Heterogeneity in MIR21 content of MCF-7 exosomes and comparative study of their thermal stability

Globally, the subject of exosomes (defined as extracellular nano-vesicles (30 nm to 130 nm in diameter) secreted by many types of living cells and found in all body fluids, including blood, urine, and saliva) has received significant attention over the past 15 years, but with more intensive research during the last five years. The promising future of exosomes in diagnostics and therapeutics can be seen through the growing number of acceptable clinical trials using exosomes, mainly for the early-stage detection of cancer. This research on exosomes involves two projects: The first project is to characterize tumor exosomes secreted by female breast cancer cells (MCF-7) using three biophysical properties (size, mass, and density) that are correlated to one molecular property: the level of miR21, a standard cancer biomarker. Our developed approach revealed that tumor (MCF-7) exosomes are heterogeneous in miR21. The second project aims to characterize exosomes through the influence of their thermal stability. Using our approach, Time-Temperature Method (TTM), we were able to demonstrate that distinguishing tumor (MCF-7) and normal-like (MCF-10A) exosomes by their thermal stability is possible with 33% average difference and a significant statistical result (P-value < 0.001). Accordingly, our results show, for the first time to our knowledge, that exosome-thermal stability is an important biophysical property for differentiation between different populations of exosomes. The new methods and results presented in this study could, with further research, be developed into a point-of-care device for the early detection of cancer.

HETEROGENEITY IN MIR21 CONTENT OF MCF-7 EXOSOMES AND COMPARATIVE STUDY OF THEIR THERMAL STABILITY by Samer Mohammed Hakami A dissertation submitted to the faculty of The University of Utah in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Chemical Engineering The University of Utah December 2017 Copyright © Samer Mohammed Hakami 2017 All Rights Reserved The University of Utah Graduate School STATEMENT OF DISSERTATION APPROVAL The dissertation of Samer Mohammed Hakami has been approved by the following supervisory committee members: Mikhail Skliar , Chair 4/6/2017 Date Approved Philip Bernard , Member 4/6/2017 Date Approved Jules Magda , Member 4/6/2017 Date Approved Leonard Pease , Member 4/6/2017 Date Approved Swomitra Mohanty , Member 4/6/2017 Date Approved and by the Department/College/School of Milind Deo , Chair/Dean of Chemical Engineering and by David B. Kieda, Dean of The Graduate School. ABSTRACT Globally, the subject of exosomes (defined as extracellular nano-vesicles (30 nm to 130 nm in diameter) secreted by many types of living cells and found in all body fluids, including blood, urine, and saliva) has received significant attention over the past 15 years, but with more intensive research during the last five years. The promising future of exosomes in diagnostics and therapeutics can be seen through the growing number of acceptable clinical trials using exosomes, mainly for the early-stage detection of cancer. This research on exosomes involves two projects: The first project is to characterize tumor exosomes secreted by female breast cancer cells (MCF-7) using three biophysical properties (size, mass, and density) that are correlated to one molecular property: the level of miR21, a standard cancer biomarker. Our developed approach revealed that tumor (MCF-7) exosomes are heterogeneous in miR21. The second project aims to characterize exosomes through the influence of their thermal stability. Using our approach, Time-Temperature Method (TTM), we were able to demonstrate that distinguishing tumor (MCF-7) and normal-like (MCF-10A) exosomes by their thermal stability is possible with 33% average difference and a significant statistical result (P-value < 0.001). Accordingly, our results show, for the first time to our knowledge, that exosome-thermal stability is an important biophysical property for differentiation between different populations of exosomes. The new methods and results presented in this study could, with further research, be developed into a point-of-care device for the early detection of cancer. iv TABLE OF CONTENTS ABSTRACT ....................................................................................................................... iii LIST OF TABLES ............................................................................................................ vii LIST OF ACRONYMS AND ABBREVIATIONS ........................................................ viii ACKNOWLEDGEMENTS ............................................................................................... ix Chapters 1. INTRODUCTION ........................................................................................................ 1 1.1 Motivation ....................................................................................................... 1 1.2 Background and Literature Review .................................................................. 2 1.3 Formation of Exosomes .................................................................................... 3 1.4 Size of Exosomes .............................................................................................. 4 1.5 Shape of Exosomes ........................................................................................... 4 1.6 Molecular Cargo of Exosomes.......................................................................... 4 1.7 Biomarkers of Exosomes .................................................................................. 5 1.8 Isolation of Exosomes ....................................................................................... 6 1.9 Diversity of Extracellular Vesicles ................................................................... 7 1.10 Functions of Exosomes ................................................................................... 8 1.11 Exosome Characterization and Quantification Methods ................................ 9 2. HETEROGENEITY IN MIR21 LOADING OF MCF-7 EXOSOMES ..................... 27 2.1 Abstract ........................................................................................................... 27 2.2 Introduction ..................................................................................................... 28 2.3 Materials and Methods .................................................................................... 30 2.4 Results ............................................................................................................. 40 2.5 Discussion and Conclusion ............................................................................. 44 3. COMPARISON OF THERMAL STABILITY OF MCF-7 AND MCF-10A EXOSOMES ............................................................................................................... 63 3.1 Abstract ........................................................................................................... 63 3.2 Introduction ..................................................................................................... 63 3.3 Materials and Methods .................................................................................... 66 3.4 Results ............................................................................................................. 72 3.5 Discussion ....................................................................................................... 82 3.6 Implications..................................................................................................... 87 3.7 Conclusion ...................................................................................................... 89 4. GENERAL CONCLUSIONS AND FUTURE WORK ........................................... 112 4.1 General Conclusions ..................................................................................... 113 4.2 Future Work .................................................................................................. 114 REFERENCES ............................................................................................................... 116 vi LIST OF TABLES 1.1. Exosome isolation methods………………………………………………………19 2.1 Run conditions applied for the CFFF experiment ……………………………… 48 2.2 Summary of the collected CFFF mass fractions …..…………………………… 49 2.3 Run conditions applied for the AFFF experiment ..…...………………………... 51 2.4 Summary of the collected AFFF mass fractions …..………………….….…..… 52 3.1 Incubation conditions applied to study the thermal stability of exosomes …..…. 97 3.2 Experimental procedure (sampling) applied to study the thermal stability of tumor (MCF-7) and normal-like (MCF-10A) exosomes ………..…………………….……...98 3.3 The experimental procedure of the blind, randomized experiment…………..…102 LIST OF ACRONYMS AND ABBREVIATIONS MVB Multivesicular Body EVs Extracellular Vesicles MVs Microvesicles FACS Flow Cytometric Analysis LC Liquid Chromatography MS Mass Spectrometry SEM Scanning Electron Microscopy TEM Transmission Electron Microscopy CFFF Centrifugal Field-Flow Fractionation AFFF Asymmetrical Field-Flow Fractionation NTA Nanoparticle Tracking Analysis dPCR Digital Droplet PCR cDNA Complementary DNA DLS Dynamic Light Scattering ES-DMA Electrospray-Differential Mobility Analysis ACKNOWLEDGEMENTS “Success is measured not so much by the position one has reached in life, as by the obstacles one has overcome while trying to succeed.” - Booker T. Washington. The five years I spent to finish my Ph.D. were not a simple journey at all. Indeed, this journey was a real challenge, full of exciting adventures. During these years in my program, I was between two points: the first day, which started with a lack of real knowledge, and the final destination that ended with profound knowledge, knowledge that should lead to a profound change in my life. Therefore, I would like to express my gratitude and sincere appreciation to my research advisor, Dr. Mikhail Skliar, and to my supervisory committee: Dr. Philip Bernard, Dr. Jules Magda, Dr. Leonard Pease, and Dr. Swomitra Mohanty, for their help, continuous support, and encouragement during my program. My thankfulness goes also to my research group: Dr. Vasiliy Chernyshev, Dr. Soheyl Tadjiki, Rakesh Rachamadugu, and Inge Stijleman, for their teamwork and support offered in a number of ways. My acknowledgments are also extended to all people in my department of Chemical Engineering. I am also very thankful and grateful for the support provided by two Universities: Jazan University (Jazan, Saudi Arabia) for sponsoring and funding my education throughout these years, and The University of Utah for the assistantship and the opportunity to be one of the U’s students. Finally, I will not forget to send my special thanks and profound gratitude to my family members and relatives not only for their moral, and great, support, but also for their motivation throughout the entire course of my Ph.D. program. x CHAPTER 1 INTRODUCTION 1.1 Motivation This study was motivated by the need for a convenient, effective, and noninvasive tool for an early detection of cancer, leading to an improved survival rate of patients. To fulfill this motivation, the main purpose of the presented research was to develop a novel method that can be used to characterize exosomes (biological nanoparticles), due to their potential in the early diagnosis of abnormalities. When a disease, in particular cancer, affects certain body cells, the exosomes released by these tumor cells contain some molecular contents and biomarkers. These biomarkers can be used for early cancer detection. This profound knowledge about exosomes has resulted in an intense competition between research centers around the world to develop a simple, cheap, and reliable exosome-based tool with easy protocol for the early detection of cancer. The number of exosome publications is growing rapidly, as shown in Figure 1.1. During the first 20 years after their discovery in 1981, there was little interest in exosomes. Only in recent years have we seen a dramatic increase in the number of publications. This explosive growth reflects the recognized importance of exosomes and the many promising applications, particularly in cancer biomarker discovery. 2 1.2 Background and Literature Review Exosomes are extracellular nano-vesicles (30 nm to 130 nm in diameter) secreted by many types of living cells and found in all body fluids, including blood, urine, and saliva. Their molecular cargo, such as nucleic acids and proteins, derived from a secreting cell and delivered to the recipient cell is a signaling mechanism with strong regulatory effects on many biological processes, including immune response and cellular survival.1,2 Different type of cells, including cancer cells, continuously release extracellular vesicles into the extracellular space.1 Trams et al. (1981)3 were the first to coin the term “exosomes” as a reference to the smallest of the membrane vesicles. C. Harding et al. (1983)4 described the release of small vesicles from reticulocytes of rats. Two years later, Pan et al. (1985)5 used an electron microscopic to study the exocytosis of exosomes in sheep. Pan et al. (1985) demonstrated the initial genesis of exosomes and characterized their size (about 50-nm vesicular round bodies). Johnstone et al. (1987)6 were the first to isolate and harvest exosomes by centrifugation (at 100,000 X g for 90 minutes). They also confirmed the ability of exosomes to retain different biological activities and characteristics. The research group of Valadi et al. (2007)7 published an article showing that exosomes can also contain nucleic acids. These nucleic acids carried by exosomes are mainly RNA species, including messenger RNA (mRNA) and microRNAs (miRNAs). They also demonstrated that RNAs are confined inside exosomes, not on their external structures. More interestingly, the group revealed that translation of mRNA is possible if exosomal mRNA is transferred into another cell. Another milestone in the field was achieved when Chevillet et al. (2014)8 reported that only a very small fraction of exosomes in a given population contains even a single miRNA molecule. Exosome 3 formation, compositions, and biological functions remain active areas of research. 