UvA-DARE is a service provided by the library of the University of Amsterdam (http://dare.uva.nl)

UvA-DARE (Digital Academic Repository)

Extracellular vesicles for clinical diagnostics of nervous system diseases

Atai, N.

Link to publication

Citation for published version (APA):Atai, N. (2014). Extracellular vesicles for clinical diagnostics of nervous system diseases.

General rightsIt is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s),other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).

Disclaimer/Complaints regulationsIf you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, statingyour reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Askthe Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam,The Netherlands. You will be contacted as soon as possible.

Download date: 03 Jul 2020

!

Chapter 9: Analysis of extracellular vesicles in vitreal fluid. Development of diagnostic

and prognostic biomarkers of vitreal diseases

Running title: Extracellular vesicles in vitreal fluid

Nadia A. Atai1,2,3,4, Michael S. Hughes2, Sara Sivaraman1,2, Jamie L. Metzinger3, Henk A. van

Veen4, Vimal Sarup3, Claudia Castiblanco3, Sandra Hu-Torres3, Rienk Nieuwland5, Reinier

Schlingemann6, Ingeborg Klaassen6, Cornelis J.F. Van Noorden4,6, C. Stephen Foster3, Fred

H. Hochberg2,7*

1Department of Neurosurgery, Massachusetts General Hospital, Boston, Massachusetts, USA, 2Department of Neurology and Program in Neuroscience, Massachusetts General Hospital and

Harvard Medical School, Boston, Massachusetts, USA. 3Massachusetts Eye Research and

Surgery Institute (MERSI) and Harvard Medical School, Boston, Massachusetts

4Department of Cell Biology and Histology, The Van Leeuwenhoek Center for Advanced

Microscopy (LCAM), Academic Medical Center, University of Amsterdam, Amsterdam, The

Netherlands. 5Department of Clinical Chemistry, Academic Medical Center, University of

Amsterdam, Amsterdam, The Netherlands. 6Ocular Angiogenesis Research Group,

Departments of Ophthalmology and Cell Biology and Histology, Academic Medical Center,

University of Amsterdam, Amsterdam, The Netherlands. 7Pappas Center, Massachusetts

General Hospital, Boston, Massachusetts.

Keywords: Exosomes, extracellular vesicles, vitreous, central nervous system tumors, ocular

lymphoma, biomarker, uveitis

*Corresponding author:

Fred H. Hochberg, MD,

Massachusetts General Hospital Cancer Center

Stephen E. & Catherine Pappas Center for Neuro-Oncology

55 Fruit Street, YAW 9E

Boston, Massachusetts 02114

E-mail: fhochberg@partners.org

129

!

Abstract

Uveitis masquerade syndromes of the vitreous often present diagnostic dilemmas, especially

in the face of partial or complete failure to respond to standard immunosuppressive therapies.

The gold standard in diagnostic testing has been the vitreous biopsy, despite variabilities in

results; the yield of variation has been reported between 14.3-61.5%. Extracellular vesicles

(EVs) released from various cell types into biofluids such as the blood and cerebrospinal fluid

(CSF) contain genetic information that can serve as diagnostic biomarkers of diseases. In the

present study, we have investigated whether EVs are present in vitreous and can be used for

diagnosis and monitoring of response to therapy, using biomarkers such as vascular

endothelial growth factor (VEGF). Vitreous specimens taken at the time of diagnostic

vitrectomy for inflammatory/malignant uveitis were studied in 23 patients. Seven specimens

were collected and processed using NanoSight and transmission electron microscopy (TEM)

and 16 specimens were used for detection of total vitreous-derived EV-RNA using a

bioanalyzer followed by qRT-PCR for detection of human GAPDH, CD63, VEGF and EGFR

mRNA. We demonstrate that EV assays of vitreous can be used for diagnosis of eye diseases

from pauci-cellular ocular fluids. NanoSight and TEM revealed the presence of EVs in 7 out

