infrared Imaging System (LI-COR Biosciences, Cambridge, UK). For identification by transmission electron microscopy (model JEOL Jem1010, 100 kV), exosomes were fixed in 2% PFA (w/v) in 200 mM phosphate buffer (pH 7.4). Fixed exosomes were dropped onto a formvar/carbon-coated grid and left to dry at room temperature for 20 min. After washing in PBS, the exosomes were fixed in 1% glutaraldehyde for 5 min, washed in water, and stained with saturated aqueous uranyl oxalate for 5 min. Samples were then embedded in 0.4% (w/v) uranyl acetate and 1.8% (w/v) methylcellulose and incubated on ice for 10 min. The excess liquid was then removed. The grid was dried at room temperature for 10 min and viewed at 380,000 and 3120,000. The LM10 nanoparticle characterization system (NanoSight, Amesbury, UK) was used for realtime characterization and quantification of the vesicles.
EXOSOME CARGO FROM PLASMA AND BAL IN NSCLC
To quantify the exosomes, we measured the activity of acetylcholinesterase (AChE), an enzyme that is present in these vesicles. This method has been successful in the analysis of exosomes isolated from human lung adenocarcinoma, human mucoepidermoid bronchial carcinoma and colon and pancreas carcinoma cell lines (Rieu et al., 2000; Savina et al., 2002; Gastpar et al., 2005; Cantin et al., 2008; Chalmin et al., 2010; Merendino et al., 2010). Briefly, 40 lL of the exosome fraction was added to individual wells on a 96-well flat-bottomed microplate. 1.25 mM acetylthiocholine and 0.1 mM 5,50 -dithiobis(2-nitrobenzoic acid) were added to exosome fractions in a final volume of 300 lL. Changes in absorbance at 412 nm were monitored for 30 min. RNA Extraction and Retrotranscription
Total RNA was extracted from exosomes by the mirVanaTM miRNA Isolation Kit (Ambion Inc., TX) and retro-transcribed, using miScript II RT Kit (Qiagen, Valencia, CA) for miRNA quantitative PCR Array and TaqMan microRNA Reverse Transcription Kit (Applied Biosystem, Foster City, CA) for miRNA quantification in a validation set. cDNAs were quantified by NanoDrop ND1000 (Thermo Scientific, Wilmington, DE); and concentrations were equalized to 100 lg/lL to standardize per cDNA amount. miRNA Quantitative PCR Array and Real-Time PCR for Independent Series Validation
miRNA expression profiles were analyzed in R Green PCR each sample by the miScript SYBRV Kit (Qiagen), according to the manufacturer’s instructions. This system accurately analyzes the levels of 84 miRNA sequences in the same time as a real-time PCR instrument (Roche Diagnostics, Mannheim, Germany). We then compared the data from four pooled samples using relative quantifications. Ratios were calculated using 2Dct. For validation set analysis, relative quantificaR Microtion was performed using the TaqManV RNA Assay specific to each miRNA (Applied Biosystems), in line with the manufacturer’s instructions. The relative concentrations of each miRNA were calculated by interpolation using a standard curve for each respective miRNA generated by the serial dilution of cDNA from cell lines. Samples were validated independently for each miRNA, and were normalized using the mean value of miRNA evaluated.
3
miRNA Pathway Analysis and Statistical Analysis
miRNA-targeted pathway analysis was performed using the DNA Intelligent Analysis (DIANA)-miRPath v2.0 database. miRNA and pathway-related information for this database was obtained from miRBase 18 and the Kyoto Encyclopedia of Genes and Genomes (KEGG) v58.1. In silico target prediction was performed by DIANA-microT-CDS. The Union of Pathways mode was used to identify all the pathways significantly targeted by the selected microRNAs. We performed the False Discovery Rate method as a correction for multiple hypothesis testing and we selected Union of Pathways analysis. The server combines the previously calculated significance levels and provides a merged P-value for each pathway, by applying Fisher’s method (Vlachos et al., 2012). Concentrations of exosomes measured by NanoSight were contrasted by the paired Student’s t-test. AChE levels in patients were contrasted by the Mann–Whitney test. miRNA levels in the validation set were analyzed by the Mann– Whitney test. Results were considered significant where P < 0.05.
RESULTS Characterization and Quantification of Exosome Levels in Plasma and BAL Samples from Patients with NSCLC and Nontumor Pathologies
Initially, exosomes of plasma and BAL samples from 30 patients with NSCLC and 75 patients with nontumor pathology were isolated. Next, to confirm specific isolation of exosomes, four pools (each one mixing 15 cases) of exosomes isolated from plasma and BAL of patients with lung cancer (Tpl and Tbal pools, respectively), and from plasma and BAL of patients with nontumor pathology (NTpl and NTbal pools, respectively), were analyzed by a variety of methods. These pools were balanced for age, gender, and smoker/ nonsmoker. The identity of exosomes from the four pools was confirmed by electron microscopy (Fig. 1A). The presence of CD63, a specific exosome marker, with Calnexin as negative control, was confirmed by Western blot. CD63 protein was detected in plasma samples from both groups of patients and, to a lesser extent, in BAL samples, whereas all samples were negative for Calnexin, an endoplasmic reticulum marker (Fig. 1B). In addition, size distribution and levels of exosomes in the four pools were examined by NanoSight Genes, Chromosomes & Cancer DOI 10.1002/gcc
4
RODRIGUEZ ET AL.