1.3 Formation of Exosomes Originally, exosomes were assumed to be “garbage bags” ejected form cells with unnecessary intercellular contents. This original view has evolved. At the present time, exosomes are seen as nano-vesicles participating in cellular signaling by mediating intercellular communication between paracrine and distal cells. Many, if not all, types of cell release exosomes, including epithelial and endothelial cells, T and B cells, platelets and mast cells. Both normal and afflicted cells released exosomes, including tumor cells, which is the rational basis of diagnostic potential of the exosomes in the approach known as liquid biopsy.1,9–12 Figure 1.2 illustrates the mechanism of exosome biogenesis, followed by their release into circulation.9 It starts with the cell membrane budding inward to form an early endosome -- a small membrane-bounded compartment, 200-500 nm in diameter. The cytoplasmic content of the early endosome is internalized in a process called “inward invagination of the endosomal membrane” to form exosomal vesicles 30-130 nm in diameter. The described sequence of transformations results in the formation of a multivesicular body (MVB), which contains many exosomal vesicles. In the final step, when the MVB fuses with the cell membrane, exosomes are released out of the parent cell into the extracellular space and potentially into circulating blood or other body fluids. The figure also illustrates the formation of larger vesicles (microvesicles above 500 nm) by membrane shedding. Figure 1.3 confirms the described biogenesis by a sequence of high-resolution TEM images.12 4 Extracellular vesicles (EVs) released into the extracellular environment are classified into two main categories10,13: 1- Exosomes are membrane vesicles originating from endosomal pathway. 2- Microvesicles (MVs), on the other hand, are membrane vesicles that bud directly from the cell membrane. 1.4 Size of Exosomes It is usually assumed that the exosome diameters are in the range between 30 to 130 nm. The corresponding volume is 4.2–380 yl (yoctoliter, 1yl=10-24 liter). The expected total cargo per exosome is less than 10,000 net nucleotides of nucleic acid and less than 100 proteins.1 1.5 Shape of Exosomes There are different studies and review articles 10,14,15 that report on the morphology of exosomes. They are either spherical or near-spherical. Figure 1.4 shows the spherical shape of exosomes under the electron microscope.14 1.6 Molecular Cargo of Exosomes The molecular cargo of exosomes is divided into three major families: proteins, nucleic acids, and lipids. Each family is divided into different categories, as shown in Figures 1.5 and 1.6. The wide diversity in the molecular cargo shuttled by exosomes is attributed to different factors 16–19: 1- The cellular origin (cell type) from which the exosomes were secreted, 5 2- The health status of the secreting cell, and 3- Extracellular motivators (or stimuli). The diversity in exosome molecular content plays an important role in cell signaling and cell-cell communication. It is believed that the cargo of exosomes is transferred and delivered from parent cells to target cells as the mechanism of intercellular communication. These target/recipient cells could be either neighboring cells that are close to the exosome-secreting cells, or remote cells, as in the case of cancer metastasis.16,20 1.7 Biomarkers of Exosome Exosome biomarkers are mainly surface/membrane proteins that differentiate exosomal vesicles from other extracellular vesicles and determine their endosomal origin and lipid composition.21,22 One of the common methods used to confirm exosomal identity is through the expression of CD63 -- a surface protein often used as an exosome biomarker. Other common exosomal biomarkers are ICAM1 (intercellular Adhesion Molecule 1), ANX A5 (a specific protein in the annexin group), and TSG101 (Tumor susceptibility gene 101 protein) (Figure 1.6). 1.7.1 Exosomal Proteins Exosomes contain different types of membrane and cytoplasmic proteins such as transport proteins, fusion proteins, and lipid-related proteins.21,22 A detailed proteomic analysis of different types of exosomes has been performed. One of the references23 highlighted that there are more than 4400 different types of proteins that have been 6 recognized to be associated with exosomes. Another study24 reported 1132 proteins present in urinary exosomes and described in a database.16 1.7.2 Exosomal Nucleic Acids Nucleic acids are used as one of the positive biomarkers for exosome analysis and characterization, as it has been shown that the luminal volume of exosomes contain mRNAs and miRNAs (non-coding RNAs).7 Some authors25–27 found that the exosomal RNA cargo is different from their parental cells. In contrast, another group of authors28,29 found that the miRNA cargo in tumor exosomes is comparable to that of the originating cancer cells. Hence, miRNA content in circulating exosomes could be used to detect the presence of tumor cells and as a diagnostic tool for cancer. 1.8 Isolation of Exosomes Exosomes can be isolated from different sources, including a cell-line growth media and different body fluids, such as urine, blood, saliva, and breast milk. Currently, there are different experimental procedures used to isolate, concentrate/enrich, and purify exosomes from the sample fluid1,30–32 based on such properties as their size and/or density. There are different methods in use to isolate exosomes from biological fluids, such as differential centrifugation and size exclusion chromatorography.33 Table 1.1 shows mechanisms, advantages, and disadvantages of the major methods, which are currently used for exosome isolation.33 Despite multiple options, our ability to obtain highly purified exosomal nanoparticles is still limited. This is because both blood and 7 cell cultural media contain many non-vesicular nanoparticles with the same size as exosomes.1 Therefore, the subject of having different alternative techniques to isolate and purify exosomes is still very important. 1.9 Diversity of Extracellular Vesicles Extracellular vesicles (EVs) is a generic term used to describe all kinds of extracellular vesicles, including exosomes, ectosomes, oncosomes, prostasomes, microvesicles, apoptotic bodies, and matrix/calcifying vesicles.34 Several analytical and fractionation techniques, some of which are listed below, can be used to confirm the enrichment of exosomes in the isolated EV population and further purify the sample. 1- Electron microscopy: The electron micrographs can confirm the rounded structure of the exosomes with an approximate size range 30–80 nm (Figure 1.7). 2- Flow cytometric analysis (FACS): this analysis can confirm the identity of exosomes through the presence of surface protein CD63, which is a common exosomal maker.34 3- Detailed protein analyses (proteomic analysis): this analysis is usually performed using the combination of liquid chromatography (LC) with mass spectrometry (MS), LC-MS/MS.30 4- Other techniques such as gel electrophoresis and western blotting can be used to detect exosome markers, such as CD63.7 As a rule, it is necessary to provide additional evidence to confirm the enrichment of exosomes among many other isolated particles. 8 1.10 Functions of Exosomes The evidence is beginning to emerge that cells use exosomes to convey complex messages needed for intercellular communication. It is held that exosomes are able to travel between cells to transfer their cargo to recipient cells.1 A relatively recent discovery is the packaging of mRNA and miRNA into exosomes,7 which indicated that exosomes can deliver genetic materials from one cell to another. Figure 1.8 illustrates three pathways of exosome-mediated signaling10: 1- Exosomes are internalized and get inside the recipient cells by the process of endocytosis35 2- Exosomes fuse with the plasma membrane of the target cell to deliver their surface proteins and all internal content/cytoplasm to the recipient cell,36,37 or 3- Exosomes associate with the plasma membrane and remain stably bound. The signaling can also occur through receptor–ligand interactions, which may involve antigen presentation.30 The exosomal communication is believed to occur under both healthy and diseased18,38,39 conditions. Exosomes also play a role in expelling undesirable RNA and proteins, as was originally hypothesized. This cell maintenance function is believed to be essential in protecting cells and keeping their biological performance healthy.6,40 It is believed that there are two main advantages of exosomes as mediators of intercellular communication18: 1- The exosomal message can be targeted into either specific or multiple locations of recipient cells. 2- The multiple miRNAs encased in exosomes make exosomes able to deliver 9 multiple messages simultaneously. Such messages may be related to different cellular processes, such as gene expression, cell growth, division, survival, apoptosis, etc. It is assumed that exosomes could have specific adhesion molecules (surface legends) that enable exosomes to target a specific host cell. Overall, the functional role of exosomes in health and disease is not well understood and remains the subject of intensive research. 1.11 Exosome Characterization and Quantification Methods Cryo-transmission electron microscopy (cryo-TEM) is the gold standard in characterizing the size and the morphology of exosomal vesicles. Other electron microscopy techniques, such as scanning electron microscopy (SEM) and transmission electron microscopy (TEM), have also been widely used. The main advantage of the cryo-TEM compared to the alternatives is that the sample is imaged in its innate hydrated state. A recent study14 published by our group, summarized in Figure 1.9, discusses the analytical techniques available for exosome sizing. The characterization methods used in the current study can be divided into two categories: 1- Methods to characterize the biophysical and thermal properties of exosomes. To separate and characterize the exosomes by their mass and hydrodynamic diameter, we used centrifugal field-flow fractionation (CFFF) and asymmetrical field-flow fractionation (AFFF), respectively. The concentration and the distribution of the hydrodynamic diameters were characterized by nanoparticle tracking analysis (NTA) before and after fractionation, and prior 10 and after subjecting the sample to thermal treatment inside a digital dry block heater. 2- The molecular analysis of the exosomes was limited to the characterization of surface biomarkers using antibody array and the number of luminal copies of miR-21, a well-known onco-microRNA, quantified by digital droplet PCR (ddPCR). 1.11.1 Size Fractionation by AFFF Asymmetrical Field-Flow Fractionation (AFFF)41–43 separates heterogeneous population of nanoparticles based on their hydrodynamic diameter. The principal of AFFF operation is summarized in Figure 1.10. The separation takes place in a narrow flow channel. When particles travel in a parabolic pressure-driven carrier flow, a perpendicular external field flows across the channel pushes particles against the bottom wall of the channel. The higher diffusivity of smaller particles increases the probability of their diffusion into the region of higher flow rates. Consequently, the smallest particles are eluted first, followed by the fractions of increasingly higher hydrodynamic sizes. 1.11.2 Mass Fractionation by CFFF Centrifugal Field-Flow Fractionation (CFFF)41–43 separates nanoparticles based on their mass. The separation principle is shown in Figure 1.11. At the start of the separation, particles are pushed against the channel wall by centrifugal field imposed by rotating the channel at speeds up to 5000 rpm. The parabolic carrier flow first elutes the lightest particles that are more likely to diffuse into the region of higher flow rates 11 because of lower centrifugal forces acting on them. Heavier particles are eluted at later times. 1.11.3 Measurements of Hydrodynamic Sizes by NTA Nanoparticle Tracking Analysis (NTA)13,44–46 was used to measure the concentration and hydrodynamic size distribution of exosomes. NTA uses scattered light to track the number of particles and their motion in the field of view. The motion of each individual particle is then used to calculate its diffusivity and the corresponding hydrodynamic diameter. The size distribution is obtained from the particle-by-particle analysis of a large number of exosomes of different hydrodynamic sizes present in the sample. The NTA concept is illustrated in Figure 1.12. The laser beam passing through a liquid sample injected into the test chamber is scattered by nanoparticles. The scattered light is collected by 20x microscope objective and captured with a digital video camera at an approximate rate of 30 frames per second (fps). The Brownian motion captured in video is used to estimate the diffusivity of each particle from which the corresponding diameter is estimated assuming a spherical shape of the particles. A typical video frame showing several particles in the field of view of the instrument is given in Figure 1.13(A). Figure 1.13(B) shows the movement of individual particles in the field of view. The size measurement starts by identifying the center of each observed particle illuminated by a laser beam. The average distance that the particle moves in the x and y plane is used to estimate its diffusion coefficient (D). Then by applying Stokes-Einstein equation !! (𝐷 = !!"!! , where: η is solvent viscosity, T is the sample temperature, KB is Boltzmann’s ! constant, and dh is the hydrodynamic diameter), the hydrodynamic diameter (dh) of the 12 spherical particle is obtained. The particle concentration in the sample is estimated from the average number of particles in the field of view of known volume, which is 100 µm by 80 µm wide and 10 µm in depth. 1.11.4 miR21 Quantification by PCR The luminal abundance of onco-microRNA miR21 was performed by digital droplet PCR. After the miRNA extraction and the synthesis of complementary DNA (cDNA), the sample with added PCR reagents is broken down into 10 million droplets. The initial sample concentration is adjusted to the level at which multiple copies of the target are unlikely to be present in a single droplet. After the PCR is performed on individual droplets, the integrated fluorescence probes are used to provide high-resolution read/detect for every droplet to determine whether it contains a copy of the target, miR21. The results are then used to estimate the average number of target copies per exosome in the sample.47 13 1500" 1400" 1300" 1200" 1100" Publica(ons, 1000" 900" 800" 700" 600" 500" 400" 300" 200" 100" 0" 1981" 1984" 1987" 1990" 1993" 1996" 1999" 2002" 2005" 2008" 2011" 2014" 2017" 2020" Years, Figure 1.1. Exosome-research timeline: It shows the significant growth in number of publications in recent years. Data are based on PubMed-referenced papers (http:/ncbi.nlm.nih.gov/pmc) using search words “exosomes” and “exosome”. 