of 7 vitreous samples of patients. We also show that mRNA of GAPDH, CD63 and VEGF

can be detected in vitreous-derived EVs in 16 out of 16 vitreous samples of patients with

various eye diseases. In conclusion, we have identified EVs in all vitrectomy fluids

investigated of patients with vitreal and retinal diseases. We have developed techniques for

the isolation and characterization of these EVs from small volumes of pauci-cellular vitreal

fluid. We have developed techniques for the isolation and characterization of RNA contents

of EVs in all the vitreal fluid samples investigated to provide a novel diagnostic approach to

vitreal and retinal diseases. We have developed techniques to measure the total RNA of

vitreous-derived EVs and to identify levels of mRNA of GAPDH, CD63, VEGF and EGFR.

Therefore, we propose EVs in vitreous fluid as the new gold standard for diagnosis and

monitoring of response to therapy.

130

!

Introduction

Uveitis “masquerade syndromes” are ocular diseases seemingly inflammatory processes1, for

which vitrectomy is performed. These uveitis syndromes fall into two categories:

inflammation of vitreous and malignant invasion of tumor cells such lymphoma and

leukemia2;3;4;5. Cytopathological and polymerase chain reaction (PCR) diagnoses are limited

by the paucity of fluid and the fragility of cells in specimens. Diagnostic tests are not

available for the vast majority of viral, inflammatory or vasculitic diseases of the vitreous and

the retina. Vitrectomy is performed more than 100,000 times each year prior to therapy with

intraocular corticosteroids or immunosuppressive agents6. Moreover, criteria of response

based upon vitreal studies are limited.

Extracellular vesicles (EVs) are formed by all types of cells and are released into the

extracellular environment and this process is conserved in prokaryotes and eukaryotes which

plays a major role in development7-9. EVs transport RNA and protein within and between

cells to facilitate cell-to-cell communication10;11. The specific proteins and genetic

information from EVs mediate immune functions, coagulation, angiogenesis and regulate

features of tumor progression10;12. We hypothesized that EVs and their contents, present in

vitreal fluid, can be evaluated as biomarkers for the diseased eye. EVs have been identified in

biofluids: (1) urine13; (2) serum8 (3) plasma14; (4) cerebrospinal fluid15; (5) nasal lavage

fluid16; (6) seminal fluid17 and (7) saliva18.

The objective of this study was to create novel approaches to analysis of small volume, pauci-

cellular vitreal fluid from which EV detection and characterization could be performed.

Material and Methods

Subjects

With Institutional Review Board (New England Institutional Review Board) approval,

physician investigators (CSF, VS) prospectively identified patients about to undergo and

consented for vitrectomy at the Massachusetts Eye Research and Surgery Institute as well as

the Massachusetts Eye and Ear Infirmary. Patients were included if: !18 years old, had

provided prior consent for vitrectomy and were able to comprehend the English language

consent form defining the proposed scientific use of discarded vitrectomy fluid. The exclusion

criteria were: <18 years old; known contra-indication for diagnostic and/or therapeutic

vitrectomy. Gender, race, and ethnicity did not influence a patient’s eligibility for inclusion in

131

!

study. Twenty three diagnostic vitrectomy patients provided fluid for study (Table 1). The age

range of patients was between 18 years and 92 years. Pre-vitrectomy diagnoses included

intermediate uveitis (n=4), Birdshot chorioretinopathy (n=4), panuveitis (n=3), macular

degeneration (n=2) ocular B cell lymphoma (n=1), posterior uveitis (n=1), autoimmune

uveitis (n=1), sarcoidosis (n=1), chronic pars planitis (n=1), iritis (n=1), anterior uveitis (n=1)

and unknown ocular pathology (n=1). Fifteen patients were known to have previously

received systemic medications and all patients have been provided ocular medications. One

patient (V-RNA 2) was the recipient of a therapeutic vitrectomy. Potentially predisposing

illnesses included cancer (n=5), viral infection (n=3), arthritis (n=3), sarcoidosis (n=2) colitis

(n=1), polymyalgia (n=1), skin infection (n=1), B27+ arthropathy (n=1) and unknown (n=8).

