Wouter Woud

Fantastic Vesicles and How to Find Them Direct Detection of Single Extracellular Vesicles in Complex, Clinically Relevant Biofluids Wouter Willem Woud

The research described in this thesis was performed at the Erasmus MC Transplant Institute, Department of Internal Medicine, University Medical Center, Rotterdam, The Netherlands. Cover design: BernArt Visuals Lay-out: Publiss | www.publiss.nl Print: Ridderprint | www.ridderprint.nl Copyright: Wouter W. Woud, 2023 Printing of this thesis was financially supported by the Nederlandse Transplantatie Vereniging (NTV), Erasmus Universiteit Rotterdam, Cytek Biosciences, and Cees Woud Natuursteen. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, by photocopying, recording, or otherwise, without the prior written permission of the author.

Fantastic Vesicles and How to Find Them Direct detection of single extracellular vesicles in complex, clinically relevant biofluids Fantastische vesikels en hoe ze te vinden Directe detectie van individuele extracellulaire vesikels in complexe, klinisch relevante biovloeistoffen Thesis To obtain the degree of Doctor from the Erasmus University Rotterdam, by command of the Rector Magnificus Prof. dr. A. L. Bredenoord and in accordance with the decision of the Doctorate Board. The public defence shall be held on Wednesday 28 June 2023 at 10:30 hrs by Wouter Willem Woud born in Zaanstad, The Netherlands

Doctoral Committee Promotor: Prof. dr. C. C. Baan Other members: Prof. dr. E. J. Hoorn Prof. dr. R. J. Porte Dr. E. N. M. Nolte-‘t Hoen Co-promotors: Dr. M. J. Hoogduijn Dr. ir. K. Boer

TABLE OF CONTENTS Fantastic Vesicles and How to Find Them Chapter 1 General Introduction and Objectives 9 Chapter 2 Nanoparticle Release by Extended Criteria Donor Kidneys during Normothermic Machine Perfusion – Transplantation, 2019 23 Part I – Development: The Needle in the Haystack Chapter 3 An Imaging Flow Cytometry-Based Methodology for the Analysis of Single Extracellular Vesicles in Unprocessed Human Plasma – Communications Biology, 2022 29 Chapter 4 Isolation-free Measurement of Single Urinary Extracellular Vesicles by Imaging Flow Cytometry – Nanomedicine: Nanotechnology, Biology, and Medicine, 2023 83 Part II – Validation: Detection of EVs in Kidney Transplantation Chapter 5 Extracellular Vesicles Released During Normothermic Machine Perfusion are Associated with Human Donor Kidney Characteristics – Transplantation, 2022 121 Chapter 6 Direct Detection of Circulating Donor-Derived Extracellular Vesicles in Kidney Transplant Recipients – Scientific Reports, 2022 149 What have we learned? Chapter 7 Summaries 175 - English Summary 176 - Nederlandse Samenvatting 179 Chapter 8 General Discussion and Future Perspectives 183 Chapter 9 About the Author 199 - Curriculum Vitae 200 - List op Publications 201 - PhD Portfolio 202 - Acknowledgements 204



Chapter 1 10 The wide wealthy world of communication The living world is a collection of ecosystems in which the interactions between individual components determine the emergent properties of complex biological systems 1. Such interactions are mainly driven through communication between or within species and organisms, which are subject to changing environmental conditions. As such, the state of a system is reflected by its occupying species, which may be studied to infer environmental changes. To illustrate: bees are considered an indicator species meaning that their vibrancy on earth reflects environmental conditions and aids in gauging the health of ecosystems. Simply put, reduced numbers of bees in a given hive indicate a decline in environmental conditions. Similar to studying bees to infer environmental status, studying biomolecules released by cells – as means of intercellular communication – allows researchers to infer cellular status, and, by extension, the status of cellular systems as a whole. This is of paramount importance in the context of health and disease. Broadly speaking, cellular communication is performed through either direct contact with neighboring cells or the excretion of biomolecules into the extracellular space. A relatively recently discovered and exciting modality of such an excreted means of communication, and the subject of this thesis, are extracellular vesicles (EVs). A brief history of EV-erything EVs were first described in a series of manuscripts which identified potential structures that would retrospectively be described as EVs 2. In 1946, Chargaff and West reported the discovery of a ‘particulate fraction’ which sedimented from human plasma at 31,000 g (but remained at solution at 5,000 g). At the time, these particles were suggested to be a form of cellular waste 3. In 1967, Peter Wolf described a “material in minute form, sedimentable by high-speed centrifugation and originating from platelets, but distinguishable from intact platelets” – which we now know as the EV fraction. Wolf provided electron microscopy images of these particles, which he described as ‘platelet dust’ 4. A few years later (1971), Neville Crawford published further images of these particles – which were now being described as ‘microparticles’ – and showed that these particles contained lipids and carried cargo such as the cellular energy source adenosine triphosphate (ATP) 5, thus suggesting that these particles were more than cellular debris or waste particles. These pioneering experiments were the first to describe the presence and structure of such cell-free components and hinted at their potential biological importance.

General Introduction and Objectives 11 1 Since then, EVs have been identified as a heterogeneous group of lipid bilayered membrane structures (30-8000 nm in diameter 6), and are classified into three major subtypes based on their mode of biogenesis (Figure 1). In addition to their mode of release, EV size is often used for characterization: exosomes are regarded as the smallest type of EVs (30 – 100 nm in diameter), microvesicles range from 100 to 1000 nm in diameter, and apoptotic bodies may reach sizes up to several micrometers 7. However, there is some controversy on nomenclature and sizes as different types of EVs overlap in their size distribution 8, 9. Therefore, the term ‘Extracellular Vesicle’, as used in this thesis, is used as a generic term for all secreted vesicles. EVs carry proteins on their surface and a variety of macromolecules as cargo (e.g., lipids, proteins, enzymes, nucleic acids, protein-coding mRNAs and regulatory microRNAs 10, 11), which are thought to reflect the status of their cell of origin. Upon contact with or internalization by recipient cells, EVs have the ability to transfer information from one cell to another, thus modulating recipient cell behavior 12. Therefore, EVs have been recognized as mediators of intercellular communication during both normal physiological as well as in pathological processes 10, 13. As EVs are excreted by virtually all cell types in the human body, they can be found in all body fluids, such as the blood 6, saliva 14 and urine 15, 16. Figure 1 - Exosomes, Microvesicles, and Apoptotic bodies. Exosomes (left) are released into the extracellular domain through fusion of multivesicular bodies (MVB) with the cell membrane. Microvesicles (middle) are formed through outward budding of the cell membrane, and apoptotic bodies (right) are fragments of cells which have undergone apoptosis. Adapted from Karpmann, et al. 17.