14 Figure 1.2. Mechanisms of formation and release of exosomes and microvesicles. (A) Secretion of proteins out from the cell through the endoplasmic reticulum and Golgi body. (B) Exosomes are formed first inside an early endosome (also known as multivesicular body, MVB). Upon fusion with the cell membrane, exosomes are released into the extracellular space. The “Y”-shapes on the exosome membrane represent membrane proteins inherited from the parent cell. (C) Microvesicles (500–1000 nm) are formed directly by shedding from the cell membrane into extracellular space. Adapted from9 15 Figure 1.3. The morphological analysis of exosome biogenesis under TEM. It shows the stages of exosome formation inside a CD34 human cell followed by the exosome release. (i) The cytoplasm of the cell with MVBs, which encircles many exosomes (inset, arrows). (ii) Arrows indicate the beginning of exosome formation (the inward invagination) from the MVB membrane. (iii) Fusion of MVB with the cell membrane. (iv) Release of exosomes out from the cell. Adapted from12 16 Figure 1.4. The image shows the spherical shape of exosomes derived from pancreatic cancer cells, which are visualized with cryo-transmission electron microscopy (TEM). Arrows indicate a faint density surrounding some exosomal particles, suggesting the presence of surface macromolecules. Adapted from14 17 Figure 1.5. The molecular cargo of extracellular vesicles, including exosomes, is divided into three major families: Proteins, Nucleic acids, and Lipids. Adapted from19 18 Figure 1.6. Molecular cargo of exosomes include miRNA, DNA, surface, and luminal proteins. Adapted from16 19 Table 1.1. Summary of exosome isolation methods. Adapted from33 20 Figure 1.7. Electron micrographs of exosomes isolated from MC/9 and HMC-1 cells. Both images show small vesicles approximately 50–80 nm in diameter (100 nm scale bar). HMC-1 exosomes were labeled with anti-CD63 nanoparticles. Adapted from7 21 Figure 1.8. Interactions of exosomes with recipient cells happen through three pathways. Pathway 1: Exosomes remain stable and dock at the plasma membrane (1); Pathway 2: Exosomes fuse with the plasma membrane (2); Pathway 3: Exosomes are endocytosed by the plasma membrane (3). Then endocytosed exosomes fuse with the membrane of an endocytic compartment (4). The delivery of proteins and RNA into cytosol of the target cell occurs through pathways 2 and 3. Symbols: Membrane-associated proteins (triangles), Transmembrane proteins (rectangles), RNAs (curved symbol). Adapted from10 22 Figure 1.9. Methods of size and shape characterization of hydrated and desiccated exosomes. (A) Cryo-TEM produces images of two-dimensional projections for exosome geometry in their hydrated state. (B) Nanoparticle tracking analysis (NTA) and dynamic light scattering (DLS) can be used to characterize hydrodynamic sizes of exosomes, which were found to be larger than exosome geometric sizes. (C) Electrospraydifferential mobility analysis (ES-DMA) separates and sizes particles based on their charge-to-size ratio after aerosol desiccation. (D) SEM produces images for exosome geometry in their desiccated state. Exosome surface desiccation results in a “cup-shaped” appearance. Adapted from14 23 Figure 1.10. Separation mechanism of asymmetrical field flow fractionation (AFFF): The parabolic flow profile of the carrier fluid has high velocity away from the channel walls. Initially, the flow across the channel pushes particles against the wall of the channel. Smaller particles with higher diffusion more readily migrate towards the center of the channel and are eluted before the larger particles. Adapted from42 24 A B Figure 1.11. Separation mechanism of centrifugal field-flow fractionation (CFFF): (A) First the centrifugal field is applied to push particles against the wall of the channel. The parabolic carrier flow sweeps the particles downstream. Lighter particles experience smaller centrifugal force and are more likely to diffuse towards the center of the channel. As a result, they are eluted before heavier particles. (B) The centrifugal force (Fc) acting on the particles is proportional to particle mass (m), rotational speed (ω2), and particle diameter (r): Fc=mω2r. Adapted from42 25 Figure 1.12. Nanoparticle tracking analysis (NTA) estimates the size of the particles based on their displacement imaged as a scattered light. Adapted from13 26 (A) (B) Figure 1.13. (A) Representative image for the video used to determine the size of individual exosomes. The red points show the position of individual particles as objects in the frame. (B) A path of one tracked particle with time is indicated by the black spots. The position of each spot is determined from the video captured at 30 frames per second. The total time used to track this particle before it disappeared from the focal plane of the camera was 4 seconds, which corresponds to 120 video frames captured in 4 seconds. Adapted from45,48 27 CHAPTER 2 HETEROGENEITY IN MIR21 LOADING OF MCF-7 EXOSOMES 2.1 Abstract It reasonable to expect that exosomes released by the same cell type and cultured under controlled conditions would secrete exosomes with similar molecular content and biophysical properties. We tested this hypothesis for exosomes secreted by MCF-7 female breast cancer cells. The exosomes were characterized and fractionated by their hydrodynamic size, mass, and density. The content of onco-miroRNA miR21 in different fractions was then quantified by the digital droplet PCR. Our results were at variance with our expectations and showed heterogeneous miR21 loading. We found that lighter (in the average-mass range 0.2-0.25 femtogram, fg) and smaller (average hydrodynamic diameter 50-70 nm) exosomes are more abundant in miR21 than the heavier (mass ~ ≥ 0.3fg) and larger (diameter ~ ≥ 80nm) MCF-7 exosomes. Sequential fractionation of exosomes, first by their mass, followed by subfractionation by size, allowed us to obtain density fractions of the exosomes. Our preliminary results indicate that certain density fractions are substantially enriched in miR21. 28 2.2 Introduction Characterization of biological nanoparticles by size, mass, density, and their molecular content is difficult. Molecular composition of exosomes broadly falls into three main types: nucleic acids (mainly RNA species: miRNAs and mRNAs), proteins, and lipids. The molecular heterogeneity of exosome (both in level and composition) are attributed to three factors: (1) cells of origin49,50; (2) the health status of individuals, with cancer patients having elevated levels of circulating miRNAs compared to normal individuals29,51,52; (3) environmental conditions, such as hypoxia and stress.53,54 Several studies examined the molecular composition of exosomes as a function of the cell type that secretes them.55–57 The review by Ferguson et al.58 notes that, even if exosomes are isolated from the same parental cells, the content and type of exosomal nucleic acids, proteins, and lipids are heterogeneous. The investigation of exosomal RNAs (mainly non-coding miRNAs) has received significant attention.58 However, studies focused on the heterogeneity of exosomal miRNAs are rare.58 Chevillet et al.8 showed in their quantitative study that the exosomal miRNAs are distributed non-uniformly among the exosome population. Using different stoichiometric models, the authors illustrated that many exosomes are devoid of miRNA copies and, on average, it takes 100 exosomes to find a single miRNA copy. Ferguson et al.58 showed that, although endothelial exosomes are usually more enriched in miRNAs (~90% of all non-coding RNAs) than non-endothelial exosomes, this does not hold for exosomes isolated from breast cancer cells. They found that the enrichment of miRNAs in exosomes released by metastatic breast cancer cells (MDA-MB-231) and breast cancer cell line (MCF-7) are only ~5% and ~1%, respectively, of all non-coding RNAs. 29 Exosome isolation methods in current use produce a heterogeneous population of vesicles, which may also be contaminated by microvesicles and other non-exosomal particles. For example, the canonical differential ultracentrifugation method was reported to co-purify vesicles of endosomal and non-endosomal origin.59 Several approaches use exosomal biomarkers to purify the exosomes even though the surface biomarkers themselves are known to be heterogeneously expressed.21–24 As reported,55,58,60 these new approaches can be broadly classified into label-based methods, such as fluorescenceactivated cell sorting (FACS) and immunocapture, and label-free methods, such as Laser Tweezer Raman Spectroscopy (LTRS). Despite a significant development effort, there is currently no single isolation method that guarantees that the isolate contains all exosomes present in the original biofluid sample without contamination by non-endosomal particles.58 The objective of our study was to characterize the distribution of onco-microRNA miR21 in different subpopulations of extracellular particles in the range of exosomal sizes. We specifically examined the distribution of miR21 in mass, size, and density fractions of exosomes isolated from the growth medium of MCF-7 breast cancer cells. To our knowledge, this is the first attempt to characterize miRNA distribution in biophysically different fractions of exosomes. Prior work by Paulaitis et al.61 examined the exosomal miRNA in fractions defined by different expression of surface biomarkers. To achieve the goal of the study, the population of the isolated MCF7 exosomes was fractionated by their mass via Centrifugal Field Flow Fractionation (CFFF). The same population was also fractionated by their hydrodynamic diameter using Asymmetrical Field Flow Fractionation (AFFF). The distinct density fractionations were 30 obtained by first performing mass fractionation by CFFF, followed by AFFF size subfractionation of one of the mass fractions. The hydrodynamic size of the particles in the obtained mass, size, and density fractions was characterized by Nanoparticle Tracking Analysis (NTA). The downstream molecular analysis to quantify for the abundance of miR21 in different fractions was performed by digital droplet PCR (ddPCR). Our literature review indicates that AFFF has been previously used to size fractionate the exosomes,17–66 but the current work is the first to performed the CFFF mass fractionation and the density fractionation by the sequential FFF mass-size fractionation. 2.3 Materials and Methods 2.3.1 Cell Culture and Isolation Prior to use, MCF-7 breast cancer cells (American Type Culture Collection (ATCC), Manassas, VA, USA) were stored in liquid nitrogen. Using an ExoQuick kit (SBI, Mountain View, CA, USA ), exosomes were isolated from 1 ml of serum. Following the manufacturer’s protocol, 30 ml of the growth medium was centrifuged at 3000×g for 15 minutes in order to remove cells and cell debris. The supernatant was transferred to a new vial contained 6ml ExoQuick-TC precipitating solution and the mixture incubated overnight at 4°C. The exosomes were pelleted by centrifuging the mixture at 1,500×g for 30 minutes, and the supernatant was discarded. After the isolation, the exosomes were re-suspended in either 450µl 2mM ammonium acetate or 450µl phosphate buffered saline (PBS) solution. The sample was subdivided into multiple 100µl aliquots, which were stored at -80°C prior to use.14,47 31 2.3.2 Exosome Biomarkers Known surface biomarkers were used to confirm that the obtained isolate is enriched in exosomes. The expression of eight known exosome biomarkers (CD81, CD63, ICAM1, ALIX, FLOT1, TSG101, ANXA5, and EpCam) and a negative control (cytoplasmic protein GM130) was characterized by commercial antibody assay (ExoCheck, SBI). First, the concentration of exosomal proteins was measured by spectrophotometry (NanoDrop ND-8000). After verifying that the concentration is in an acceptable range, a mixture of 600 µl of exosome lysis buffer (EXORAY, SBI) and 500µg of exosome protein were mixed by vortexing for 15 seconds. Then, in a fresh testing tube containing 9.4ml of Exosome Array Binding buffer, the above mixture was added and gently mixed (inverting the tube three times). In a clean tray containing 5ml distilled water at room temperature, the antibody array on polyvinylidene fluoride (PVDF) membrane was saturated for 2 minutes. After discarding the water, the Exosome lysate with a volume of 10 ml was added to the antibody array and incubated overnight on a rocker at 4°C. After discarding the liquid, 10 ml of 1X Array Wash Buffer was added to the antibody array and gently shaken at room temperature for 5 minutes. After repeating this step two additional times, the Wash Buffer was removed and 10 ml of Detection Buffer was added to the antibody array. After gentle shaking for 2 hours at room temperature, the Detection Buffer was removed, and 2 ml of Developer Solution was added and incubated for 2 minutes. The array was then imaged for 10 seconds using the Bio-Rad ChemiDoc XRS Imager System.14,47 32 2.3.3 Transmission Electron Microscopy The following procedure was used to image MCF-7 tumor exosomes. A copper mesh grid coated with Formvar-carbon was rendered hydrophilic by a glow-discharge treatment of the surface. A drop of an exosome solution (~ 3.5 µl) was then placed on the grid for approximately 1-2 minutes and then blotted away with filter paper. Then, a droplet (~3.5 µl) of 1% uranyl acetate solution was placed on the grid for ~20 seconds and then blotted away. The specimen was allowed to dry prior to imaging. In some cases, an intermediate washing step (2-3 seconds) in deionized water was applied before application of the stain. The dried specimen was imaged in a transmission electron microscope (JEM1400Plus, JEOL USA, Peabody, MA) at 120 kV. Images were recorded on a Gatan Orius camera (Gatan, Inc., Pleasanton, CA, USA).14,67 2.3.4 Exosomes Concentration and Size Measurements by NTA Within 24 hours of thawing the MCF-7 exosome samples stored at -80°C, the measurements of exosome concentration and size distribution were performed using nanoparticle-tracking analysis (NTA) (model LM10 with 40 mW 405 nm violet laser, Nanosight Instruments, Salisbury, UK). A high-sensitivity sCMOS camera (OrcaFlash2.8, Hamamatsu C11440) was used to capture the light scattered by the exosomes. The results were analyzed using the software provided by the manufacturer (Nanosight Version 3.0). When necessary, some samples were diluted to ensure the optimal concentration for the NTA analysis. The dilutions were performed using the same buffer as used during the fractionation (1/4 PBS). The pH of this buffer was measured to be approximately equal to 7.4. The buffer's viscosity was assumed to be 33 equal to the viscosity of water. The needle temperature probe was used to manually measure the temperature inside the NTA flow cell, with the results used during the NTA analysis. During all experiments, the temperature remained at ~ 20°C, implying a near constant viscosity of the solution at 1cP. The NTA analysis of each exosome fraction obtained from CFFF or AFFF experiments was performed within 5 minutes of collecting fractions.14,67 After diluting a sample to the point when approximately 30-100 particles/frame (2x108 - 2x109 particles/ml) are visible in the field of view, the sample was injected with a 1 ml sterile syringe into the NTA test-cell. A set of five videos (60-seconds each) was recorded for each sample (i.e., each sample was characterized five times). The results were averaged to obtain exosome concentration and size distribution (mode, mean, and the standard deviation). The NTA software settings for the analysis were as follows: camera level was set to 15; the acquisition rate was 25 frames per second; the shutter speed was set to 30 milliseconds (ms); the slider gain was equal to 500; the maximum jump mode, blur, and the minimum track length were set to Auto; the detection threshold for video processing was set to 4; and numbers of frames in each video was 1500. During the NTA analysis, exosomes were assumed to have a spherical shape.14,67 After the NTA data acquisition, the sample was aspirated from the test cell. Before analysis of a new sample, the test cell was rinsed with 3ml of 1X PBS to avoid sample cross contamination. After all samples were analyzed, the test cell was thoroughly rinsed with deionized water (DI) and wiped with 70% EtOH.14,67 34 2.3.5 Centrifugal Field Flow Fractionation (CFFF) The fractionation of MCF7 exosomes into different mass fractions was performed using the centrifugal field flow fractionation (CFFF, Figure 2.1). A diluted PBS (1:4 dilution) with reduced ionic strength (~0.03 mol/L) and neutral pH (~7.4) was used as a carrier fluid to prevent particle agglomeration and sticking to the CFFF flow channel. The background contamination of the carrier fluid was characterized by the NTA prior and was found to contain background particles at the concentration 1x106-107 particles/ml. The buffer was then subjected to the mass fractionation to ensure that the background particles are not concentrated in a narrow mass fraction. The operating conditions and run parameters used during the exosome CFFF experiments are summarized in Table 2.1 and Table 2.2, which indicates that the retention time for each fraction had a midpoint equal to 1 minutes. At our fractionation conditions, the first peak in the CFFF fractogram appears approximately 3 minutes after sample injection. It is known as a “Void Peak” and corresponds to non-retained/unseparated particles and molecules, such as proteins. The exosome containing CFFF mass fractions started with fraction #3 (a third fraction eluted between 3-4 minutes, as shown in Table 2.2) and continued until fraction #10. For each exosome fraction, the concentrations, the size distributions, and its mode and mean were measured by the NTA. The fractionation experiments were repeated three times. 