Isolation of EV from vitreal fluid

EV isolation techniques were optimized for small volume viscous vitreal fluid with sparse

cells. Fluid (between 200-500µl) was obtained within a 3ml syringe without anticoagulant,

prior to washing, using a 23 guage vitrectomy canula. The sample was immediately labeled

and kept at 4°C followed by storage at -80°C. The schematic novel methodology is provided

(Figure 1). Next, for experimental purposes vitreous samples were thawed at 4°C and diluted

to a total volume of 2ml with cold twice filtered PBS (2xfPBS). 0.2 µm filters (Thermo

Scientific, Waltham MA) were used to filter cold PBS. The diluted volume of vitreous fluid in

2xfPBS was filtered using an 0.8 µm filter (Thermo Scientific, Waltham MA). The resulting

fluid was re-diluted to a final volume of 2.3 ml and then centrifuged at 100,000 xg for 2hrs at

4°C in a MLA-55 rotor (Beckman Coulter, Indianapolis, IN). The pellet of vitreal EVs, was

used for further experiments.

NanoSight, nanoparticle tracking analysis (NTA)

NTA analysis (Fig. 2) is a state-of-the-art laser detection system for sizing and quantification

EVs in a fluid. Purified pelleted EVs were resuspended in 100µl 2xf PBS and 50µl of the

sample was used for determination of vitreal EV size and concentration using NTA with a

nanoparticles analyzer (NanoSight LM10, Amesbury, UK) utilizing automode NTA software

version 2.2. To prepare the vitreal EVs for NTA analyses we optimized a unique dilution

(1:100 in 2xfPBS) of seven (patients N/T1-7, Table 1) vitreal samples.

132

!

Transmission electron microscopy (TEM)

The remaining 50µl of vitreal EVs in 2xf PBS for each sample (N/T1-7 table 1) was fixed

with 4% freshly-prepared para-formaldehyde before being processed (HvV, CJFvN). Next,

fixed vitreal EVs were washed with 100 mM HEPES buffer (pH 7.3) containing 150 mM

NaCl, 2 mM CaCl2 and 10 mM MgCl2. Ten µl of the suspension was placed on coverslips

after which freshly carbon-coated grids were placed for 60 sec. The grids were lifted and

excess suspension was removed by touching with a filter paper at an angle of 45°. Grids were

placed for 60 sec on top of a drop of 2% uranyl acetate after which grids were air-dried. The

grids were examined with a Technai-12 G2 Spirit Biotwin transmission electron microscope

(FEI, Eindhoven, The Netherlands).

RNA extraction from EVs

Purified pelleted EVs of next 16 patients were used for RNA extraction and analysis. Again,

RNA extraction of vitreous EV pellets was performed after the EVs had been treated with

DNAse (DNA-free kit; Ambion®, Life Technologies, Grand Island, NY) and RNAase

inhibitors (Fermentas, Thermo Scientific). EVs were then lysed in 700 "l of QIAzol reagent

(Qiagen, Valencia, CA) NS Nucleic acids extracted using miRNeasy kit (Qiagen) according

to the manufacturers’ recommendations. Total RNA, eluted in 30 "l of RNAse-free water,

was characterized by bioanalysis using the RNA 6000 pico chip (2100 Bioanalyzer, Agilent,

Santa Clara, CA).

cDNA synthesis

Available amounts of RNA (200-2000pg RNA) were used as input for the cDNA reaction

using the SuperScrip® VILO™ (Invitrogen, Life Technologies) in a total volume of 20 "l.