Chapter 1 12 The relative stability of EVs (their cargo is protected from fragmentation and degradation by the lipid bilayer 18), and their ubiquitous presence in (relatively) easily obtainable bodily fluids have sparked the interest in EVs as potential biomarkers for disease diagnosis and prognosis 19. As EVs exist at numbers exceeding 1000 particles for each cell of origin, their analysis offers quantitative advantages over less abundant entities such as circulating tumor cells 20, donor-derived cell-free DNA 21, or antibodies against cytoplasmic proteins 22. EVs as potential biomarkers in kidney transplantation Kidney transplantation is the preferred treatment for patients suffering from irreversible, end-stage renal disease - providing increased patient survival over dialysis 23. However, the shortage of available donor kidneys (grafts), the increasing number of patients on the waiting list, and the general aging of the population has led to an increased use of expanded-criteria donor (ECD) grafts as well as grafts procured from donation after circulatory death (DCD) 24 – both of which are associated with poorer transplant outcomes when compared to organs from standard criteria donors 25, 26. An essential problem with the usage of these kidneys is the lack of quality measures needed to guide the clinician in deciding whether to accept or decline the organ. In the past decade, hypothermic machine perfusion (HMP) has gained interest as a promising preservation technique for deceased donor organs 27, showing improved clinical outcomes after kidney transplantation compared with static cold storage 28. The most recent development in organ preservation is normothermic machine perfusion (NMP). In contrast to HMP, NMP aims to restore cellular metabolism and function to the organ, which is achieved through circulation of a warm, oxygenated red blood cell based solution through the organ prior to transplantation 29, 30. Because metabolism is activated, NMP offers the possibility to assess graft status prior to transplantation through monitoring of the perfusion dynamics and analysis of biomarkers (such as EVs) in the perfusion fluids 25, 29, 31, 32. After transplantation - despite potent immunosuppressive therapy - acute rejection of the graft occurs in as much as 21% of transplantations during the first 6 months after transplantation 21. Though the function of kidney allografts are routinely monitored through serum creatinine and urea, and urinary protein concentrations, these markers are relatively insensitive for allograft rejection as a rise in their concentrations does not specifically indicate immunologic rejection

General Introduction and Objectives 13 1 33. Consequently, an elevation of these markers is often followed by a kidney transplant biopsy, which, despite being the gold standard to diagnose rejection, is an invasive procedure with a risk of complications including bleeding and infection 34. Combined, these issues reveal the critical need for more accurate, early and minimally invasive biomarker platforms to diagnose kidney allograft rejection. The potential of EVs as biomarker for the detection of allograft rejection has been described – in animal models – by a few groups 35-40. These studies have shown that donor-derived EVs are released into the circulation post transplantation, and provide indications that concentrations of donor-derived EVs diminish during rejection well before alterations in classical biomarkers or histologic manifistation of injury can be observed. These findings suggest that detection and monitoring of donor-derived circulating EVs may herald rejection in a more time-sensitive manner compared to classical markers. Challenging to measure Despite the interest in and clinical relevance of EVs as biomarker, EV analysis is hampered by a variety of factors. First of all, their physical characteristics, such as their small size, low epitope copy number 41, the variety of protein markers depending on the cell source, the confinement of some markers to the luminal side of the EV, and the low abundance of pathological EVs 11, 42 all contribute to the complexity of EV analysis. Additionally, no unique antigens representative for specific EV classes and subpopulations have been reported to date. Instead, tetraspanins (CD9/CD63/CD81) are recognized as common EV antigens. These proteins are enriched on EVs and are involved in EV biogenesis, cargo selection, and cell targeting 43, 44. Second, the identification of EVs in blood plasma is further hindered by the molecular complexity of plasma, which contains multiple elements (e.g. protein aggregates, cell debris and the far more abundant lipoproteins) that interfere with EV analysis 11, 45. Lipoproteins are submicron structures of lipids and apolipoproteins that are excreted into the circulation by the liver and intestines. They are classified into several subgroups, and their biophysical properties in terms of size and density largely overlap with those of EVs. However, a distinguishing feature is the presence of an aqueous core in EVs, whereas the core of lipoproteins is comprised of lipids 11.

Chapter 1 14 Third, apart from their biological diversity and their large overlap in biophysical properties with other entities, a lack of robust EV detection methods and ambiguities in how data should be interpreted for EV analysis makes interpretation between studies challenging 46, 47. Currently, the gold standard approach for EV analysis is based on the isolation or concentration of EVs. Ultracentrifugation, density-gradient, and size exclusion chromatography are the most widely used EV isolation techniques 48, despite yielding low-purity EV samples due to the coisolation of non-desired molecules such as lipoproteins 11, 45. Additionally, a variety of analytical platforms are available. Nanoparticle tracking analysis (NTA) allows the determination of the size distribution and a rough indication of the concentration of individual nanoparticles in suspension49, but provides limited phenotyping capabilities. In turn, transmission electron microscopy (TEM) is able to image particles <1 nm, but is time consuming and not suitable for looking at shifts in EV populations. Other methods, such as ELISA and Western blot analysis, offer bulk phenotyping abilities but lack quantification 9, 15, 50, 51. Thus, a tool for the accurate determination of the concentration and phenotyping of single EVs in complex samples such as plasma represents an unmet need. The holy grail: direct detection of single EVs in complex samples The only technique that has the potential to detect, size, and phenotype thousands to millions of EVs per minute is flow cytometry (FC) 52. However, most clinical flow cytometers, and their corresponding assays, are designed for cell measurements and are not readily adapted to measure EVs; as the majority of EVs are <300 nm in diameter, conventional FCs struggle to discriminate these particles from background signals 11, 53, 54. Another problem with flow cytometry is that the generated signals are expressed in arbitrary units, which hinders comparison of results between different instruments 47. To address these issues, more sensitive instruments are introduced into the field, and guidelines regarding methods and data reporting are being developed for both flow cytometry (Minimum Information about a Flow Cytometry experiment, MIFlowCyt) and EV research 47. In summary, the ideal EV analysis platform would be able to 1) detect and discriminate single EVs <300 nm in diameter above background signals of the instrument, 2) determine the size, concentration and phenotype of single EVs, 3)