2.3.6 Asymmetrical Field Flow Fractionation (AFFF) The asymmetrical flow field-flow fractionation (AFFF, Figure 2.2) was used to fractionate MCF7 exosomes by their hydrodynamic sizes. Similar to the CFFF case, the 35 background contamination in the carrier fluid was confirmed to be substantially lower than the concentration of the exosomes in the analyzed fractions. The operating conditions and run parameters during the AFFF experiments are summarized in Table 2.3 and Table 2.4, which shows that the retention time for each fraction had a midpoint of 1 minute. The unseparated particles and molecules formed a Void Peak, which appeared 3 minutes after the carrier flow was started. The exosome containing factions started with fraction #5 (fifth eluted fraction, collected between 5-6 minutes, as shown in Table 2.4) and continued until fraction #10. For each collected AFFF fraction, the concentrations, the size distributions, and its mode and mean were measured by the NTA. The fractionation experiments were repeated three times. 2.3.7 Density Fractionation by Sequential CFFF and AFFF Fractionations One of the obtained CFFF mass fractions was further sub-fractionated by AFFF (Figure 2.3). Specifically, the 1 ml CFFF Fraction #3, which contained the highest concentration of exosomes, from three repeat runs was pooled to form a 3ml sample for the sequential AFFF sub-fractionation. The AFFF was operated at the conditions listed in Table 2.3 using 1.5 ml of the pooled Fraction #3. The downstream miR21 quantification by ddPCR was performed for the AFFF Fraction #5 derived from CFFF Fraction #3. 2.3.8 Isolation and Extraction of Exosomal miRNA miRNA was extracted using the Total Exosome RNA and Protein Isolation Kit (Invitrogen) following the manufacturer’s procedures with a final elution volume of 40 µl. Samples were combined with 200 µl of 2X Denaturing Solution, lysed by vortexing, 36 and then incubated for 5 minutes on ice. 400 µl of Acid-Phenol:Chloroform was added and vortexed for 60 seconds. Samples were centrifuged (5 minutes at 10,000xg) to separate the mixture into aqueous and organic phases, and the upper aqueous phase was transferred to a fresh tube. 133 µl of 100% EtOH was added, vortexed, and transferred onto a column. After centrifugation (30 seconds at 10,000g), the flow-through, which contains the small RNA, was combined with 300µL of 100% EtOH, transferred onto a new column, and centrifuged for 30 seconds at 10,000xg. The flow-through was discarded and the column was placed into a fresh tube and underwent two washing steps. The RNA was eluted in 40 µl RNase-free water and stored at -80 °C for downstream RNA profiling by digital droplet PCR.47 2.3.9 miR21 Analysis by Digital Droplet PCR miR21 content in the original population of the exosomes and its various fractions was quantified by dPCR. TaqMan miRNA assay was used. For miR-21 amplification, Applied Biosystems provided primers and probes of TaqMan assay. Proprietary (assay ID#44279775) represents the actual sequences. A final volume of 25 µl was prepared for dPCR reactions using the following inputs: 12.5 µl TagMan Fast Advanced Master Mix, 1 µl TagMan Advanced miRNA Assay (20X), 0.5 µl RNase free water, and 1 µl Drop Stabilizer (RainDance Technologies). Using the Raindrop Source chip (RainDance Technologies), droplets were generated and then eluted into PCR tubes. For the amplification on a thermocycler, the following settings were applied: 95 °C (10 minutes), 50 cycles of 95°C (15 seconds), 58°C (15 seconds), 60°C, slow ramp speed 0.5°C/seconds (45 seconds), and 95°C (10 minutes). Once the PCR was completed, the 37 collected samples were placed into the Raindrop Source chip to detect miR21 using single fluorescent droplet detection. The RainDrop analyst Software software (RainDance Technologies, Version 2.0) was used to analyze the data by recording a count of positive versus negative droplets.47 2.3.10 Quantification of miR21 per an Exosome The level of miR21 per exosome was calculated as follows. First, the concentration measurements were performed by NTA for the collected mass and/or size fractions. Second, the produced average-concentration (number of exosomes per ml) of each aliquot was multiplied by its volume (~1000µl used for miRNAs extraction) to obtain the number of exosomes that were lysed for miRNAs extraction. Third, a volume of ~10µl aspirated from ~14µl (the total volume of miRNAs extracted from the 1000µllysed-MCF-7 exosomes) was further diluted (1:3). Fourth, about ~10µl of the diluted sample was analyzed by ddPCR to obtain the number of miR21 copies per 10µl of the sample. Finally, the produced number of miR21/10µl was multiplied by 3 and by the total volume of the extracted miRNAs (14µl), and then divided by the number of exosomes. The result of the described steps was the estimate of the number of miR21 copies per exosome. 2.3.11 Calculations of the Mass and Density of MCF-7 Exosomes The mass and density of MCF-7 exosomes were calculated using the measurements and the well-established theory of CFFF separation. The workflow applied to obtain the mass and density of MCF-7 exosomes was as follows. The first step is to 38 obtain the experimental value of the retention ratio, R, as !! R=!  (2.1) ! where to is the void time of unretained particles associated with the void peak and tr is the retention time of retained particle associated with the retained peak. This equation is commonly used to describe the retention ratio of particles inside the CFFF separation channel.68 For each eluted fraction “i", equation 2.1 becomes: R !" = ! !! !,!"  (2.2) where RFi is the retention ratio of fraction “i" and tr,Fi is the retention time of same fraction. Since the void time (to) and the retention times (tr,Fi) are known and tabulated in Table 2.2, the retention ratio of each collected mass fraction (RFi ) can be easily obtained using equation 2.2. The retention ratio, R, can also be expressed as: R= !!"#$ !!!    (2.3) where vzone is the average velocity of the particle zone (the volumetric flow rate of the particle zone divided by the cross-sectional area of the CFFF separation-channel) and <v> is the constant average velocity of the carrier liquid. Substituting for vzone in equation 39 2.3, using the distribution of concentration of particles across the CFFF separationchannel, followed by the analytical integration, the following exact solution of R is obtained68: ! R = 6𝜆 coth( ) − 2𝜆 !!  (2.4) where λ is the retention parameter. Equation 2.4 is the classical retention equation, which relates the experimentally measured retention ratio, R, to the retention parameter λ. The physical meaning of λ is to indicate the degree of zone compression of the particles due to the applied centrifugal-force field. For each eluted mass fraction “i”, equation 2.4 becomes: R !" = 6λ!" coth( Once λ Fi ! !!!" ) − 2λ!"  (2.5) is obtained numerically, the average-mass and average-density of MCF-7 exosomes for a single fraction “i” can be easily calculated, through the subsequentcomputation of the following equations68: 𝑚!,!" = !!   !  !!"  !    !  (2.6a) 𝑚!" = 𝑚!,!"   +     𝑚!"#$  !"##$%,!" (2.6b) 𝑚!"#$  !"##$%,!"       =   ρ!   V!"              (2.6c) ! V!"   =  d!"   !   ! (2.6d) 40 ρ!"   = !!" !!" (2.6e) where  𝑚!,!" is the average effective mass (the mass corrected for buoyancy) of particles of fraction “i” (in kg),  𝑚!" is the average true mass of particles of fraction “i”,  𝑚!"#$  !"##$%,      !" is the average mass of displaced buffer of fraction “i”, 𝑉!" is the average volume of particles of fraction “i” (in m3), 𝑑!" is the average diameter of particles of fraction “i” (in m), ρl is the density of the carrier fluid (water density of 1000 kg/m3 was used), KB is the Boltzmann’s constant= 1.38064852 × 10-23 m2 kg s-2 K-1, T is the absolute temperature (in K), w is the thickness of the channel (in m),  𝐺 is the centrifugal acceleration calculated as 𝐺 = 𝜔! 𝑟  , where ω is the angular rotation frequency equal to 2π×rpm/60, and r is the radius of the rotor (in m). The diameter 𝑑!" in equation 2.6d was obtained from the NTA measurements as the average mode diameter of exosomes in fraction i. Further details and the complete derivation of the above equations are found in the Field-Flow Fractionation Handbook.68 2.4 Results 2.4.1 TEM Characterization of MCF-7 Exosomes Figure 2.4 shows the typical TEM image of MCF-7 exosomes after staining and desiccation. It indicates that their membrane vesicles are in the range between 20 and 50 nm.67 A much smaller geometric size of the exosomes relative to their hydrodynamic diameter has been noted previously by Chernyshev et al.14 41 2.4.2 Mass Fractionation of MCF-7 Exosomes Using CFFF Figure 2.5 gives the averaged results of CFFF fractionation experiments repeated three times (the data for each run is shown in Figures 2.6-2.8). The exosome-containing CFFF fractions were characterized to find the changing concentration (in number of particles per ml) with elution time (Figure 2.5A), the change in the mean and mode diameters (Figure 2.5B), the particle mass (calculated from equation 2.6a and shown in Figure 2.5C), and the density calculated from the particle mass and the volume estimated based on the measurements of the average mode diameters of the exosomes contained in the fraction (Figure 2.5D). The general trend in Figure 2.5 shows that earlier fractions contain more exosomes but they have relatively small mass and density. The mass of the particles increases with the fractionation time, as expected during the centrifugal-field fractionation. The density of the exosomes tends to be higher in the later-eluted fractions, and the size increases with the elution time. The profile of the average density (kg/m3) of MCF-7 exosomes, shown in Figure 2.5D, increases linearly with the elution time, except for a large drop in density observed for the last analyzed fraction (Fraction #7, eluted after 9 minutes of fractionation). The trend has changed with Fraction #7 because of a faster increase in particle volume compared to the rate at the mass increase. 2.4.3 Molecular Characterization of Mass Fractions The miR21 content of mass fractions (fractions 3 through 9 in Table 2.2) were characterized by ddPCR. The results in Figure 2.9 are expressed as a number of miR21 copies per exosome and shown superimposed biophysical properties of the mass 42 fractions. The miR21 abundance (shown in red in Figure 2.9) decreases with elution time. The highest abundance was observed in small light exosomes of the lowest density. 2.4.4 miR21 Abundance vs. Biophysical Properties of Mass Fractions Figure 2.10 re-interprets the data on miR21 abundance given in Figure 2.9 as a function of the concentration, hydrodynamic diameter, mass, and density of the CFFF fractions. In a different way, this figure shows the relative abundance of miR21 copies in high-concentration (Figure 2.10A), small (Figure 2.10B), light (Figure 2.10C), and least dense (Figure 2.10D) fractions. 2.4.5 Molecular Characterization of Size Fractions The same conceptual framework was applied to size fractions of exosomes obtained by AFFF. The analysis of six exosome-containing size fractions (fractions #5 through #10 in Table 2.4) is summarized in Figure 2.11. The average mode and mean hydrodynamic diameters increased linearly with the elution time (minutes), consistent with the expectation that smaller-size particles are less retarded by the asymmetric crosschannel flow inside the AFFF instrument. Figure 2.11B indicates that a higher miR21 abundance in smaller exosomes (mode and mean diameters~ ≤ 80nm) eluted early during the size fractionation process. 2.4.6 Density Fractions by Sequential CFFF-AFFF Fractionation By subjecting CFFF Fraction #3 (eluted between 3-4 minutes) to AFFF subfractionation, several density fractions were obtained. In other words, Fraction #3, which 43 represents a sample with a fixed mass (on average, 0.25 fg) but with heterogeneous sizes and densities, was separated based on the property of size from smaller to larger sizes. Fraction #3 was selected because it has the highest concentration (see Figure 2.5) and the highest abundance of miR21 (see Figure 2.9). Figure 2.12A shows the result of the size sub-fractionation of mass Fraction #3 for three runs (Run 1, Run 2, Run 3). In each run, the collected size sub-fractions were analyzed separately using NTA and dPCR. Thus, different size sub-fractions, which correspond to different densities, were obtained. The figure shows that the average size (mode size in green) increases almost linearly with elution time. This result was expected because it is identical with the separation principle of the AFFF technique. On the other hand, the content miR21/exosome shows the following results: (1) For the three runs (in black, red, and blue), the smallest exosomes (eluted in the first fraction) have the highest abundance of miR21. One explanation behind this result is that, Fraction #3 is a population of exosomes with approximately equal mass. As a result, the enrichment through the size reduction should make the first eluted fraction (with the smallest particles) more enriched in miR21. The general trends shown in Run#1, Run#2, and Run#3 confirm the above explanation where the content of miR21/exosome decreases with elution time. The offset/inconsistency in miR21 shown in Run#3 (in blue) is most likely due to an experimental error during one of the experimental stages. (2) The direct correlation of miR21 with density is expected because nucleic acids have higher density than other cellular macromolecules of exosomes. The above result allows us to calculate the factor of miR21 enrichment after applying CFFF-AFFF fractionation. For the three runs shown in Figure 2.12A, the first 44 eluted fractions have abundant miR21. Consequently, Figure 2.12B shows the factors of miR21 enrichment (red), only, for the first eluted fractions in Run#1, Run#2, and Run#3. The miR21 enrichment factors for Run#1, Run#2, and Run#3 compared to the original MCF-7 exosomes (blue) are 0.25X, 0.6X, and 10X, respectively. This variance in miR21 enrichment between the three runs is attributed into two reasons: (1) the input volumes of the original aliquots in Run#1, Run#2, and Run#3, were 20 µl, 20 µl, and 100 µl, respectively. Accordingly, the highest produced enrichment factor (10X) was associated with the highest input volume (100 µl). That is, more fragments of miR21 are always associated with higher exosome concentration (higher input volume, 100 µl). (2) The population of MCF-7 exosomes in the circulation is heterogeneous in its molecular content of miR21. In general, more experiments are needed to verify the repeatability of the above results and to confirm the expectation that smaller denser exosomes are enriched in miR21 and perhaps other microRNAs. 