The cDNA synthesis program consisted of 1cycle at 95°C for 3 min, 40 cycles at 95°C for 30

min sec, 60°C for 30s, 70°C for 30s, 1 cycle 70°C for 7 min.

qRT-PCR

Before using human samples to detect mRNA of interest in low levels of vitreal derived EV-

RNA/cDNA, the Taqman probes and primers (Applied Biosystems, Life Technologies) for

GAPDH (primer sequences: Hs03929097_g1), CD63 (primer sequences: Hs00156390_m1),

EGFR and VEGF (primer sequences: Hs00900055_m1) were optimized in low levels of

133

!

cellular RNA/cDNA and EV-RNA/EV-cDNA derived from epidermoid carcinoma, a VEGF-

expressing cell line A431 (Supp. Fig. 1). Next, human vitreal deriver EV-RNA/cDNA was

used to measure the expression levels of GADPH, CD63, VEGF and EGFR. One "l of vitreal

cDNA reaction volume (containing approximately 22pg cDNA per reaction) was used for

each qRT-PCR reaction. qRT-PCR reactions were performed in 25 "l reaction volumes

including fast TaqMan MasterMix (Applied Biosystems, Life Technologies) and nuclease-

free water. Amplification conditions consisted of 50°C, 2 min; 95°C, 10 min; 40 cycles of

95°C, 15 s, 60°C, 1 min on standard mode performed using ABI PRISM 7500 (Applied

Biosystems, Life Technologies).

Statistics

Statistical analysis was performed using Graphpad Prism® software (version 5.01; Graphpad,

La Jolla, CA) and Excel (Microsoft, Redmond, WA). We performed unpaired t-test evaluation

of cycle threshold (Ct) values. The Ct value is the number of PCR cycles required for the

fluorescent signal to exceed the background signal level.

Results

EV detection in vitreal fluids

Vitreal fluid at average volume of 200-500"l at diagnostic/therapeutic vitrectomy was

obtained from 23 patients (Table 1). Materials were obtained according to our standard

operating procedures in special tubes, labeled, ultracentrifuged and evaluated. From the first

seven patients EVs were characterized using NTA and TEM. By NTA analyses all 7 vitreous

samples contained EVs sized between 50 and 500nm. Figure 2A shows NTA analysis of a

single exemplar fluid revealing that EV size in this sample ranges between 71nm (arrow 1)

and 600nm (arrow 2); a spectrum seen in majority of processed samples. NTA analyses for

EV concentration in these vitreal samples varied between 125 and 450 EVs x 108/ml (Figure

2B). By TEM all seven samples exhibited vesicles surrounded by bilayer lipid membranes

(Figure 3). TEM showed larger EVs (50-800 nm) compared with/to NTA sizing (50 and

500nm). In addition, TEM images indicated various shapes of EVs such as round and

elongated and some samples contained clustered of small EVs (N/T 2 and N/T 7). However,

estimated EV concentrations detected by NTA were similar to those by TEM imaging (data

not shown).

134

!

RNA extraction from EVs

Vitreal EVs of 16 patients contained EV-derived RNA (V-RNA 1-16) detected by bioanalyzer

with large amounts of small RNAs and ribosomal RNAs (18s and 28s) as shown in figure 4.

EV-RNA concentrations varied in patient samples between 0.154ng/ul-215ng/ul. Majority of

the specimen contained low levels of EV-RNA independent of the vitreal volume. The highest

EV-RNA concentrations were identified in patients with intermediate uveitis (V-RNA 5 and

6), Birdshot chorioretinopathy (V-RNA 3) and in V-RNA 4 with unknown ocular diagnosis.

Lower EV-RNA concentrations were identified in patients with anterior uveitis, macular

degeneration, panuveitis, posterior uveitis, sarcoidosis uveitis, chronic pars planitis and iritis.

Fourteen of 23 specimens with either high concentrations of EVs (N/T 1-7) or high

concentrations of EV-RNA (V-RNA1-16) showed clinical vitreal cell infiltration (Table 1)

and 12 out of 23 patients demonstrated by fluorescein angiogram ‘leaky’ vessels (vasculitis).