General Introduction and Objectives 15 1 operate at a relatively high-throughput rate of thousands to millions of EVs per minute, 4) be used without the need for prior EV isolation (thus omitting sample selection biases whilst simultaneously reducing sample handling time), and 5) discriminate the identified EVs from other contaminating components in the biofluid of interest. Objectives of this thesis Sensitive and standardized methods for single EV analysis are needed if EVs are to be translated into clinical practice. In recent years, imaging flow cytometry (IFCM) has emerged as a technique that enables the discrimination and analysis of single EVs with increased sensitivity compared to conventional FC. The ability of IFCM to detect submicron particles has been demonstrated using fluorescent polystyrene beads 55-58 or the use of cell supernatant-derived EVs 50. At the moment of writing, several studies have reported the detection of EVs - obtained after performing isolation procedures - from plasma using IFCM 55, 56, 58, 59. However, the used isolation procedures may have changed some EV properties: ultrafiltration might disintegrate larger EVs (thus generating smaller particles which, in turn, skew EV quantification upward) 60 whereas ultracentrifugation might cause aggregation and encapsulation of EVs (skewing EV quantification downward) 61. Thus, it is unlikely that these results represent all EVs in plasma 62. The main objective of this thesis is to explore whether IFCM is a suitable platform for the direct detection of single EVs in molecular complex samples such as plasma without the need to perform EV isolation techniques. This thesis is composed of two parts: ⦁ Part I presents the development of a standardized IFCM-based methodology which allows for the direct detection, characterization and quantification of single EVs in molecular complex samples such as perfusate, plasma or urine. ⦁ Part II aims to validate the standardized methodology to detect EV subsets in the context of kidney transplantation. In chapter 2, we aim to validate the proof-of-concept that kidneys release nanoparticles (such as protein aggregates and EVs) ex-vivo. In this chapter, we examine the release of nanoparticles into the perfusion fluid by expanded-criteria donor (ECD) kidneys during normothermic machine perfusion (NMP). To this end,

Chapter 1 16 perfusate samples taken before, during, and after the NMP procedure are analyzed with nanoparticle tracking analysis (NTA) to quantitate and determine the size distribution of nanoparticles released during NMP. Though NTA currently is a gold-standard technique for EV-quantitation and size analysis, it is unsuitable for complex samples such as plasma or urine (due to its limited phenotyping capabilities). In chapter 3, we aim to provide a standardized (size- and fluorescence calibrated) IFCM-based methodology which is able to discriminate, phenotype, and determine the concentration of individual human plasma-derived EVs ≤400 nm in diameter – without prior isolation of EVs. This methodology aims to discriminate EVs from contaminating agents such as lipoproteins and protein aggregates in molecular complex samples such as plasma, and forms the backbone of this thesis. In chapter 4, we present an adaptation of this methodology aiming to detect single EVs in urine – another complex bio fluid with its own set of challenges in the context of EV detection. In chapter 5, we characterize the nanoparticles released by ECD kidneys during NMP, and confirm that these are representative of EVs. Following the identification of EVs in the perfusion fluids, we aim to identify distinct EV subsets and examine whether these are potentially correlated with donor and NMP viability characteristics. As a first step towards clinical applicability, we next set out to determine whether the developed methodology is able to detect and follow-up single, (donor) tissuederived EVs in plasma samples of kidney transplant recipients (chapter 6). Chapter 7 provides a summary of the results described in this thesis. Chapter 8 discusses these results and provides a perspective on future implications of our findings.

General Introduction and Objectives 17 1 REFERENCES 1. Friedman J, Gore J. Ecological systems biology: The dynamics of interacting populations. Current Opinion in Systems Biology. 2017;1:114-21. 2. Couch Y, Buzas EI, Di Vizio D, Gho YS, Harrison P, Hill AF, et al. A brief history of nearly EVerything - The rise and rise of extracellular vesicles. J Extracell Vesicles. 2021;10(14):e12144. 3. Chargaff EW, R. The biological significance of the thromboplastic protein of blood. Journal of Biological Chemistry. 1946;166(1):189-97. 4. Wolf P. The Nature and Significance of Platelet Products in Human Plasma. Brit J Haemat. 1967;13(3):269-88. 5. Crawford N. The Presence of Contractile Proteins in Platelet Microparticles Isolated from Human. Brit J Haemat. 1971;21(53):53-69. 6. Arraud N, Linares R, Tan S, Gounou C, Pasquet JM, Mornet S, et al. Extracellular vesicles from blood plasma: determination of their morphology, size, phenotype and concentration. J Thromb Haemost. 2014;12(5):614-27. 7. Gyorgy B, Szabo TG, Pasztoi M, Pal Z, Misjak P, Aradi B, et al. Membrane vesicles, current state-of-the-art: emerging role of extracellular vesicles. Cell Mol Life Sci. 2011;68(16):266788. 8. Gould SJ, Raposo G. As we wait: coping with an imperfect nomenclature for extracellular vesicles. J Extracell Vesicles. 2013;2. 9. Witwer KW, Buzas EI, Bemis LT, Bora A, Lasser C, Lotvall J, et al. Standardization of sample collection, isolation and analysis methods in extracellular vesicle research. J Extracell Vesicles. 2013;2. 10. Pitt JM, Kroemer G, Zitvogel L. Extracellular vesicles: masters of intercellular communication and potential clinical interventions. J Clin Invest. 2016;126(4):1139-43. 11. Simonsen JB. What Are We Looking At? Extracellular Vesicles, Lipoproteins, or Both? Circ Res. 2017;121(8):920-2. 12. Kalra H, Drummen GP, Mathivanan S. Focus on Extracellular Vesicles: Introducing the Next Small Big Thing. Int J Mol Sci. 2016;17(2):170. 13. Abels ER, Breakefield XO. Introduction to Extracellular Vesicles: Biogenesis, RNA Cargo Selection, Content, Release, and Uptake. Cell Mol Neurobiol. 2016;36(3):301-12. 14. Han Y, Jia L, Zheng Y, Li W. Salivary Exosomes: Emerging Roles in Systemic Disease. Int J Biol Sci. 2018;14(6):633-43. 15. Salih M, Zietse R, Hoorn EJ. Urinary extracellular vesicles and the kidney: biomarkers and beyond. Am J Physiol Renal Physiol. 2014;306(11):F1251-9. 16. Wang S, Kojima K, Mobley JA, West AB. Proteomic analysis of urinary extracellular vesicles reveal biomarkers for neurologic disease. EBioMedicine. 2019;45:351-61. 17. Karpman D, Stahl AL, Arvidsson I. Extracellular vesicles in renal disease. Nat Rev Nephrol. 2017;13(9):545-62. 18. Minciacchi VR, Zijlstra A, Rubin MA, Di Vizio D. Extracellular vesicles for liquid biopsy in prostate cancer: where are we and where are we headed? Prostate Cancer Prostatic Dis. 2017;20(3):251-8. 19. Roy S, Hochberg FH, Jones PS. Extracellular vesicles: the growth as diagnostics and therapeutics; a survey. J Extracell Vesicles. 2018;7(1):1438720.