2.5 Discussion This study has found that the exosomes secreted by the same MCF-7 cell type are heterogeneous in their hydrodynamic size, mass, and the density. The miR21 content was found to be non-uniformly distributed as these biophysical properties vary between the fractions. It was particularly surprising that higher density of the exosomes is not positively correlated with the miR21 abundance. Figures 2.9B and 2.10B indicate that the number of miR21 copies changed with the mass, density, and size of the exosomes. The number of miR21 copies dropped by a factor of almost 10 as the mode of the 45 hydrodynamic size distribution increased from ~60 nm to ~75 nm. The lighter exosomes in the average buoyant mass range between 0.20 and 0.25 fg are more abundant in miR21 than the heavier ones. The density is the derivative properties of the measured hydrodynamic diameter and the theoretically calculated average mass of exosomes in different CFFF fractions. Strong (cubic) dependence of the density on the diameter of the particles sensitized the results on the density-dependence of the miR21 abundance to the sizing errors. It is also possible that the size of the vesicles containing miR21 is substantially smaller than the measured hydrodynamic size used in density calculations. The compared average mass of exosomes inferred from the CFFF results with the quartz crystal microbalance (QCM) measurements shows that the average mass of MCF-7 exosomes determined by CFFF is 0.97± 0.21 fg, which is found to be larger than the average mass determined by QCM (0.40 ± 0.074 fg) (unpublished results).67 Our results indicate that certain exosome factions are relatively enriched in a known cancer biomarkers, miR21.69,70 This observation may be important in detecting the presence of miR21 in complex heterogeneous samples. Recently, many studies have indicated the importance of miRNAs as biologicallyactive species involved in cell-to-cell communication, gene expression, cancer initiation/progression, and as novel biomarkers in diagnostics.71–74 The potential relationship between exosomes and miRNAs, as seen through different studies, highlighted the important role of circulating exosome miRNAs for cancer diagnosis and monitoring. For example, Ling Li et al.,75 reported that under stimulation of hypoxic microenvironment, tumor-hypoxic cells release miR21–rich exosomes delivered to normoxic cells. This cell-to-cell communication using miR21–rich exosomes can 46 promote pro-metastatic behaviors and eventually increase aggressiveness of solid tumors. On the other hand, decreasing levels of miR-21 in exosomes can significantly reduce invasiveness of tumor cells. Hannafon et al.,76 demonstrated that human breast cancer exosomes are enriched with certain microRNA species, such as miR21. Overall, new strategies are needed in detecting the presence of onco-microRNAs in complex biological samples. To this end, the sample fractionation by biophysical properties correlated with higher abundance of target microRNAs appears to be a promising strategy. 47 A B Figure 2.1. Mass fractionation of MCF-7 exosomes: (A) The original aliquot of MCF-7 exosomes (volume of 100 µl) was fractionated by the CFFF into different mass fractions (1 ml-volume each). (B) The schematic illustration of the mass fractionation process. The mass of an exosome increases with the increase in the vesicle size and the luminal cargo (~). 48 Table 2.1. Run conditions during CFFF experiments. Item Input volume Injection delay Relaxation time Starting time to collect mass fractions Channel flow rate Collection duration for each mass fraction Centrifugal speed Detector wavelength Number of collected samples Total time for the experiment Value Description/ Comments 100 µl of the original MCF7 (~1012 particles/ml) 30 seconds It is the time needed for the sample to pass the CFFF loop to the separation channel. 10 minutes It is the time needed by the exosome particles to settle down at the bottom of the CFFF channel under the effect of centrifugal forces before the carrier fluid flow is initiated. Right after the relaxation time (10 minutes) 1 ml/ minute 1 minute 4900 rpm 219 nm 9-10 samples (1ml each) 20 minutes 49 Table 2.2. Summary of the collected CFFF mass fractions showing the fraction number, collection period, and the corresponding retention time used in mass calculation. Fraction # Collection period 1st 1-2 minutes Retention time (midpoint) 1.5 2nd 2-3 minutes 2.5 3rd 3-4 minutes 3.5 4th 4-5 minutes 4.5 5th 5-6 minutes 5.5 6th 6-7 minutes 6.5 7th 7-8 minutes 7.5 8th 8-9 minutes 8.5 9th 9-10 minutes 9.5 10th 10-11 minutes 10.5 50 A B Figure 2.2. Size fractionation of MCF-7 exosomes: (A) The original aliquot of MCF-7 exosomes (volume of 100 µl) was fractionated by the AFFF into different size fractions (2 ml-volume each). (B) The schematic illustrates that the size fractionation of the original aliquot of MCF-7 exosomes contains heterogeneous size fractions. Note that the size separation by AFFF is not affected by the luminal cargo (~) and the fractionation is based on the hydrodynamic size of the exosomes (vesicle size is shown). 51 Table 2.3. Run conditions during AFFF experiments. Item Input volume Focusing time Value Description/ Comments 100 µl of the original MCF7 (~1012 particles/ml) 3 minutes It is the time needed by the exosome particles to settle down at the bottom of the AFFF separation channel before the carrier fluid is initiated. Right after the focusing time (3 minutes) 2 ml/ minute 0.5 ml/ minute 1 minute Starting time to collect size fractions Channel flow rate Cross flow rate Collection duration for each size fraction Detector wavelength 219nm UVDAD (diode array detection), 280nm UVDAD, and 254nm UVDAD Number of collected 9-10 samples (2 ml each) samples Total time for the 20 minutes experiment 52 Table 2.4. Summary of the collected AFFF size fractions showing the fraction number, the collection period, and the corresponding retention time. Fraction # Collection period 1st 2nd 1-2 minutes 2-3 minutes Retention time (midpoint) 1.5 2.5 3rd 3-4 minutes 3.5 4th 4-5 minutes 4.5 5th 5-6 minutes 5.5 6th 6-7 minutes 6.5 7th 7-8 minutes 7.5 8th 8-9 minutes 8.5 9th 9-10 minutes 9.5 10th 10-11 minutes 10.5 53 A B Figure 2.3. Sequential fractionation of MCF-7 exosomes: (A) Sequential fractionation by CFFF and AFFF. (B) The selected mass fraction (Fraction #3 in our case) eluted from CFFF was further sub-fractionated by AFFF to produce different size sub-fractions of particles with approximately equal mass but different density. (~) indicates luminal cargo. 54 Figure 2.4. TEM image of negatively stained MCF-7 exosomes. Adapted from67 55 120# 1.80E+10% Mean#diameter#(nm)# A 1.60E+10% 110# B b Mode#diameter#(nm)# 100# 1.20E+10% Diameter((nm)( Concentra)on*(#/mll)* 1.40E+10% 1.00E+10% 8.00E+09% 6.00E+09% 90# 80# 70# 4.00E+09% 60# 2.00E+09% 50# 0.00E+00% 3% 4% 5% 6% 7% 8% 9% 3# 10% 4# 5# 7# 8# 9# 10# 8# 9# 10# 1040# 0.70# C c 0.60# D 1035# 0.50# 1030# 0.40# Density((kg/m3)( Mass$(femto+gram)$ 6# Elu/on(/me((nm)( Elu)on*)me*(min)* 0.30# 0.20# 1025# 1020# 1015# 0.10# 1010# 0.00# 3# 4# 5# 6# 7# Elu2on$2me$(min)$ 8# 9# 10# 3# 4# 5# 6# 7# Elu3on(3me((min)( Figure 2.5. Biophysical characterization of CFFF mass fractions. (A) The average concentration, expressed as a number of particles/ml decreased gradually with the elution time. (B) The size, expressed as a hydrodynamic diameter, increased with the elution time. (C) The average mass of particles was higher in the later fractions. (D) The average density of the eluted particles was obtained using the average mass and the corresponding diameter of each fraction. 56 2.50E+10% Run-1 Concentra)on*(#/mll)* 2.00E+10% 1.50E+10% 1.00E+10% 5.00E+09% 0.00E+00% 3% 4% 5% 6% 7% 8% 9% 10% 8% 9% 10% 8% 9% 10% Elu)on*)me*(min)* 9.00E+09% Run-2 8.00E+09% Concentra)on*(#/mll)* 7.00E+09% 6.00E+09% 5.00E+09% 4.00E+09% 3.00E+09% 2.00E+09% 1.00E+09% 0.00E+00% 3% 4% 5% 6% 7% Elu)on*)me*(min)* 1.20E+10% Run-3 Concentra)on*(#/mll)* 1.00E+10% 8.00E+09% 6.00E+09% 4.00E+09% 2.00E+09% 0.00E+00% 3% 4% 5% 6% 7% Elu)on*)me*(min)* Figure 2.6. Biophysical characterization of CFFF mass fractions. The three runs of concentration, expressed as a number of particles/ml decreased gradually with the elution time. 57 110# Mean#diameter#(nm)# Run-1 Mode#diameter#(nm)# 100# Diameter((nm)( 90# 80# 70# 60# 50# 3# 4# 5# 6# 7# 8# 9# 10# 8# 9# 10# 8# 9# 10# Elu/on(/me((nm)( 120# Mean#diameter#(nm)# 110# Run-2 Mode#diameter#(nm)# Diameter((nm)( 100# 90# 80# 70# 60# 50# 3# 4# 5# 6# 7# Elu/on(/me((nm)( 130# Mean#diameter#(nm)# 120# Run-3 Mode#diameter#(nm)# Diameter((nm)( 110# 100# 90# 80# 70# 60# 50# 3# 4# 5# 6# 7# Elu/on(/me((nm)( Figure 2.7. Biophysical characterization of CFFF mass fractions. The three runs of sizes, expressed as a hydrodynamic diameter, generally increased with the elution time. 58 1050# Run-1 1045# 1040# Density((kg/m3)( 1035# 1030# 1025# 1020# 1015# 1010# 3# 4# 5# 6# 7# 8# 9# 10# 8$ 9$ 10$ 8$ 9$ 10$ Elu3on(3me((min)( 1028$ Run-2 1026$ 1024$ Density((kg/m3)( 1022$ 1020$ 1018$ 1016$ 1014$ 1012$ 3$ 4$ 5$ 6$ 7$ Elu3on(3me((min)( 1026$ Run-3 1024$ Density((kg/m3)( 1022$ 1020$ 1018$ 1016$ 1014$ 1012$ 3$ 4$ 5$ 6$ 7$ Elu3on(3me((min)( Figure 2.8. Biophysical characterization of CFFF mass fractions. The three runs of density of the eluted particles generally increased with the elution time, and then decreased after 9 minutes of fractionation. 59 3.50E+10% 120% 2.50E'04% A 2.50E'04% 110% 3.00E+10% Mode%diameter%(nm)% 2.00E'04% miR21/exosome% 1.50E+10% 1.00E'04% Diameter+(nm)+ 1.50E'04% 2.00E+10% 1.50E'04% 90% 80% 1.00E'04% 70% 1.00E+10% 5.00E'05% 5.00E'05% 60% 5.00E+09% 0.00E+00% 50% 0.00E+00% 4% 5% 6% 7% 8% 9% 0.00E+00% 3% 10% 4% 5% 10% 2.50E'04% 1035% 2.00E'04% 0.40% 0.30% 1.00E'04% 0.20% 5.00E'05% 0.10% 0.00% 0.00E+00% 9% 10% 2.00E'04% 1030% Density+(kg/m3)+ 1.50E'04% miR21/exosome+ 0.50% 6% 7% 8% Elu8on+8me+(min)+ 9% D 0.60% 5% 8% 1040% 2.50E'04% C 4% 7% Elu6on+6me+(nm)+ Elu2on+2me+(min)+ 0.70% 6% 1.50E'04% 1025% 1.00E'04% 1020% miR21/exosome+ 3% Mass+(femto1gram)+ miR21/exosome+ 100% 2.50E+10% miR21/exosome+ Concentra2on+(#/mll)+ 2.00E'04% 3% B Mean%diameter%(nm)% 5.00E'05% 1015% 1010% 0.00E+00% 3% 4% 5% 6% 7% 8% 9% 10% Elu8on+8me+(min)+ Figure 2.9. Molecular characterization of miR21 abundance (expressed as a number of miR21 copies per exosome) in different CFFF fractions is superimposed on the results shown in Figure 2.5 and scaled at right in red. 60 2.50E'04% 2.50E'04% A B 2.00E'04% miR21/exosome+ miR21/exosome+ 2.00E'04% 1.50E'04% 1.00E'04% 1.50E'04% 1.00E'04% 5.00E'05% 5.00E'05% 0.00E+00% 0.00E+00% 0.00E+00% 5.00E+09% 1.00E+10% 75% 1.50E+10% 77% 79% 81% 83% 2.50E'04% 89% 91% 93% 2.50E'04% C D 2.00E'04% miiR21/exosome+ 2.00E'04% miR21/exosome+ 87% Mode+diameter+(nm)+ Concentra2on+(#/ml)+ 1.50E'04% 1.00E'04% 5.00E'05% 0.00E+00% 0.20% 85% 1.50E'04% 1.00E'04% 5.00E'05% 0.25% 0.30% 0.35% Mass+(femto1gram)+ 0.40% 0.45% 0.00E+00% 1010% 1015% 1020% 1025% 1030% 1035% Density+(kg/m3)+ Figure 2.10. miR21 abundance as a function of biophysical properties of CFFF fractions. (A) The abundance is higher in fractions with higher exosome concentration. (B) Smaller exosomes are more enriched in miR21. (C) Exosomes with smaller mass are more abundant in miR21. (D) The average density is inversely correlated with miR21 abundance. 61 120% Mean#diameter#(nm)# 110# Mode#diameter#(nm)# 110% 100# 90# 80# B Mode%diameter%(nm)% 2.00E'04% miR21/exosome% 100% Diameter+(nm)+ Diameter((nm)( 2.50E'04% Mean%diameter%(nm)% A 1.50E'04% 90% 80% 70# 70% 60# 60% 1.00E'04% miR21/exosome+ 120# 5.00E'05% 50% 50# 5# 6# 7# 8# 9# 10# 0.00E+00% 5% 11# 6% 7% 9% 10% 11% Elu6on+6me+(min)+ Elu/on(/me((min)( 1.60E'04% 1.60E'04% C 1.40E'04% D 1.40E'04% 1.20E'04% 1.20E'04% miR21/exosome+ miR21/exosome+ 8% 1.00E'04% 8.00E'05% 6.00E'05% 1.00E'04% 8.00E'05% 6.00E'05% 4.00E'05% 4.00E'05% 2.00E'05% 2.00E'05% 0.00E+00% 0.00E+00% 60% 65% 70% 75% 80% 85% 90% Mode+diameter+(nm)+ 95% 100% 105% 110% 60% 65% 70% 75% 80% 85% 90% 95% 100% 105% 110% Mean+diameter+(nm)+ Figure 2.11. Biophysical-Molecular characterization of the AFFF size fractions and the abundance of miR21 (in red). (A) The average hydrodynamic diameter (expressed as the mode and the mean of the size distribution) increased with the elution time. (B) Smaller exosomes are more abundant in miR21 (number of miR21 copies per exosome is shown as a red line). (C) The observed dependence of miR21 as a function of the mode diameter of the exosome size distribution in different AFFF fractions. (D) The observed dependence of miR21 as a function of the mean diameter. 62 A 110.0% 100.0% Mode%size% miR21/exo%(Run%3)% miR21/exo%(Run%1)% miR21/exo%(Run%2)% A5FFF+of+C5FFF+frac:on+3+ 1.50E'04% miR21/exosome+ 90.0% Diameter+(nm)+ 2.00E'04% 80.0% 1.00E'04% 70.0% 60.0% 5.00E'05% 50.0% 40.0% 0.00E+00% 4% 5% 6% 7% 8% Time+(min)+ 9% 10% 11% 12% B 1.40E'04% 1.20E'04% Original%miR21/exosome% Enrichment%of%FracaEon#3% 10X& 0.6X& 3.7X& miR21/exosome& 1.00E'04% 8.00E'05% 0.25X& 6.00E'05% 4.00E'05% 2.00E'05% 0.00E+00% Run#1% Run#2% Run#3% Avg%of%3%runs% Figure 2.12. Results of miR-21 enrichment after the sequential fractionation: (A) The trend of miR-21 for three runs (black, red, and blue) for applying CFFF-AFFF fractionation. In three runs, the first eluted fraction between 4 and 6 minutes with the smallest exosomes (size in green) has the highest content of miR21. (B) Using the content of miR21/exosome of the first eluted fraction of three runs shown in A, the figure shows enrichment factors of miR21 (red) compared to each control (blue). The input volumes of the original aliquots used for Run#1, Run#2, and Run#3 were 20 µl, 20 µl, and 100 µl, respectively. 