Characterization of mRNA in vitreous EVs

We optimized qRT-PCR method for low levels of EV–RNA from vitreal fluid using qRT-

PCR analytics based upon the cell line A431 expressing high levels of VEGF. A431 cellular

RNA and EV-deriver RNA were used to create dilution curves for qRT-PCR experiments

using concentrations as low as 1pg per reaction and as high as 1000 pg recommended by the

manufacturer (Suppl. table 1). We were able to detect gene products of interest such as

GAPDH, CD63, VEGF as low as 10pq per qRT-PCR reaction. Based on these studies we

used 100pg of EV-RNA per reaction qRT-PCR to optimize the Taqman probes and primers

for our gene products. These probes were evaluated in positive and negative controls derived

from cell-line RNA and cell-line EV-RNA including, an astrocyte cell line (cell line RNA:

relatively positive for GAPDH, and negative for CD63, VEGF and EGFR),U87-WT glioma

cell line (cell line RNA; positive for GAPDH, CD63, EGFR) and A431 (cell line RNA

elatively positive for GAPDH, VEGF, CD63 EGFR) (Suppl. Table 2).

Next, this novel method was applied to study expression of gene product in vitreal human

samples. qRT-PCR studies were performed for identification of vitreal EV mRNA for

GAPDH, CD63, VEGF and EGFR. GAPDH is a metabolic enzyme, generally accepted a

“house-keeping gene product” which functions as a transcription factor and induces cell

apoptosis such as in Parkinson’s disease19. We demonstrated EV-mRNA GAPDH in 14

vitreous specimens. Surprisingly, two specimens (V6; V9) did not contain GAPDH mRNA

(Fig. 5A). The qRT-PCR Ct values for GAPDH mRNA were used to calculate the EV-mRNA

135

!

GAPDH copy number for each vitreal sample (Table 2). Copy numbers of EV-mRNA

GAPDH above 1000 were detected in specimens V1; V2; V12; V15; V16. Specimens V6 and

V9 did not contain GAPDH copies.

CD63, a transmembrane protein, associated with Hermansky-Pudlak syndrome20, plays a role

in tumorigenesis and is currently a marker of cellular EVs. Twelve specimens (Fig. 5B)

contained CD63 mRNA. EV-mRNA CD63 copies above 100 were identified (specimens V-

RNA1;2;7;8;15 and 16). Four specimens, including two negative for GAPDH, contained no

CD63 mRNA (V5; V6; V9: V11).

Next, we investigated “first in man” EV-derived mRNA VEGF and EGFR from vitrectomy

specimens of 16 patients and 6 patients, respectively. (Fig.5 D and C). The VEGF gene family

consists of 5 members, VEGF-A, VEGF-B, VEGF-C, VEGF-D and placental derived growth

factor (PlGF), of which VEGF-A, VEGF-B and PlGF are involved in angiogenesis and

VEGF-C and VEGF-D regulate lymphatic vasculogenesis21. Studies on angiogenesis have

extensively investigated the role of VEGF-A, a heparin-binding homodimeric glycoprotein

which is an endothelial cell mitogen and it has been shown that it correlates with tumor

progression such as in glioblastoma22. In this studie on vitreal EV-RNA we found that 9

specimens, including two without GAPDH or CD63 mRNA (V6; V9) contained VEGF

mRNA. The VEGF copy number above 20 were identified in specimens V-RNA

2;4;6;9;10;15;16. The remaining specimens contained either no VEGF mRNA or fewer

copies.

Four out of 6 vitreal specimens contained EGFR mRNA above 20 copies (V-RNA7;

14;15;16). EGFR, a small molecular growth factor is mainly produced by platelets and

macrophages and is present in urine, saliva, milk and plasma. It binds to its receptor EGFR,

whose discovery led to the Nobel Prize conveyance. The receptor is present in lung, anal and

glial brain tumors.