Chapter 1 18 20. Krebs MG, Hou JM, Ward TH, Blackhall FH, Dive C. Circulating tumour cells: their utility in cancer management and predicting outcomes. Ther Adv Med Oncol. 2010;2(6):351-65. 21. Verhoeven J, Boer K, Peeters AMA, Clahsen-van Groningen MC, Roodnat JI, van de Wetering J, et al. A Novel High-Throughput Droplet Digital PCR-Based Indel Quantification Method for the Detection of Circulating Donor-derived Cell-free DNA After Kidney Transplantation. Transplantation. 2022. 22. Kierny MR, Cunningham TD, Kay BK. Detection of biomarkers using recombinant antibodies coupled to nanostructured platforms. Nano Rev. 2012;3. 23. Harhay MN, Rao MK, Woodside KJ, Johansen KL, Lentine KL, Tullius SG, et al. An overview of frailty in kidney transplantation: measurement, management and future considerations. Nephrol Dial Transplant. 2020;35(7):1099-112. 24. Summers DM, Watson CJ, Pettigrew GJ, Johnson RJ, Collett D, Neuberger JM, et al. Kidney donation after circulatory death (DCD): state of the art. Kidney Int. 2015;88(2):2419. 25. Brat A, Pol RA, Leuvenink HG. Novel preservation methods to increase the quality of older kidneys. Curr Opin Organ Transplant. 2015;20(4):438-43. 26. Callaghan CJ, Harper SJ, Saeb-Parsy K, Hudson A, Gibbs P, Watson CJ, et al. The discard of deceased donor kidneys in the UK. Clin Transplant. 2014;28(3):345-53. 27. Rijkse E, de Jonge J, Kimenai HJAN, Hoogduijn MJ, de Bruin RWF, van den Hoogen MWF, et al. Safety and feasibility of 2 h of normothermic machine perfusion of donor kidneys in the Eurotransplant Senior Program. BJS Open. 2021;5(1). 28. Moers C, Smits JM, Maathuis MH, Treckmann J, van Gelder F, Napieralski BP, et al. Machine perfusion or cold storage in deceased-donor kidney transplantation. N Engl J Med. 2009;360(1):7-19. 29. Hosgood SA, Saeb-Parsy K, Wilson C, Callaghan C, Collett D, Nicholson ML. Protocol of a randomised controlled, open-label trial of ex vivo normothermic perfusion versus static cold storage in donation after circulatory death renal transplantation. BMJ Open. 2017;7(1):e012237. 30. Vallant N, Wolfhagen N, Sandhu B, Hamaoui K, Cook T, Pusey C, et al. A Comparison of Pulsatile Hypothermic and Normothermic Ex Vivo Machine Perfusion in a Porcine Kidney Model. Transplantation. 2021;105(8):1760-70. 31. Moers C, Varnav OC, van Heurn E, Jochmans I, Kirste GR, Rahmel A, et al. The value of machine perfusion perfusate biomarkers for predicting kidney transplant outcome. Transplantation. 2010;90(9):966-73. 32. Xu J, Buchwald JE, Martins PN. Review of Current Machine Perfusion Therapeutics for Organ Preservation. Transplantation. 2020;104(9):1792-803. 33. Josephson MA. Monitoring and managing graft health in the kidney transplant recipient. Clin J Am Soc Nephrol. 2011;6(7):1774-80. 34. Morgan TA, Chandran S, Burger IM, Zhang CA, Goldstein RB. Complications of UltrasoundGuided Renal Transplant Biopsies. Am J Transplant. 2016;16(4):1298-305. 35. Campana S, De Pasquale C, Carrega P, Ferlazzo G, Bonaccorsi I. Cross-dressing: an alternative mechanism for antigen presentation. Immunol Lett. 2015;168(2):349-54. 36. Golebiewska JE, Wardowska A, Pietrowska M, Wojakowska A, Debska-Slizien A. Small Extracellular Vesicles in Transplant Rejection. Cells. 2021;10(11).