63 CHAPTER 3 COMPARISON OF THERMAL STABILITY OF MCF-7 AND MCF-10A EXOSOMES 3.1 Abstract To the best of our knowledge, this is the first study that systematically compared the thermal stability of tumor and normal-like exosomes. The thermal stability of exosomes was investigated by measuring the decrease in their concentration after exposure to different temperatures and incubation times. The results were obtained for the exosomes isolated from the growth medium of MCF-7 ER+ breast cancer cells and normal-like MCF-10A breast cells. The difference in the thermal stability of MCF-7 and MCF-10A was evaluated. A statistically significant difference (P-value < 0.001) was found to occur with a higher exposure temperature (57°C) and a longer incubation time (24 hours). At these conditions, MCF-7 exosomes were found to be more stable than MCF-10A exosomes. 3.2 Introduction A few studies have discussed the effect of temperature on exosome stability as a function of incubation time, and that indicates the lack of research in this particular area. For example, in one study,13 the comparison of plasma and urine exosomes indicates that 64 after 5 days of incubation at room temperature and 4 °C, the size of urine and plasmaderived exosomes changed, but the reduction in the concentration was considerable (see Figure 3.1). The concentration for plasma exosomes has dropped faster, suggesting their lower stability at the considered experimental conditions. This study suggests that the reduction/loss in exosome concentration “most likely” was due to the exosomes degradation into small fragments (under 25 nm). This conclusion was based on the NTA limitation for minimum size detection (30nm). A different group of authors48 presented a study to measure the size and integrity of three different human exosomes, where each of them was stored at three different temperatures (37 °C, 4°C, and −20 °C) with different periods of time. The main conclusion of the study shows that the exosome size depends strongly on both the storage temperature and storage time as shown in Figure 3.2. The authors reattributed the size reduction (shown in Figure 3.2) to two reasons: either a structural change or degradation of exosomes. Generally, this study indicates that the stability/integrity of exosomes decreases as a function of temperature and time. Finally, another study77 investigated the thermal stability of urine-derived particles including exosomes using Nanoparticle tracking analysis (NTA). One of the results presented by this study is shown in Figure 3.3, where the concentration of urine particles (between 20-100 nm in diameter) was analyzed at different temperatures and incubation times. The figure shows clearly that the concentration of urine particles including exosomes depends strongly on the temperature and the incubation time. In our presented study, the difference in thermal stability of tumor and normallike exosomes was investigated for extracellular vesicles secreted by normal-like human 65 epithelial breast cells (MCF-10A line) and metastatic estrogen- and progesterone-positive adenocarcinoma human breast cells (MCF-7 line). We tested our hypothesis, which states that it is possible to find the temperature and incubation time at which thermal stability of normal and tumor exosomes is statistically different. The lack of significant difference between normal and tumor exosomes is our null hypothesis. The basis of our hypothesis is formed by two considerations. First, tumor cells are metabolically more active than normal cells.78 Accordingly, the higher metabolic activity of tumor cells with the faster proliferation leads to a higher release of energy and a higher temperature than normal cells. This suggested that tumor exosomes, with their content derived from tumor cells, will maintain the tolerance to elevated temperatures. Second, two different studies provided indirect confirmation of by observing that the denaturing of plasma of cancer patients, which inevitably contains tumor exosomes, occurs at higher temperatures than normal plasma.79,80 In this study, the thermal stability of exosomes secreted by normal-like MCF-10A and ER+ MCF-7 breast tumor cells was compared. The combined impact of two factors – the incubation time and temperature – was investigated. The impact of four temperature levels (RT, 37, 47, and 57° C) and the exposure over 30 minutes, 4, 8, and 24 hours were considered. The thermal stability was assessed by characterizing the reduction in the particle concentration measured by the NTA. 66 3.3 Materials and Methods 3.3.1 Cell Culture and Isolation Prior to use, both MCF-7 ER+ breast cancer and normal-like MCF-10A breast cells (American Type Culture Collection (ATCC), Manassas, VA, USA) were stored in liquid nitrogen. The cells were grown following standard protocol.47 The exosomes were isolated from the growth medium using the precipitation method (ExoQuick-TC, SBI) Briefly, cell media with a volume of 30 mL was centrifuged at 3000×g for 15 minutes in order to remove/pellet all cells. The supernatant (~25mL) was transferred to a new vial containing 6 mL of ExoQuick-TC precipitation solution, mixed, and incubated overnight at 4°C. After the incubation, each tube was centrifuged at 1,500×g for 30 minutes to pellet exosomes. The supernatant was discarded. Following the isolation, exosomes were re-suspend in either 450 µL 2 mM ammonium acetate or 450 µL phosphate buffered saline (PBS). Using a commercial antibody assay (ExoCheck, SBI), the exosome enrichment in the sample was confirmed by positive staining of several known biomarkers. The overall sample was divided into 100 uL aliquots and stored at -80°C for a subsequent use.14,47 3.3.2 Exosomes Concentration and Size Measurements Using NTA Within 24 hours of thawing an MCF-7 and MCF-10A exosome aliquot, the measurements of exosome concentration and size distribution were performed using nanoparticle tracking analysis (NTA) (Nanosight model LM14 with 50 mW red 635 nm wavelength laser). A high-sensitivity sCMOS camera (OrcaFlash2.8, Hamamatsu C11440) was used to capture the light scattered by exosomes. The results were analyzed 67 using the software provided by the manufacturer (Nanosight Version 3.2). The buffer used for exosome dilution was 1X PBS (Phosphate Buffered Saline, 1X Solution, Fisher Scientific). During the NTA analysis, the viscosity was assumed to be equal to water viscosity. The room temperature remained around 20°C throughout the tracking of experiments. At such conditions, the viscosity of water remains nearly constant at 1 cP. The NTA analysis was performed after serial (1:500 and then 1:335) dilution with the PBS in order to have approximately 70-100 particles in each video frame (this corresponds to an approximate concentration of 1.4-2x109 particles/ml, 70-100 particles/frame). The NTA analysis of each diluted exosome sample was performed within 5 minutes of the initial dilution. Each diluted sample was injected into the test cell using a 1 mL sterile syringe (Becton and Dickinson, Ref. No. 309659). A set of five videos, each 60-seconds long, was recorded for each sample and the results were averaged to obtain exosome concentration, size distribution, and its properties (mode, mean, and the standard deviation). The NTA software settings for the analysis were as follows: camera level=16, frames per second=25, shutter speed=30 ms, slider gain=500, detection threshold for video processing=4, and numbers of frames in each video=1500. Maximum jump mode, Blur, Minimum track length were set to Auto. For NTA analysis, exosomes were assumed to have a spherical shape. During the NTA analysis, the sample was flown using a syringe pump, which allowed us to obtain the results based on a larger number of analyzed particles. The required volume of the sample for the NTA analysis was at least 800 ul. After loading 300 ul into the NTA cell using a 1 mL sterile syringe, 100 ul of the sample was advanced 68 by the syringe pump at rate of 1000 units (each unit = 0.003ul/s) to flush the test cell. The data was not acquired during the flush. Then, the syringe pump was adjusted to provide the flow rate of 80 units (0.27 ul/s) at which the data was collected. Once the NTA data acquisition was completed, the sample was aspirated from the test cell, and the cell was rinsed twice using 3ml of 1X PBS to avoid sample crosscontamination during the next run. At the end of the experiment, the test cell was thoroughly rinsed with DI water and, and wiping with 70% EtOH using KimWipes. The described procedure was the same for all samples.14,67 3.3.3 Conventional Heating of Exosomes Using Digital Dry Block Heater A thermal metal dry bath (ThermoQ, Model: CHB-T1-A, Hangzhou Bioer Technology Co., Ltd.) with temperature range of 0°C to 100°C was used to heat the MCF-7 and MCF-10A exosome samples. Temperature was programmable and was controlled by using the ThermoQ program (Version 1.0) with digital recording. Temperature uniformity, temperature control accuracy, and temperature fluctuation was ≤ 0.5°C. The sample-displayed temperature by the program was verified by using an external temperature probe connected to a stand-alone thermometer. The taper holes of the heating block, for a tube size 1.5ml, were cleaned regularly to guarantee the complete contact between the taper hole and tested tubes for good thermal conductivity. 3.3.4 Design of Experiments The design of exosome thermal experiments was based on a two-factorial design. As shown in Figure 3.4, there are two quantitative factors for the thermal process of 69 exosomes: temperature and incubation time. Exosome concentration is the main measure of the thermal effect on exosomes. The minimum number of testing points to study a two-factorial design is four testing points (four runs). For data reproducibility, each experiment is replicated three times; hence there are 12 runs.81 However, for the purpose of having a precise thermal experiment, we expanded our testing points of the two-factorial design into 14 testing points, as shown in Figure 3.5. These testing points were designed to cover all treatment combinations by having the points of corners, points of in-between, and one point in the outer range including the control. This decision of having this combination of data points was important to decide the best experimental parameter needed for a successful hypothesis testing. Based on this specific two-factorial design shown in Figure 3.5, concentrations of MCF-7 and MCF-10A exosomes were measured as a function of four different temperatures (RT, 37°C, 47°C, and 57°C) and four incubation times (30 minutes, 4 hours, 8 hours, and 24 hours). In total, we conducted 1080 runs to investigate the 14 testing points, including three experimental repetitions needed for data reproducibility. 3.3.5 Sampling According to our two-factorial design, Table 3.1 shows the sample incubation conditions (14 testing points) used to study the thermal stability of exosomes (MCF-7 and MCF-10A). The protocol of sampling is shown in Tables 3.2A, 3.2B, 3.2C, and 3.2D, where the 14 testing points are separated into 4 experimental sets (four tables) with a control (in each set) for normalization. 70 Prior to analysis, the cleanliness of the used buffer (1X PBS, Fisher Scientific) and tubes (Natural, 1.5ml-Microcentrifuge Tubes, Fisherbrand™ Premium, Cat. No. 05408-129) was verified by NTA. To assure cleanliness, the maximum amount of particles should not exceed 10 particles/frame. Within 24 hours of analysis, the original aliquot of exosomes (either MCF-7 or MCF-10A) stored at -80°C was thawed at 4°C. Then, exosomes were diluted in a 10ml1xPBS stock (using sterile 15ml-centrifuge tubes, Fisherbrand, Cat. No. 07-200-886). The tube was gently mixed by pipetting about 10 times to assure good mixing and solution homogeneity. The dilution factor of tumor (MCF-7) and normal-like exosomes (MCF-10A) were 1:500 and 1:335, respectively. A volume of 100µl of the original aliquots of MCF-7 and MCF-10A exosomes was enough for three repetitions. Then, the 10ml-stock of the diluted exosomes was distributed into “ten” 1.5ml centrifuge vials each containing 900µl. The first run (vial#1), as shown in Tables 3.2A-D, was analyzed by NTA within 5 minutes of the initial dilution. After the heating process, each tube was gently mixed right before the NTA analysis began. The protocol of sampling used for the randomized experiment is shown Table 3.3. This randomized thermal experiment involves 15 aliquots: seven aliquots for MCF-7 exosomes, seven aliquots for MCF-10A exosomes, and one 50/50 mixed sample (1:1 ratio in volume and concentration). The 15 aliquots were labeled from 1 to 15, randomized, and tested as blind samples at 57°C for 24 hours (57°C/24 hours). The controls (for MCF-7 and MCF-10A exosomes) were incubated at room temperature for 24 hours (RT/24 hours). 71 3.3.6 Hypothesis Testing For the hypothesis testing in our study, there are three possible scenarios: (1) The positive hypothesis: tumor (MCF-7) exosomes, significantly, are more thermally stable than normal-like (MCF-10A) exosomes; (2) The negative hypothesis: normal-like (MCF10A) exosomes, significantly, are more thermally stable than tumor (MCF-7) exosomes; (3) The null hypothesis: the thermal stability of both exosomes (MCF-7 and MCF-10A) is identical and there is no significant difference between them. Among these three scenarios, we selected the positive hypothesis for two reasons: First, since tumor cells tend to function at high temperature, we expect that exosomes released from tumor cells could have higher thermal stabilities than the ones released from normal cells; Second, cancer plasma, as reported by different studies,79,80 start to denature at higher temperatures than normal plasma. As cancer plasma contains tumor exosomes, it could interact with other surrounding content of plasma like proteins. Such a process of interaction/modification is believed to shift cancer plasma towards higher denaturation temperatures (higher thermal stability). The results revealed from the applied two-factorial design (see the section of the results), demonstrated that if tumor (MCF-7) and normal-like (MCF-10A) exosomes are treated at 37°C/24 hours, 47°C/24 hours, and 57°C/24 hours, the difference in their thermal stability becomes statistically significant (P-value =0.02). Accordingly, we decided to re-validate this significant result by testing exosomes at the best experimental parameter (57°C/24 hours). Hence, for this hypothesis testing, the null hypothesis, states that: statistically, there is no significant difference between the concentrations of tumor (MCF-7) and normal-like (MCF-10A) exosomes if they are treated at 57°C/24 hours. On 72 the other hand, the alternative hypothesis states that: statistically, there is a significant difference between the concentrations of tumor (MCF-7) and normal-like (MCF-10A) exosomes if they are treated at 57°C/24 hours. 