Discussion

Vitreal EV-derived Gene products

We have identified EVs in vitreal fluid from 23 patients with inflammatory, autoimmune and

malignant diseases. These EVs, presumably of platelet, lymphocytic, phagocytic, endothelial

or retinal origin, were identified by accepted methodologies including NTA analyses (n=7),

Transmission EM (n=7), and characterized in sixteen patients (EV-derived total RNA and

EV-mRNAs). We observed that intermediate uveitis and birdshot chorioretinopathy contained

higher levels of RNA compared to other type of uveitis or autoimmune diseases. In addition

136

!

we show here that as with cerebrospinal fluid, the vitreal fluid represents a unique source of

relatively acellular fluid, rich in EVs and EV-derived gene products such as GAPDH, CD63,

VEGF, EGFR.

The handling procedures for vitreous fluid have not previously been reported and in this study

we introduce a novel protocol (Figure 1), which was used to isolate EVs from as little as

200ul of vitreous fluid; a volume which provides between 25-90 EVs x 108/ml of fluid. This

calculation underscores the advantage associated with EV-based studies in comparison to

vitreal cell-based and vitreal fluid-based protein studies. To optimize this methodology we

performed dilution and optimization studies based upon evaluation of EVs obtained from

supernatants of tissue cultured cell lines. These provided our detection limits and represented

both ‘positive’ and ‘negative’ controls for EV-based qRT-PCR studies. In principle, our

preparative protocol can also be applied to small volumes of other body fluids. The size and

appearance and concentration of EVs varied from patient to patient and although they may be

disease dependent, our mixed population prevented distinctions. In addition there exists the

potential to perform vitreal EV-gene product sequencing studies for illnesses such as Behcets

and Birdshot retinopathy. We observed that NTA biased towards identification of smaller EVs

than the larger ones observed by TEM. Currently in the EV field there exists no commonly

accepted ‘gold standard’ for determination of EV size and concentration.

From sixteen vitreal fluid samples we isolated EVs and were able (Table 2) to extract a total

RNA. We demonstrate that vitreal fluid volumes as low as 200"l can be used for isolation,

characterization and quantification of EVs. Isolation and characterization of the extracted

mRNA from these EVs was analyzed by qRT-PCR for identification and quantification of 4

gene product mRNAs (CD63, GAPDH, VEGF, EGFR). The identification of these gene

products supports the concept of molecular diagnostics of vitreal diseases based upon

analyses of EV-derived mRNA . These diagnostics can be adapted to detect vitreal infections

of bacterial, viral and protozoan origin; autoimmune diseases; cancer; degenerative diseases;

coagulation disorders and vascular damage. We were able to detect as few as 15 copies of

VEGF in vitreal fluid; a figure which compares favorably to the literature detection of as few

as 5 copies of herpes mRNA in cells from the nervous system. qRT-PCR based diagnostic

biomarker studies utilizing vitreal fluids are analogous to studies based upon urine (prostate

cancer), sputum (reactive airway disease and cancer), and plasma (pregnancy outcome,

eclampsia, stroke and coronary event outcome). This study suggests, vitreo-retinal specialists

could obtain vitreal fluid, harvest for cell studies, the pellet after centrifugation (1500rpm) and

137

!

use the otherwise discarded supernatant for EV-based studies. This approach would

supplement the current diagnostic accuracy of cytopathologic, immunocytopathologic and cell

sorting approaches. This study suggests, qRT-PCR studies of EV-based gene products should

be more sensitive than similar studies performed on non-EV vitreal fluids as EV-based gene

products are not exposed to RNAases or PCR inhibitors likely present in vitreal fluids.