General Introduction and Objectives 19 1 37. Gunasekaran M, Xu Z, Nayak DK, Sharma M, Hachem R, Walia R, et al. Donor-Derived Exosomes With Lung Self-Antigens in Human Lung Allograft Rejection. Am J Transplant. 2017;17(2):474-84. 38. Kennel PJ, Saha A, Maldonado DA, Givens R, Brunjes DL, Castillero E, et al. Serum exosomal protein profiling for the non-invasive detection of cardiac allograft rejection. J Heart Lung Transplant. 2018;37(3):409-17. 39. Tower CM, Reyes M, Nelson K, Leca N, Kieran N, Muczynski K, et al. Plasma C4d+ Endothelial Microvesicles Increase in Acute Antibody-Mediated Rejection. Transplantation. 2017;101(9):2235-43. 40. Zhang H, Huang E, Kahwaji J, Nast CC, Li P, Mirocha J, et al. Plasma Exosomes From HLA-Sensitized Kidney Transplant Recipients Contain mRNA Transcripts Which Predict Development of Antibody-Mediated Rejection. Transplantation. 2017;101(10):2419-28. 41. Welsh JA, van der Pol E, Bettin BA, Carter DRF, Hendrix A, Lenassi M, et al. Towards defining reference materials for measuring extracellular vesicle refractive index, epitope abundance, size and concentration. J Extracell Vesicles. 2020;9(1):1816641. 42. Blijdorp CJ, Tutakhel OAZ, Hartjes TA, van den Bosch TPP, van Heugten MH, Rigalli JP, et al. Comparing Approaches to Normalize, Quantify, and Characterize Urinary Extracellular Vesicles. J Am Soc Nephrol. 2021. 43. Jankovicova J, Secova P, Michalkova K, Antalikova J. Tetraspanins, More than Markers of Extracellular Vesicles in Reproduction. Int J Mol Sci. 2020;21(20). 44. Raposo G, Stoorvogel W. Extracellular vesicles: exosomes, microvesicles, and friends. J Cell Biol. 2013;200(4):373-83. 45. Karimi N, Cvjetkovic A, Jang SC, Crescitelli R, Hosseinpour Feizi MA, Nieuwland R, et al. Detailed analysis of the plasma extracellular vesicle proteome after separation from lipoproteins. Cell Mol Life Sci. 2018;75(15):2873-86. 46. Konoshenko MY, Lekchnov EA, Vlassov AV, Laktionov PP. Isolation of Extracellular Vesicles: General Methodologies and Latest Trends. Biomed Res Int. 2018;2018:8545347. 47. Welsh JA, Van Der Pol E, Arkesteijn GJA, Bremer M, Brisson A, Coumans F, et al. MIFlowCytEV: a framework for standardized reporting of extracellular vesicle flow cytometry experiments. J Extracell Vesicles. 2020;9(1):1713526. 48. Gardiner C, Vizio DD, Sahoo S, Théry C, Witwer KW, Wauben M, et al. Techniques used for the isolation and characterization of extracellular vesicles: results of a worldwide survey. Journal of Extracellular Vesicles. 2016;5:10.3402/jev.v5.32945. 49. van der Pol E, Coumans FA, Grootemaat AE, Gardiner C, Sargent IL, Harrison P, et al. Particle size distribution of exosomes and microvesicles determined by transmission electron microscopy, flow cytometry, nanoparticle tracking analysis, and resistive pulse sensing. J Thromb Haemost. 2014;12(7):1182-92. 50. Gorgens A, Bremer M, Ferrer-Tur R, Murke F, Tertel T, Horn PA, et al. Optimisation of imaging flow cytometry for the analysis of single extracellular vesicles by using fluorescence-tagged vesicles as biological reference material. J Extracell Vesicles. 2019;8(1):1587567. 51. Lener T, Gimona M, Aigner L, Borger V, Buzas E, Camussi G, et al. Applying extracellular vesicles based therapeutics in clinical trials - an ISEV position paper. J Extracell Vesicles. 2015;4:30087.

Chapter 1 20 52. Kuiper M, van de Nes A, Nieuwland R, Varga Z, van der Pol E. Reliable measurements of extracellular vesicles by clinical flow cytometry. Am J Reprod Immunol. 2021;85(2):e13350. 53. Larson MC, Luthi MR, Hogg N, Hillery CA. Calcium-phosphate microprecipitates mimic microparticles when examined with flow cytometry. Cytometry A. 2013;83(2):242-50. 54. Gyorgy B, Modos K, Pallinger E, Paloczi K, Pasztoi M, Misjak P, et al. Detection and isolation of cell-derived microparticles are compromised by protein complexes resulting from shared biophysical parameters. Blood. 2011;117(4):e39-48. 55. Erdbrugger U, Rudy CK, Etter ME, Dryden KA, Yeager M, Klibanov AL, et al. Imaging flow cytometry elucidates limitations of microparticle analysis by conventional flow cytometry. Cytometry A. 2014;85(9):756-70. 56. Headland SE, Jones HR, D’Sa AS, Perretti M, Norling LV. Cutting-edge analysis of extracellular microparticles using ImageStream(X) imaging flow cytometry. Sci Rep. 2014;4:5237. 57. Lannigan J, Erdbruegger U. Imaging flow cytometry for the characterization of extracellular vesicles. Methods. 2017;112:55-67. 58. Mastoridis S, Bertolino GM, Whitehouse G, Dazzi F, Sanchez-Fueyo A, Martinez-Llordella M. Multiparametric Analysis of Circulating Exosomes and Other Small Extracellular Vesicles by Advanced Imaging Flow Cytometry. Front Immunol. 2018;9:1583. 59. Ricklefs FL, Maire CL, Reimer R, Duhrsen L, Kolbe K, Holz M, et al. Imaging flow cytometry facilitates multiparametric characterization of extracellular vesicles in malignant brain tumours. J Extracell Vesicles. 2019;8(1):1588555. 60. Jang SC, Kim OY, Yoon CM, Choi DS, Roh TY, Park J, et al. Bioinspired exosome-mimetic nanovesicles for targeted delivery of chemotherapeutics to malignant tumors. ACS Nano. 2013;7(9):7698-710. 61. Linares R, Tan S, Gounou C, Arraud N, Brisson AR. High-speed centrifugation induces aggregation of extracellular vesicles. J Extracell Vesicles. 2015;4:29509. 62. Shao H, Im H, Castro CM, Breakefield X, Weissleder R, Lee H. New Technologies for Analysis of Extracellular Vesicles. Chem Rev. 2018;118(4):1917-50.

General Introduction and Objectives 21 1

NANOPARTICLE RELEASE BY EXTENDED CRITERIA DONOR KIDNEYS DURING NORMOTHERMIC MACHINE PERFUSION Wouter W. Woud1, Ana Merino1, Martin J. Hoogduijn1, Karin Boer1, Martijn W.F. van den Hoogen1, Carla C. Baan1, Robert C. Minnee2 1 Department of Internal Medicine, section Nephrology and Transplantation, 2 Department of Surgery, Division of Hepato-Pancreato-Biliary and Transplant Surgery, Erasmus MC, University Medical Centre Rotterdam, The Netherlands Transplantation 103(5):p e110-e111, May 2019.