3.4 Results 3.4.1 Part-1 3.4.1.1 Thermal Stability of MCF-7 and MCF-10A Exosomes Our results on the thermal stability of exosomes (MCF-7 and MCF-10A) were obtained according to experimentations described in the two-factorial design. Figure 3.6 discloses the overall trend of the thermal stability of MCF-7 (red) and MCF-10A (blue) exosomes. The thermal stability of exosomes is determined by the reduction in exosome concentration as a function of four different temperatures (RT, 37°C, 47°C, and 57°C) and four incubation times (30 minutes, 4 hours, 8 hours, and 24 hours). The figure shows four segments of time (0.5 hour, 4 hours, 8 hours, and 24 hours). Each segment includes three temperatures (RT, 37°C, 47°C), excluding the last segment (at 24 hours) where it has the additional temperature of 57°C, according to the two-factorial design. The general trend of the figure shows that the normalized concentration of both exosomes (MCF-7 and MCF-10A) decreases gradually as a function of temperature and incubation time. The thermal behaviors of exosomes (MCF-7 and MCF-10A) presented in Figure 3.6 are described as follows: The first segment (at half an hour incubation time, 0.5 hours) shows that the percentages of concentration reduction (degradation levels) of MCF-7 exosomes compared to control (RT-0 hour) are 2% (at RT), 11% (at 37°C), and 15% (at 47°C). For 73 MCF-10A exosomes, the percentages of concentration reduction compared to control are 1% (at RT), 9% (at 37°C), and 11% (at 47°C). Statistically, the result of exosome thermal stability (for the two samples: MCF-7 and MCF-10A at 0.5 hour) is insignificant, Pvalue =0.65. The second segment (at 4 hours incubation time) shows that the percentages of concentration reduction of MCF-7 exosomes compared to control (RT-0 hour) are 13% (at RT), 23% (at 37°C), and 26% (at 47°C). For MCF-10A exosomes, the percentages of concentration reduction compared to control are 8% (at RT), 10% (at 37°C), and 14% (at 47°C). Statistically, the result of exosome thermal stability (for the two samples: MCF-7 and MCF-10A at 4 hours) is insignificant, P-value =0.08. The third segment (at eight hours incubation time) shows that the percentages of concentration reduction of MCF-7 exosomes compared to control (RT-0 hour) are 28% (at RT), 31% (at 37°C), and 36% (at 47°C). For MCF-10A exosomes, the percentages of concentration reduction compared to control are 22% (at RT), 24% (at 37°C), and 34% (at 47°C). Statistically, the result of exosome thermal stability (for the two samples: MCF-7 and MCF-10A at 8 hours) is insignificant, P-value =0.31. The summary from these three time segments (0.5 hours, 4 hours, and 8 hours) indicates the following: As long as the incubation time is below 8 hours, the thermal stability of MCF-10A exosomes is higher (by 1%-13%) than the thermal stability MCF-7 exosomes, but with insignificant P-values: 0.65, 0.08, and 0.31, respectively. The fourth segment (at 24 hours incubation time) shows that the percentages of concentration reduction of MCF-7 exosomes compared to control (RT-0 hour) are 34% (at RT), 35% (at 37°C), 42% (at 47°C), and 40% (at 57°C). For MCF-10A exosomes, the 74 percentages of concentration reduction compared to control are 36% (at RT), 50% (at 37°C), 58% (at 47°C), and 67% (at 57° C). It can be noticed clearly that the thermal behavior of exosomes (at this fourth segment of 24 hours) starts becoming more significant with several important observations: First, after 24 hours of incubation at 37°C, 47°C, and 57°C, the difference between the normalized concentration of MCF-7 and MCF-10A exosomes increases significantly from 15% to 27%. Second, after 24 hours of incubation, the domination at higher thermal stability of normal-like (MCF-10A) exosomes significantly changed to favor tumor (MCF-7) exosomes. Third, the result of exosome thermal stability for two samples (MCF-7 and MCF-10A) obtained at 37°C, 47°C, and 57°C is statistically significant, P-value= 0.02. 3.4.1.2 The Thermal Profiles of Exosomes and Data Analysis Figure 3.7A shows the difference in concentration measurements between tumor (MCF-7) and normal-like (MCF-10A) exosomes according to the 14 testing points in the two-factorial design (as shown in Figure 3.5). This result indicates clearly that the difference in thermal stability between exosomes (MCF-7 and MCF-10A) increases significantly as the temperature goes from 37°C to 57°C at the highest incubation time, 24 hours. Figure 3.7B re-interprets the data shown in Figure 3.7A. In a different way, Figure 3.7B shows the values of normalized concentration of MCF-7 (red) and MCF-10A (blue) exosomes, through which the differences in Figure 3.7A were obtained. The chart, which displays three dimensions of data: x=temperature, y=incubation time, z=bubble size (normalized concentration of exosomes), indicates that below 8 hours of incubation 75 and between 20°C and 47°C, the differences between bubble sizes of tumor and normallike exosomes is minor (i.e., the bubbles almost have the same size). However, the variation between the bubble sizes compared to the controls (located at x=0, y=0 and with a size of 100) is significant after 24 hours of incubation and between 37°C and 57°C. Because of this significant result (P-value =0.02) produced by treating the two groups (MCF-7 and MCF-10A) at 37°C/24 hours, 47°C/24 hours, and 57°C/24 hours, we wanted to re-validate this result using a randomized-blind experiment (with a larger sample size, 14 samples instead of three samples). The best parameter condition applied for this verification experiment was determined after performing our data analysis using Figure 3.7A. In this data analysis, a similar example from the textbook of “Design and Analysis of Experiments” (by D.C. Montgomery, 6th edition)81 was applied for the following goals: (1) determining the primary factor of interest in our two-factorial design: either temperature or incubation time; (2) determining if there is a significant interaction between the two experimental factors (temperature and incubation time); (3) determining the best levels of temperature and incubation time that could produce a statistically significant result needed for the verification experiment. Therefore, the following calculations were performed using the results shown in Figure 3.7A: 1- Determining the primary factor of interest: In any factorial design, the effect of a factor means the process of representing the factor response due to a change in the factor level.81 Measuring the effect of a factor (in our case temperature is factor A and incubation-time is factor B) can be obtained by the difference between its average-responses produced at the low and high levels.81 Numerically, this can be obtained as follows (using Figure 3.7A): 76 𝐴 =   !!!" 𝐵 =   !!!" ! ! −   !!! −   !!! ! ! = 8.5  𝑢𝑛𝑖𝑡𝑠 (3.1) = 6.5  𝑢𝑛𝑖𝑡𝑠 (3.2) The procured numbers in equations 3.1 and 3.2 mean that if the temperature (factor A) is increased from the low to the high level, it will produce a responseincrease with an average of 8.5 units. However, if the incubation time (factor B) is increased from the low to the high level, it will produce a response-increase with an average of 6.5 units.81 Thus, increasing A from the low level to the high level will increase the reduction in exosome concentration more than that of factor B. Accordingly, temperature (factor A) is our primary factor of interest (i.e., the leading factor in our two-factorial design). 2- Determining the presence of interaction between the excremental factors: The interaction effect between the factors of the experiment occurs if the responsedifference of the factor A (the temperature factor) is not the same at all levels of factor B (the incubation time).81 Numerically, this can be obtained as follows (using Figure 3.7A): At the low level of factor B (incubation time), the A effect is: 𝐴 =  4 −  1 = 3  𝑢𝑛𝑖𝑡𝑠 (3.3) At the high level of factor B (incubation time), the A effect is: 𝐴 =  16 −  2 = 14  𝑢𝑛𝑖𝑡𝑠 (3.4) 77 Calculating the magnitude of this interaction effect between temperature and incubation time (AB), is obtained by the average difference between the two effects of factor A (shown in equations 3.3 and 3.4). Numerically, this can be obtained as follows: 𝐴𝐵 =  14 −  3 = 11  𝑢𝑛𝑖𝑡𝑠 (3.5) Since the magnitude of the interaction effect is significant (AB=11 units) relative to the effects of factor A (=8.5 units) and factor B (=6.5 units), then there is a strong interaction effect between our factors of the experiment. That is, the effect of factor A (temperature) clearly depends on the selected level for factor B (incubation time). 3. Determining levels producing significant results: As stated by D.C. Montgomery,81 “when there is a significant interaction between factors A and B, then the levels of factor A must be examined, while keeping the levels of factor B fixed.” This step is important in order to generate a conclusion about the effect of factor A on the level being measured. Based on the above data analysis, we drew the following conclusions: (1) temperature is the leading factor of interest; (2) highest temperature (47°C) and highest incubation time (24 hours) lead to the highest reduction in exosome concentration; (3) increasing the level of the leading factor from 47°C to 57°C, while keeping the incubation time fixed at the highest level (24 hours) has almost doubled the yield (i.e., the reduction in exosome concentration) from 16% to 27%. 78 Accordingly, the suggested level chosen for our hypothesis testing needed for the verification experiment (the randomized-blind experiment) is 57°C/24 hours. 3.4.1.3 Exosome Sizes after Thermal Treatment Figure 3.8 shows the size changes of tumor (MCF-7) and normal-like (MCF-10A) exosomes after their thermal treatment. The overall result demonstrates clearly that, compared to the control (RT-0 hour), both the mode (Figure 3.8A) and the mean (Figure 3.8B) size changes of MCF-7 (red) and MCF-10A (blue) exosomes are within the range of ±10-20 nm. This indicates that during the entire process of thermal treatment and at all testing points (RT/30 minutes, 37°C/30 minutes, 47°C/30 minutes, RT/4 hours, 37°C/4 hours, 47°C/4 hours, RT/8 hours, 37°C/8 hours, 47°C/8 hours, RT/24 hours, 37°C/24 hours, 47°C/24 hours, and 57°C/24 hours), there is not a definitive trend/change or major effect on exosome sizes compared to the control (RT-0 hour). 3.4.2 Part-2 3.4.2.1 The Overall Thermal Stability of MCF-7 and MCF-10A Exosomes Figure 3.9 illustrates the thermal behavior of tumor (MCF-7) and normal-like (MCF-10A) exosomes treated according to the randomized-blind experiment at 57°C for 24 hours. Compared to the controls (MCF-7 and MCF-10A), the figure shows the percentages of concentration reduction (degradation levels) of MCF-7 exosomes in red, MCF-10A exosomes in blue, and a 50/50 mixed sample (1:1 ratio in volume and concentration) in green. Through results presented in Figure 3.9, it can be clearly seen that the percentages 79 of concentration reduction of MCF-7 exosomes (seven samples), compared to the MCF-7 control, are 38% (sample#1), 37% (sample#2), 32% (sample#4), 34% (sample#7), 36% (sample#9), 35% (sample#11), and 32% (sample#14). Accordingly, the concentration reduction of MCF-7 exosomes ranges between 32%-38% with an average of 35%. Similarly, the percentages of concentration reduction of MCF-10A exosomes (seven samples), compared to the MCF-10A control, are 67% (sample#3), 67% (sample#6), 70% (sample#8), 64% (sample#10), 69% (sample#12), 68% (sample#13), and 73% (sample#15). Thus, the concentration reduction of MCF-10A exosomes ranges between 64%-73% with an average of 68%. The concentration reduction of the 50/50 mixed sample depends on which control is used. In other words, the concentration reduction of the 50/50 mixed sample is 41% (in red) using the control of MCF-7 and/or 48% (in blue) using the control of MCF-10A. This randomized thermal experiment on MCF-7 and MCF-10A exosomes treated at 57°C/24 hours produces an average thermal difference of 33%. With a larger sample size (seven samples for each tested group), this verification experiment produces a significant statistical result of P-value <0.001. 3.4.2.2 The Identification Method of the 15-Blindly-Labeled Samples The 15-exosome samples were labeled from 1 to 15 and analyzed blindly without knowing which number (1 to 15) for which class of exosomes (class 1: seven samples of MCF-7 exosomes, class 2: seven samples of MCF-10A exosomes, class 3: one 50/50mixed sample). These 15 samples were identified as follows: As shown in the fourth segment of Figure 3.6 (at 57°C and 24 hours), the 80 reduction rates of MCF-7 and MCF-10A exosomes compared to their controls are 40% and 67%, respectively. Accordingly, by matching this known result in Figure 3.6 with the unknown results in Figure 3.9, we can conclude the following: (1) Among the 15 samples, any sample that has a concentration reduction between 32%-38% comparable to the result presented in Figure 3.6 will be clearly related to MCF-7 exosomes. Hence, aliquots # 1, 2, 4, 7, 9, 11, and 14 belong to the seven samples of MCF-7 exosomes. (2) Among the 15 samples, any sample that has a concentration reduction between 64%-73% comparable to the result presented in Figure 3.6 will be clearly related to MCF-10A exosomes. Hence, aliquots # 3, 6, 8, 10, 12, 13, and 15 belong to the seven samples of MCF-10A exosomes. It is worth mentioning that, by looking at the results of concentration reductions of MCF-7 and MCF-10 exosomes presented in Figure 3.6 (at 57°C/24 hours) and Figure 3.9 (basically at 57°C/24 hours), we find that these results are almost consistent. In other words, the average reduction rate of MCF-10A exosomes treated at 57°C/24 hours is 67% (Figure 3.6) compared to 68% (Figure 3.9). Similarly, the average reduction rate of MCF-7 exosomes treated at 57°C/24 hours is 40% (Figure 3.6) compared to 35% (Figure 3.9). Finally, we were able to identify the 50/50 mixed sample using either of the two controls shown in Figure 3.9. For instance, if we use the control of MCF-10A, then the concentration reduction of the 50/50 mixed sample, which is 48%, must be lower than the lowest concentration reduction among normal samples. In this case, it is sample #10 with 64% (see Figure 3.9). The reason for that is because the percentage of the more thermally stable tumor exosomes increased in the 50/50 mixed sample, and vice versa, if we use the control of MCF-7. In general, we were able to pick up the 50/50 mixed sample using the 81 following principle: At the same thermal condition, treating the 50%-50% sample containing a higher degraded material (say A) and a lower degraded material (say B) will either (1) drop the concentration below the level of 50% purely treated A, or (2) increase the concentration above the level of 50% purely treated B. 3.4.2.3 Exosome Sizes after Thermal Treatment Figure 3.10 shows the size changes of tumor (MCF-7) and normal-like (MCF10A) exosomes, after their thermal treatment at 57°C/24 hours. The overall result demonstrates clearly that, compared to the control (RT-0 hour), both mode (Figure 3.10A) and mean (Figure 3.10B) size changes for the seven samples of MCF-7 (red) and MCF-10A (blue) exosomes are within a range of ±10-20nm. This indicates that there is not a definitive trend/change or major effect on exosome sizes after their thermal treatment at 57°C/24 hours. The mode (Figure 3.10A) and the mean (Figure 3.10B) sizes each match at the 50/50 mixed sample, which essentially has a median size between the sizes of the MCF-7 and MCF-10A exosomes. Figure 3.11 shows exosome concentration versus size distribution of MCF-7 and MCF-10A exosomes after their thermal treatment at 57°C/24 hours. Comparable to the controls, the treated MCF-10A exosomes (average of seven samples in light blue) degraded significantly in contrast with the treated MCF-7 exosomes (average of seven samples in light red). The overall standard distribution of MCF-7 is narrower than that of the MCF-10A exosomes. Again, this result confirms that after the thermal treatment at 57°C/24 hours, there is no a definitive shift in the sizes of both exosomes (MCF-7 and MCF10A). However, there is a definitive drop in exosome concentration, particularly, in 82 MCF-10A exosomes. The standard distribution of the 50/50 mixed sample (in green) shows almost a median size distribution between both treated exosomes (MCF-7 and MCF-10A). 