VEGF mRNA was identified from EVs in nine (of sixteen patients). VEGF-A gene encodes a

protein, which is the major mediator of vascular permeability and angiogenesis. Avastin

(Ranibizumab and bevacizumab), a VEGF-A inhibitor has been FDA approved for treatment

of glioblastoma, cancers of lung, rectum, breast and kidney. Both Ranibizumab and

bevacizumab are used off-label to treat ocular diseases including neovascularization

associated with proliferative diabetic retinopathy, glaucoma, diabetic macular edema,

retinopathy of prematurity and macular edema secondary to retinal vein occlusions6;23.

Symptoms and clinical examination may improve but unclear is the mechanism of action of

this drug, the kinetics of drug-target interactions, the efficacy of dose and schedule and the

selection of patients for therapy. Our identification of EV-based VEGF-A raises the potential

use of this analytic as a metric of drug response such as to Avastin. In addition these

approaches may offer a strategy by which to identify patients, with high levels of EV-VEGF-

A gene product or protein who may be likely responders to therapy

Conclusion

We have identified EVs in (diagnostic and therapeutic vitrectomy fluids of patients with

various vitreal and retinal diseases. We have developed techniques for the isolation and

characterization of these EVs and their mRNA content from small volumes of pauci-cellular

vitreal fluid. We have measured enrichment of EV-contained gene products: GAPDH, CD63,

VEGF, EGFR. These studies provide the basis for a diagnostic analysis of EVs and their

contents (DNA, RNA, proteins, miRNA) of viral, bacterial, protozoan infections, tumors, and

degenerative and vascular diseases of the retina or vitreous fluid. These analyses may also

serve as a response metric for VEGF inhibitor therapy. These studies offer the potential to be

expanded to analyses of fluids from the anterior chamber of the eye.

138

!

Reference:

1. Rothova A, Ooijman F, Kerkhoff F et al. Uveitis masquerade syndromes.

Ophthalmology 2001;108:386-399.

2. Gündüz K, Demirel S, Yagmurlu B et al. Correlation of surgical outcome with

neuroimaging findings in periocular lymphangiomas. Ophthalmology 2006;113:1231.

3. Karma A, Von Willebrand EO, Tommila PV. Primary intraocular lymphoma:

improving the diagnostic procedure. Ophthalmology 2007;114:1372-1377

4. Chan CC and Wallace DJ. Intraocular lymphoma: update on diagnosis and

management. Cancer Control 2004;11:285-295.

5. Wang JG, Williams JC, Davis BK et al. Monocytic microparticles activate endothelial

cells in an IL-1#-dependent manner. Blood 2011;118:2366-2374.

6. Van Geest RJ, Lesnik-Oberstein SY, Tan H et al. A shift in the balance of vascular

endothelial growth factor and connective tissue growth factor by bevacizumab causes

the angiofibrotic switch in proliferative diabetic retinopathy. Br J Ophthalmol

2012;96:587-590.

7. Skog J, Würdinger T, van Rijn S et al. Glioblastoma microvesicles transport RNA and

protein that promote tumor growth and provide diagnostic biomarkers. Nat Cell Biol

2008;10:1470-1476.

8. Atai NA, Balaj L, van Veen H et al. Heparin blocks transfer of extracellular vesicles

recipient cells between donor and cellular cells. J Neurooncol 2013;115:343-351.

9. György B, Szabó TG, Pásztói M et al. Membrane vesicles, current state-of-the-art:

emerging role of extracellular vesicles. Cell Mol Life Sci 2011;68:2667-2688.

139

!

10. Al-Nedawi K, Meehan B, Kerbel RS et al. Endothelial expression of autocrine VEGF

upon the uptake of tumor-derived microvesicles containing oncogenic EGFR. Proc

Natl Acad Sci USA 2009;106:3794-3799.

11. Simpson RJ, Lim JW, Moritz RL et al. Exosomes: proteomic insights and diagnostic

potential. Expert Rev Proteomics 2009;6:267-283.

12. van der Pol E, Böing AN, Harrison P et al. Classification, functions, and clinical

relevance of extracellular vesicles. Pharmacol Rev 2012;64:676-705.