Chapter 2 24 The poor outcomes of transplantations with kidneys from extended criteria donors (ECD) requires new methods of organ preservation and assessment given the more severe ischemia/reperfusion injuries (IRI) compared to standard criteria donors1. Machine perfusion (MP), aimed at reducing IRI and increasing graft function, is extensively being researched and allows for the examination of the isolated kidneys ex vivo through analysis of perfusion fluids1,2. Donor-derived Extracellular Vesicles (EVs), which may reflect the conditional state of their tissue of origin, are known to be excreted in vivo in blood/urine and as such have been used to asses organ function post transplantation3. We postulate that analysis of nanoparticles, including EVs, in perfusion fluid during normothermic MP may allow for the assessment of kidney quality prior to transplantation. In this pilot trial, three ECD kidneys, (2 donors after cardiac death (DCD), 1 donor after brain death (DBD), comparable warm ischemia times of 15 minutes followed by 12 hours of cold ischemia, age 66/73/65, all male) were perfused at 37 oC for 2 hours during which perfusate samples were taken at 30 minutes intervals. Samples were centrifuged at 16.000x g for 10 minutes to discard platelets and supernatant was diluted 10x in 0.22 µm filtered PBS prior to analysis by Nanoparticle Tracking Analysis (NTA) (Figure 1A) to determine nanoparticle size and concentration (Figure 1B). Samples were measured by the Malvern Panalytical NanoSight NS300 and analyzed with NTA software version 3.2.16. In brief, NTA tracks the Brownian motion of individual nanoparticles in suspension on a frame-by-frame basis and correlates this movement with particle size through the Stokes-Einstein equation. Per sample, 10 videos of 15 seconds with 20-60 particles in the field of focus were recorded with camera level 11 and analyzed with detection threshold 5. This threshold was found to eliminate most of the protein background in our analysis and allowed us to focus on more complex particles such as EVs. In the perfusate samples the average particle size remained unchanged (~155 ± 7.6 nm, data not shown), while an ~7.75-fold increase in cumulative nanoparticle concentration was observed over time: 9.03E9 particles/mL after 120 minutes compared to 1.17E9 particles/mL after 0 minutes of perfusion (Figure 1C). Particle excretion was observed to be highest from the DBD kidney during the entire normothermic MP procedure. Whether this increased nanoparticle release reflects better kidney function requires further research; the released nanoparticles contain kidney-derived EVs which may be indicative for renal quality. These preliminary results indicate that analysis of perfusion fluid may be utilized to assess renal quality prior to transplantation.

Nanoparticle Release by Extended Criteria Donor Kidneys during Normothermic Machine Perfusion 2 25 Figure 1 - Renal nanoparticle release measured by NTA. A. Image of kidney derived nanoparticles during Nanoparticle Tracking Analysis (NTA) measurements. B. Size distribution vs. particle concentration of 1. perfusate after 0 minutes of perfusion and 2. perfusate after 120 minutes of perfusion. C. Ex vivo nanoparticle release by extended criteria donors (ECD) kidneys during Normothermic Machine Perfusion. Perfusate was obtained at 30 minutes intervals and measured with NTA. Accumulation of nanoparticles within the cumulative perfusion fluid was observed over time for all kidneys perfused, with highest excretion rate observed in the DBD kidney. REFERENCES 1. Brat A, Pol RA, Leuvenink HG. Novel preservation methods to increase the quality of older kidneys. Curr Opin Organ Transplant. 2015;20(4):438-443. 2. Hosgood SA, Saeb-Parsy K, Wilson C, Callaghan C, Collett D, Nicholson ML. Protocol of a randomised controlled, open-label trial of ex vivo normothermic perfusion versus static cold storage in donation after circulatory death renal transplantation. BMJ Open. 2017;7(1):e012237. 3. Morelli AE. Exosomes: From Cell Debris to Potential Biomarkers in Transplantation. Transplantation. 2017;101(10):2275-2276.

PART I Development: The Needle in the Haystack

AN IMAGING FLOW CYTOMETRYBASED METHODOLOGY FOR THE ANALYSIS OF SINGLE EXTRACELLULAR VESICLES IN UNPROCESSED HUMAN PLASMA Wouter W. Woud1, Edwin van der Pol2,3, Erik Mul4, Martin J. Hoogduijn1, Carla C. Baan1, Karin Boer1, Ana Merino1 1 Erasmus MC Transplant Institute, Department of Internal Medicine, University Medical Center Rotterdam, Rotterdam, The Netherlands. 2 Biomedical Engineering & Physics, Laboratory Experimental Clinical Chemistry, Vesicle Observation Center, Amsterdam UMC, University of Amsterdam, Amsterdam, The Netherlands. 3 Cancer Center Amsterdam, Imaging and Biomarkers, Amsterdam, The Netherlands. 4 Department Central Cell Analysis Facility, Sanquin Research and Landsteiner Laboratory, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands Commun Biol 5, 633 (2022)

Chapter 3 30 ABSTRACT Extracellular vesicles (EVs) are tissue-specific particles released by cells containing valuable diagnostic information in the form of various biomolecules. To rule out selection bias or introduction of artefacts caused by EV isolation techniques, we present a clinically feasible, imaging flow cytometry (IFCM)–based methodology to phenotype and determine the concentration of EVs with a diameter ≤400 nm in human platelet-poor plasma (PPP) without prior isolation of EVs. Instrument calibration (both size and fluorescence) were performed with commercial polystyrene beads. Detergent treatment of EVs was performed to discriminate true vesicular events from artefacts. Using a combination of markers (CFSE & Tetraspanins, or CD9 & CD31) we found that >90% of double-positive fluorescent events represented single EVs. Through this work, we provide a framework that will allow the application of IFCM for EV analysis in peripheral blood plasma in a plethora of experimental and potentially diagnostic settings. Additionally, this direct approach for EV analysis will enable researchers to explore corners of EV as cellular messengers in healthy and pathological conditions. KEYWORDS Unprocessed Human Plasma; Extracellular Vesicles; Imaging Flow Cytometry; Quantify; Phenotype; Diagnostic Platform