Figure 1. Characterization and quantification of exosome levels in plasma and BAL samples from patients with NSCLC and nontumor pathologies. (A) Electron Microscopy showing isolated exosomes in four pools, NTpl, NTbal, Tpl, and Tbal. (B) Western Blot of exosome-specific
protein, CD63, and Calnexin loading control in the four pools. (C) Nanoparticle Tracking Analysis showing size and concentration of isolated exosomes in the four pools (***P < 0.001). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
(Fig. 1C). Size distribution was consistent with previously described in other studies. Exosomes isolated from plasma and BAL samples of 30 patients with NSCLC and 75 patients with nontumor pathology were quantified by AChE activity. In patients with lung cancer, exosomes were detected in 97 and 93% of plasma and BAL samples, respectively. Similarly, in patients with nontumor pathology, exosomes were detected in 92 and 89% of plasma and BAL samples, respectively. Plasma samples contained significantly higher concentrations of exosomes than BAL samples in both groups of patients (Fig. 2, A and B panels; mean rank shown in Table 1, P < 0.001 in both cases). Moreover, exosome levels in BAL samples were significantly higher in tumor patients than in nontumor, while no such difference was observed in plasma samples. (Fig. 2, C and D panels; mean rank shown in Table 1, P 5 0.441 and P 5 0.040, respectively). Therefore, on the basis of the levels of exosome proteins and AChE activity, more exosomes were released into
plasma than pathology.
Genes, Chromosomes & Cancer DOI 10.1002/gcc
into
BAL,
regardless
of
the
miRNA Profiles in Exosomes
The miScript miRNA quantitative PCR arrays were performed in the four pools: Tpl, Tbal, NTpl, and NTbal. As extraction from a single sample did not provide enough material, pooling was adopted as an alternative to microRNA amplification. Moreover, examination of pooled samples, rather than individual samples, reduces intersubject variability and furnishes a stable reference condition (Churchill, 2002; Gold et al., 2004; Han et al., 2004). Of the 84 miRNAs analyzed in this PCR array, 87% were detected in Tpl and Tbal pools, 56% in NTpl, and 88% in NTbal (Fig. 3). When population type was analyzed, one miRNA was detected specifically in tumor samples (miR122), but none was specific to nontumor samples. When sample type was studied, two miRNAs were specific to plasma (miR-126 and miR-144) and two others were specific to BAL (miR-302a and miR-
EXOSOME CARGO FROM PLASMA AND BAL IN NSCLC
5
Figure 2. Box plots showing the levels of exosomes measured by AChE and their distribution in the subpopulations of patients and sample types studied.
302c). Moreover, miR-143 was detected only in Tbal, and miR-128 was specific to NTpl. However, no specific miRNA was identified in the Tpl or NTbal groups. The functions of these specific miRNAs are listed in Table 2. The top 10 enriched miRNAs with the highest levels detected in each pool are shown in Table 3. Thirty percentage of the miRNAs were detected in equal amounts in NTpl and NTbal, 40% were equal in Tpl and Tbal, 40% in Tpl and NTpl, and 80% in Tbal and NTbal. These results suggest that the miRNAs released in exosomes to plasma are different from those released into the airway, regardless of the pathology. A high percentage of miRNAs were common to BAL samples from tumor and nontumor patients, but this coincidence
was not observed in plasma samples. This contrast could be explained by a selective release in tumor pathology of miRNAs from exosomes to plasma, but not to BAL. Differences in the miRNA Levels in Plasma and BAL of Patients with NSCLC and Nontumor Pathology
We compared the concentrations of miRNAs in exosomes in the following pools: (a) plasma versus BAL in tumor patients, (b) plasma versus BAL in nontumor patients, (c) plasma of tumor versus nontumor patients, and (d) BAL of tumor versus nontumor patients. Table 4 shows fold-changes in expression of the miRNAs. In tumor patients, Genes, Chromosomes & Cancer DOI 10.1002/gcc
RODRIGUEZ ET AL.
6
TABLE 1. Associations Between Exosome Levels in Plasma and BAL Samples of NSCLC and Nontumor Patients
Tumor Nontumor Plasma BAL
Plasma BAL Plasma BAL Tumor Nontumor Tumor Nontumor
N
Mean rank
P
30 30 75 75 30 75 30 75
44 17 107.1 43.9 56.6 51.6 62.6 49.2