13. Miranda KC, Bond DT, McKee M et al. Nucleic acids within urinary

exosomes/microvesicles are potential biomarkers for renal disease. Kidney Int 2010;

78:191-199.

14. Wahlgren J, De L Karlson T, Brisslert M et al. Plasma exosomes can deliver

exogenous short interfering RNA to monocytes and lymphocytes. Nucleic Acids Res

2012;40:e130

15. Harrington MG, Fonteh AN, Oborina E et al. The morphology and biochemistry of

nanostructures provide evidence for synthesis and signaling functions in human

cerebrospinal fluid. Cerebrospinal Fluid Res 2009;6:10.

16. Lässer C, Alikhani VS, Ekström K et al. Human saliva, plasma and breast milk

exosomes contain RNA: uptake by macrophages. J Transl Med 2011;9:9.

17. Franz C, Boing A, Hau C et al. Procoagulant tissue factor-exposing vesicles in human

seminal fluid. J Reproduc Immunol 2013;98:45-51.

18. Berckmans RJ, Sturk A, Van Tienen LM et al. Cell-derived vesicles exposing

coagulant tissue factor in saliva. Blood 2011;37:146-152.

19. Hara MR and Snyder SH. Nitric oxide-GAPDH-Siah: a novel cell death cascade. Cell

Mol Neurobiol 2006; 26:527-538.

140

!

20. Oh J, Ho L, Ala-Mello S et al. Mutation analysis of patients with Hermansky-Pudlak

syndrome: a frameshift hot spot in the HPS gene and apparent locus heterogeneity.

Am J Hum Genet 1998;62:593-598.

21. Witmer AN, Vrensen GFJM, Van Noorden CJF et al. Vascular endothelial growth

factors and angiogenesis in eye disease. Progr Ret Eye Res 2003;22:1-29.

22. Miletic H, Niclou SP, Johansson M et al. Anti-VEGF therapies for malignant glioma:

treatment effects and escape mechanisms. Expert Opin Ther Targets 2009;13:455-468.

141

!

Conflict of interest

Acknowledgement

This work was supported by a 2009 AMC Scholarship, University of Amsterdam, the

Netherlands (N.A.A.). This work was supported by NCI: 2PO1CA069246-16A, Experimental

Therapeutics and BioMonitoring of Brain Tumors (F.H.H.) and The Richard Floor

Biorepository Fund and Prof. dr. Xandra O. Breakefield (NIH/NCI P01 CA069246).

142

!

Figure 1. Novel protocol for purification of EVs from vitreous fluid.

Figure 2. A. The NanoSight analysis of EVs in vitreous fluid. B. Amount of EVs expressed

as number of particles per ml in each vitreous sample of patients N/T-1-N/T-7 as measured by

NanoSight.

143

!

Figure 3. TEM images of vitreous specimens N/T-1-N/T-7 with various diseases (see Table

1). Bar = 500 nm.

Figure 4. RNA profile of a vitreous specimen showing the total RNA and the quality of the

RNA conform ribosomal RNA (18s and 28s).

144

!

F

igure 5. Ct values obtained by qRT-PCR for GAPDH (A), CD63 (B) VEGF (C) and EGFR

WT (D) mRNA in vitreous specimens of patients V1-16

145

!

Table 1. Detailed patient information. Vitreous EVs of patients N/T-1-N/T-7 have been used

for nanoparticles-tracking analysis (NTA) and transmission electron microscopy (TEM)

whereas those of patients V1-16 were used for qRT-PCR. M, male; F, female; OD, right eye;

OS, left eye; CA, carcinoma.

Table 2. Total RNA (ng) in vitreous samples of patients V1-16.

146

!

147

!

Table 3. A. The absolute copy number of each RNA per reaction. B. The absolute copy

number of each mRNA in total RNA of each specimens of patients V1-16. C. Copy number

of each mRNA in 1 ng of total RNA.

148

!

Suppl. Table 2

149