An Imaging Flow Cytometry-Based Methodology for the Analysis of Single Extracellular Vesicles 3 31 INTRODUCTION Extracellular vesicles (EVs) are lipid bilayer membrane structures (30-8000 nm in diameter 1) released by cells. They are involved in cellular communication through transfer of surface receptors and/or a variety of macromolecules carried as cargo (e.g., lipids, proteins, nucleic acids, protein-coding mRNAs and regulatory microRNAs) 2,3. As EVs are excreted by virtually all cell types in the human body, they can be found in most body fluids, such as the blood 1, saliva 4 and urine 5,6. Often regarded as a “snapshot” of the status of the cell of origin, EVs are examined for their biochemical signatures to assess the presence of various diseases, e.g., cancer or viral infections 7,8, and are considered excellent minimally invasive biomarkers in so-called liquid biopsies 9-11. While no unique antigens representative for specific EV classes and subpopulations have been reported to date, tetraspanins (CD9/ CD63/CD81) are recognized as common antigens. These proteins are enriched on EVs and are involved in EV biogenesis, cargo selection, and cell targeting 12,13. Despite the increased interest in EVs as biomarker, their quantification and characterization is hampered by physical characteristics such as their small size and low epitope copy number 14, the variety of their protein markers depending on the cell source, and the confinement of some markers to the luminal side of the EVs 3,15. The identification of EVs in blood plasma is further hindered by the molecular complexity of the plasma, which contains multiple elements (e.g., lipoproteins, cell debris and soluble proteins), that interfere with EV analysis 3,16. Moreover, a lack of robust methods and ambiguities in how data should be interpreted for EV analysis makes data interpretation between studies challenging 17,18. Currently, the gold standard approach for EV analysis is based on the isolation or concentration of EVs. Ultracentrifugation, density-gradient, and size exclusion chromatography are the most widely used EV isolation techniques 19, despite yielding low-purity EV samples due to the co-isolation of non-desired molecules such as lipoproteins 3,16. Additionally, a variety of analytical platforms are available. Nanoparticle tracking analysis (NTA) allows the determination of the size distribution and a rough indication of the concentration 20 of individual nanoparticles in suspension, but provides limited phenotyping capabilities. In turn, transmission electron microscopy (TEM) is able to image particles <1 nm, but is time consuming. Other methods, such as ELISA and Western blot analysis, offer bulk phenotyping abilities but lack quantification 5,21-23. Thus, a tool for the accurate determination of the concentration and phenotyping of single EV in complex samples such as plasma represents an unmet need.

Chapter 3 32 Flow Cytometry (FC) is a tool to quantify and phenotype particles in suspension. However, while EVs can reach sizes up to ~8000 nm in diameter, the majority of EV are <300 nm and are therefore difficult to discriminate from background noise by conventional FC 3,24,25. In recent years, imaging flow cytometry (IFCM) has emerged as a technique that enables the discrimination and analysis of single EV. The ability of IFCM to detect submicron particles has been demonstrated by several research groups using fluorescent polystyrene beads 26-29 or the use of cell supernatant-derived EV 21. To date, several studies have reported the detection of EVs - obtained after performing isolation procedures - from plasma using IFCM 26,27,29,30. However, due to the used isolation procedures, it is difficult to evaluate whether these results represent all EVs in plasma, or if some subpopulations are missed 31. To rule out selection bias or introduction of artefacts caused by EV isolation techniques, we here demonstrate an IFCM-based methodology to phenotype and determine the concentration of human plasma-derived EVs with a diameter ≤400 nm - without prior isolation of EVs. By omitting the need for sample isolation, this method is able to directly show the status of an individual, which will be greatly beneficial in the monitoring of EVs in health and disease, and will enable researchers to explore new corners of EV biology. RESULTS Outline of the article The objective of this article is to provide an assay that will allow researchers to study single EVs directly in diluted, labeled human plasma using IFCM. The following procedures were conducted to validate our assay: size calibration of the IFCM based on scatter intensities, background analysis of the IFCM, detergent treatment of EVs, dilution experiments, and fluorescence calibration. In addition, two labeling strategies based on CFSE+Tetraspanin+ and CD9+CD31+ were evaluated by mixing human plasma with mouse plasma at different ratios. Detection of sub-micron fluorescent polystyrene beads EV analysis at the single EV level requires an instrument that is able to detect a heterogeneous sub-micron sized population. To this end, we tested the ability of

An Imaging Flow Cytometry-Based Methodology for the Analysis of Single Extracellular Vesicles 3 33 IFCM to discriminate single-size populations of fluorescent sub-micron beads by measuring two commercially available mixtures of FITC-fluorescent polystyrene (PS) beads of known sizes (Megamix-Plus FSC – 900, 500, 300 and 100 nm, and Megamix-Plus SSC – 500, 240, 200, 160 nm). Within the Megamix-Plus FSC mix, we acquired a 300/500 nm bead ratio of 2.2, which is within the manufacturers internal reference qualification range (1.7 – 2.7 ratio). Next, we mixed both bead sets in a 1:1 ratio (‘Gigamix’) and performed acquisition. Figure 1a shows that IFCM is able to discern all seven fluorescent bead populations, as well as the 1 µm-sized Speed Beads (SB), via the FITC (Ch02) and side scatter (SSC - Ch06) intensities. Calibration of scatter intensities through Mie theory The output of IFCM signal intensities are presented in arbitrary units (a.u.), which hinders data comparability (and reproducibility) with different flow cytometers. Since light scattering of spherical objects is dependent on particle size and refractive index, Mie theory can be used to relate the scatter intensity of events to their size given their refractive index 32. Generally, Mie theory is applied to calibrate the scatter channels of a FC (forward- and/or sideward-scattered light - FSC or SSC, respectively); however, IFCM utilizes a brightfield detection channel (BF, Ch04) as opposed to FSC. Mie theory was applied on both scatter detection channels (BF and SSC). As a first step, we extracted the BF and SSC median scatter intensities of each identified size population of PS beads (Figure 1b). Coefficient of variation (CV) analysis for each single PS bead population showed scores ≥8% for the BF detector irrespective of bead size, whereas CV scores for the SSC detection channel were observed to increase with decreasing bead sizes – indicating that the detection of smaller particles is close to the detection limit of the SSC detector in our setup. Next, BF and SSC data of the PS beads were scaled onto Mie theory, resulting in a scaling factor (F) of 1.3518 and a coefficient of determination (R2) of 0.00 for the BF detector and a scaling factor of 8.405 and an R2 of 0.91 for the SSC detector (Figure 1c). Thus, signals from sub-micron PS beads measured with the BF detector do not provide quantitative information. The SSC detector, on the other hand, can be readily calibrated. For the SSC detector, the theoretical model indicates a plateau for EVs with a diameter between ~400 to ~800 nm, which translates into a low resolution when determining EV sizes based on SSC intensities within this region. To ensure inclusion of sub-micron EVs, a gate was set at SSC below the scattering intensity corresponding to the plateau, namely 400 nm EVs, corresponding to a value of 900 a.u. SSC intensity.

Chapter 3 34 These data show that 1) IFCM is able to readily discern sub-micron sized EVs based on their emitted fluorescence and SSC intensities, and 2) SSC – but not BF – light scattering intensities can be used to approximate particle sizes (following Mie calculations). The standardization of SSC signal intensities followed by the setting of a sub-micron gate provides a tool to selectively analyze all fluorescent EVs in complex samples such as plasma, as long as these particles emit detectable fluorescent intensities. Figure 1 – Calibration of scatter intensities through Mie theory. a) Gigamix polystyrene (PS) bead populations with sizes from 900 nm down to 100 nm were identified on the basis of SSC and FITC fluorescent intensities. b) Counts and median scatter intensities of each PS bead population as detected by the brightfield (BF) and side scatter (SSC) detectors (Ch04 and Ch06, respectively). c) Diameter vs Scattering cross section graphs. PS beads (green lines) were modelled as solid spheres with a refractive index of 1.5885 for a wavelength of 618.5 nm (brightfield) and 1.5783 for a wavelength of 785.0 nm (SSC). EVs (orange lines) were modelled as core-shell particles, with a core refractive index of 1.38 and a shell refractive index of 1.48 and a shell thickness of 6 nm for both wavelengths. The obtained scatter intensities of the PS beads as described in b were overlayed and a least-square-fit was performed to correlate theory and practice. Based on these correlations, SSC signal intensities were found to be indicative of particle size and a SSC cut-off of 900 a.u – corresponding to particles of 400 nm – was used in the rest in this work. F: scaling factor between scattering intensity and scattering cross section; n: refractive index.

An Imaging Flow Cytometry-Based Methodology for the Analysis of Single Extracellular Vesicles 3 35 IFCM gating strategy for the detection of single particles ≤400 nm in plasma EVs represent a heterogeneous group with different cellular origin. The analysis of single EVs, as well as the different subsets, will provide a better understanding of the pathophysiological state of the individual. Therefore, we designed a gating strategy to analyze individual sub-micron sized particles based on 1) the analysis of events within a pre-defined sub-micron size range, and 2) exclusion of multi-spot fluorescent events from our analysis. Based on the previous results, we selected all events with SSC intensities ≤ 900 a.u. - corresponding with particles of 400 nm and below. (Figure 2a – I). Next, we checked for multiplet detection within each separate fluorescent detection channel based on the number of fluorescent spots within the pixel grid for each acquired event: these spots were quantified by combining the “Spot Count” feature with the intensity masks for each of the channels used per experiment. Although the camera can spatially resolve signals originating from multiple simultaneously imaged EVs, the software anticipates that the signals are originating from multiple locations within 1 cell. By selecting all events that showed 0 or 1 spot, representing either negative or single-positive events for a fluorescent marker, we were able to exclude multiplet events from our analysis (Figure 2a – II, III). As a last step, we calculated the distance between individual fluorescent spots detected in different fluorescent channels to exclude any false double-positive events (defined as 2 different single-positive particles within the same event). To this end, we created a new mask by combining the intensity masks of the channels in use per experiment using Boolean logic (e.g., MC_Ch02 OR MC_Ch05), and combined this new mask with the “Min Spot Distance” feature to calculate the distance between the fluorescent spots across the detection channels used. We then excluded all fluorescent events that did not occupy the same location on the pixel grid (Figure 2a – IV). Ultimately, this gating strategy allows for the identification and subsequent analysis of single fluorescent sub-micron sized particles ≤400 nm in PPP and is applied throughout the rest of this work. Establishment of IFCM background fluorescence Given their physical characteristics, EVs yield faint fluorescent signals – compared too cells – when measured with IFCM. Therefore, we assessed the fluorescent background levels induced by our staining protocol. As no washing steps are performed, the discrimination of EVs from fluorescent background signals is

Chapter 3 36 required to exclude false-positive particles from analysis. 0.20 µm filtered PBS (fPBS - Buffer Control) and platelet-poor plasma (PPP) samples from 5 healthy individuals was stained with CFDA-SE (carboxyfluorescein diacetate succiminidyl ester) or a mixture of tetraspanin-specific antibodies (anti-CD9/anti-CD63/anti-CD81) labeled with APC. CFDA-SE is a non-fluorescent molecule converted to fluorescent CFSE (carboxyfluorescein succiminidyl ester) by intravesicular esterases. This helps to discriminate EV from lipoproteins, as the latter do not contain esterase activity. PPP samples left unstained or singly stained with CFSE (Ch02) or the tetraspaninspecific antibody mixture (Ch05) were used to set the gating areas (Figure 2b) and compensation matrix. Following our gating strategy, analysis of unstained fPBS or unstained PPP or fPBS + CFSE resulted in ~E5 single-positive objects/mL within the CFSE gating area. In contrast, PPP samples single stained with CFSE showed an average of 4.23E7 ± 7.28E6 objects/mL (mean ± standard deviation), representing a 100-fold higher CFSE single-positive particle concentration compared to the unstained samples and fPBS (Figure 2c, left panel). Similarly, analysis of positive fluorescent events upon staining with the tetraspaninspecific antibody mixture showed that fPBS + anti-tetraspanin antibodies (fPBS Mix) yielded 5.98E6 objects/mL – a 3.6-fold increase over the concentrations of fPBS Unstained (1.65E6 objects/mL). Additionally, an isotype control was added to analyze the specificity of the antibodies in the tetraspanin mixture. Positive particle concentrations were obtained for both fPBS and PPP Isotypes, (6.16E5 and 1.97E5 ± 1.07E5 objects/mL, respectively). Analysis of PPP + anti-tetraspanin antibodies (PPP Mix) revealed an average of 1.69E8 ± 1.44E8 objects/mL – a 28-fold higher particle concentration than fPBS + anti-tetraspanin antibodies, a 350-fold higher particle concentration than PPP Unstained (4.86E5 ± 2.6E5 objects/mL), and an approximate 860-fold higher particle concentration than PPP Isotypes (Figure 2c, right panel). An approximate 4-fold higher concentration of fluorescent particles was observed in the PPP Mix vs CFSE after subtraction of background concentrations before comparison. Together, these findings show that positive fluorescently stained events can be successfully discriminated from background signals and that the anti-tetraspanin antibody binding in our protocol is specific. Moreover, as unstained samples and isotype controls yielded ~E5 (for CFSE) and fPBS with anti-tetraspanin antibodies yielded ~ E6 objects/mL in their respective fluorescent channels, we established the level of the background concentrations in our setup for single positive fluorescent events at E5 and E6 objects/mL, for CFSE ant anti-tetraspanin antibodies respectively.