3.4.3 Part-3 3.4.3.1 Exosome Concentration and Their Stability Figure 3.12 shows the changes of MCF-7 exosome stability as a function of their concentration when they are incubated at room temperatures for 24 hours. Compared to the controls (at 0 hour), the figure shows that when exosomes are incubated for 24 hours at the highest concentration (concentration factor of 20x), they observe the least concentration reduction (27%). In contrast, when exosomes are incubated for 24 hours in lower concentration (concentration factor of 5x and 10x), they observe higher concentration reduction (46% and 50%, respectively). The overall result indicates that there is a proportional relationship between exosome stability and their concentration. 3.5 Discussion Based on the presented experimental design, hypothesis testing, and data analysis, it has been demonstrated that differentiation between tumor (MCF-7) and normal-like (MCF-10A) exosomes using their biophysical property of “thermal stability” is definitely possible with a 33% average difference between their relative stabilities. Applying the same baseline of exosomes concentration (± 10% difference) and treatment condition (57°C/24 hours) can produce a significant statistical difference (P-value < 0.001) between the treated exosomes. This result demonstrates, for the first time to our knowledge, that 83 different populations of exosomes can have a distinct thermal stability. The achieved results shown in Figures 3.6, 3.7, and 3.9 lead to the following conclusions: (1) The overall degradation trend of the tested exosomes (MCF-7 and MCF10A) decreases gradually with temperature and incubation time. (2) As long as the incubation time is below 8 hours (for the tested temperatures; RT, 37°C, and 47°C), the thermal stability of normal-like (MCF-10A) exosomes is higher (by 1%-13%) than the thermal stability of tumor (MCF-7) exosomes, but still insignificantly so (P-value=0.26). (3) After 24 hours of incubation (for the tested temperatures; RT, 37°C, 47°C, and 57°C), the property of “higher thermal stability” is changed in the favor of tumor MCF-7 exosomes. (4) The 33% thermal stability difference between MCF-7 and MCF-10A exosomes (obtained at 57°C/24 hours) indicates that the combined effect of the highest incubation time (24 hours) and highest temperature (57°C) is the best experimental parameter to cause a significant differentiation between tumor (MCF-7) and normal-like (MCF-10A) exosomes. Based on the above conclusions, tumor (MCF-7) exosomes are significantly more thermally stable than normal-like (MCF-10A) exosomes, if they are treated at 57°C/24 hours. 3.5.1 The Relationship between Exosome Sizes and Their Thermal Stability The achieved results shown in Figures 3.8, 3.10, and 3.11 result in the following conclusions: (1) The size changes of MCF-7 and MCF-10A exosomes range between ±10-20nm without a definitive change or major effect after the entire process of thermal treatment. This conclusion is confirmed by another study,13 in which the exosome sizes 84 did not undergo a definitive change after the treatment at two different temperatures (RT and 4°C). (2) Although both exosomes (MCF-7 and MCF-10A) were treated at a severe treatment condition (57°C/24 hours), the smaller tumor (MCF-7) exosomes exhibited a higher thermal stability (less degradation level) than the larger normal (MCF-10A) exosomes. 3.5.2 The 50/50 Mixed Sample Regarding the 50/50 mixed sample (1:1 ratio in volume and concentration), there are two main conclusions: (1) The 50/50 mixed sample can be identified easily through the changes of concentration demonstrated by either the figure of concentration profiles (Figure 3.9), and/or the figure of standard distributions (Figure 3.11). (2) We cannot have a 100%-reliance upon the property of size to identify the 50/50 mixed sample. In other words, the simple recognition of the 50/50 mixed sample using the changes of exosome concentration becomes difficult when it comes to the changes in size. This is because both mode and mean sizes of the 50/50 mixed sample can overlap with other sizes, such as the mean size of the MCF-7 control (see Figure 3.10). In general, the result of the 50/50 mixed sample suggests a new principle named “Exosome Thermal Enrichment” (ETE). This rapid degradation of normal-like (MCF10A) exosomes from the mixture (the 50/50 mixed sample) and the enrichment of tumor (MCF-7) exosomes can have two promising implications: (1) Exosome thermal enrichment can aid in the early detection of cancer. That is, after the thermal treatment, having a lower concentration reduction in the main population means a probable higher amount of tumor exosomes had existed in the original sample (the untreated sample), and 85 vice versa. (2) Exosome thermal enrichment can be employed for the detection of cancer biomarkers through the analysis of the molecular content enclosed with the enriched tumor exosomes, such as miRNAs. 3.5.3 Mechanism of Exosome Thermal Degradation We hypothesize that there are two possible scenarios affecting the thermal stability (degradation level) of exosomes. As shown in Figure 3.13, the two possible scenarios are: (1) The degradation in exosome surface proteins; and/or (2) The degradation in exosome lipid composition. The two possible mechanisms could occur individually or simultaneously. However, X. Osteikoetxea et al.82 published important information, which helps us to provide the “most likely” mechanism of exosome thermal degradation. The article provided novel parameters to characterize three subpopulations of extracellular vesicles (EVs) with different sizes: Apoptotic bodies, 1627 nm; Microvesicles, 208 nm; and Exosomes, 98 nm. The major conclusions revealed by the article are as follows: (1) The level of protein to lipid ratio increases when the size of EVs increases. Thus, exosomes have the least protein to lipid ratio. (2) Exosomes have the highest enrichment of lipids, particularly cholesterol. (3) The lipid order increases when the size of EVs decreases. Thus, exosomes have the highest lipid order. To that end, we hypothesize the following mechanism for exosome thermal degradation. As illustrated in Figure 3.14, the size of tumor (MCF-7) exosomes has been demonstrated to be smaller than normal-like (MCF-10A) exosomes (see Figures 3.8, 3.10, and 3.11). Therefore, the smaller tumor (MCF-7) exosomes are more likely to have higher membrane lipid order with higher content of cholesterols than the larger normal- 86 like (MCF-10A) exosomes. However, the abundance of surface proteins is more likely to be lower in tumor (MCF-7) exosomes than they are in normal-like (MCF-10A) exosomes. Consequently, the normal-like (MCF-10A) exosomes enriched with more surface proteins, which usually start to denture at much lower temperature (41°C) than cholesterols (148°C) degrade more rapidly. In other words, the higher sensitivity to the thermal treatment by surface proteins, especially at 57°C/24 hours, results in an intensive heat at the surface of MCF-10A exosomes. This higher potential of the heating process (due to the combined effect of temperature and time) can easily burst the wall of the MCF-10A exosomes and enhance a faster thermal degradation. In the above hypotheses, we did not focus on exosome nucleic acids as a cause of thermal degradation since we believe that the decorating proteins and lipids are more susceptible to the heat effect than the shielded nucleic acids. That is, the exosomal nucleic acids have the least effect on the thermal stability of tumor MCF-7 and normallike MCF-10A exosomes compared to their surface proteins and lipids. In general, this hypothesis needs to be validated by further characterization methods to determine the total content of exosomal proteins and lipids of both of MCF-7 and MCF-10A exosomes. A comprehensive total protein profiling (proteomics) and lipid profiling (lipidomics) should be performed together with the thermal treatment of exosomes. There are a variety of common methods82,83 to perform such characterizations such as Mass Spectrometry, Flow Cytometry, Immunoblot, Spectral Ratiometric Imaging, and NTA-based fluorescent measurements. To emphasize, utilizing the two results revealed by this study and by X. Osteikoetxea, we believe that the size of exosomes plays an important role to determine the relationship between the thermal stability of exosomes and their content of lipids and 87 proteins. In other words, having different thermal stability between different subpopulations of exosomes can be attributed to the relationship between sizes of exosomes and their membrane compositions, mainly the level of proteins and lipids. 3.5.4 The Effect of Low Concentration There is a proportional relationship between exosome stability and their concentration. This relationship can be clearly seen through the result in Figure 3.12, in which the stability of exosomes increases with their concentration, and vise versa. Although exosomes degrade naturally with time at room temperature, the speed of their degradation increases if they are present in lower concentration. The exact cause of this observation is still unknown and it might be attributed to the physical support exerted by exosomal particles to each other. Since no publications have discussed this topic, it is highly recommended to conduct further research and more experimental work in this regard. The effect of low concentration can be avoided at the beginning of the thermal treatment by starting with the highest “possible” concentration as a baseline. In this study, the concentration baseline for both types of exosome (MCF-7 and MCF-10A) has been chosen according to the maximum possible concentration measurement that can be obtained by NTA. In this case, it is 1.4-2.0x109 particles/ml (70-100 particles/frame). 3.6 Implications and Significance The presented approach of exosome thermal characterization is believed to have promising applications. It can be developed to characterize exosomes against the entire 88 exosome subpopulations by exploiting the dual power of exosomal thermal and molecular characteristics. This approach, to the best of our knowledge, has not been investigated to any extent. Although M. Lee et al.83 and H. Zhou et al.84 presented interesting studies to investigate the thermal stability of the molecular content of exosomes; the difference between our approach and their studies is still significant. For example, our approach is designed to differentiate between tumor and normal-like exosomes based on their thermal degradation levels at different temperatures and incubation times. In contrast, Lee’s conceptual work was to investigate the integrity of exosome-associated proteins, RNAs, and other exosome markers under the effect of different storage temperature for short and long incubation times. Moreover, Lee’s study applied Flow Cytometry, not NTA, to measure the degradation level of the treated exosomes. In general, Lee’s results showed that for a short period of time (30 minutes), all exosome-associated proteins mostly degraded after their incubation at 90°C. For long-term incubation (10 days), the amounts of protein and RNAs, were mostly reduced at RT compared to -70°C and 4°C. However, long-term incubation (10 days) resulted in a major loss of CD63 at 4°C and RT compared to -70°C. On the other hand, Zhou’s study investigated how to better collect, store, and preserve urinary exosomal proteins. In that investigation, the abundance of urinary exosome-associated proteins for three urine samples stored at 4°C, -20°C, and -80°C were measured. The main conclusion from this study indicated that there was a 27.4% recovery of exosome proteins for the sample stored at -20°C compared to the sample stored at 4°C. However, recovery of exosome proteins can be up to 86% for the sample 89 stored at -80°C. Accordingly, integration between our approach and both Lee’s and Zhou’s studies support the above suggestion to characterize different subpopulations of exosomes (normal and diseased exosomes) dually by utilizing their properties of thermal stabilities and molecular content (Thermal-Molecular Method). Such an approach is believed to add a new dimension to the field of exosomes, and their methods of characterization. 3.7 Conclusion The conducted research demonstrated that exosome-thermal stability is a new biophysical property that can be used to distinguish tumor (MCF-7) and normal-like (MCF-10A) exosomes. The presented results revealed that treating exosomes at the thermal parameter of 57°C/24 hours can definitely differentiate between MCF-7 and MCF-10A exosomes with a 33% average difference and a significant statistical result (Pvalue < 0.001). Since the presented analytical method is believed to be novel, it has been named the Time-Temperature Method (TTM) for exosome thermal characterization. The TTM can be developed to provide a reliable tool to get a better understanding of the thermal behavior of different exosome subpopulations, including normal and diseased ones. When the TTM was tested to differentiate between 15 blind, randomized samples (seven samples for each of MCF-7 and MCF-10A exosomes, and one 50/50 mixed sample), the method successfully determined the identities of all 15 samples based on the degradation level of exosome concentration. The presented results concluded that the mean and mode size changes of MCF-7 90 and MCF-10A exosomes remains insignificant and without a definitive change or major effect during the thermal treatment. Although both exosomes (MCF-7 and MCF-10A) were treated at the highest severe treatment conditions (57°C/24 hours), the smaller tumor (MCF-7) exosomes exhibited a higher thermal stability (less degradation) than the larger normal (MCF-10A) exosomes. The approach reveals a new principle, named, “Exosomes Thermal Enrichment, ETE”. This principle was demonstrated by the simple recognition of the 50/50 mixed sample. With further research, this principle can be utilized for the following promising applications: (1) for the early detection of cancer by measuring the percentage of the enriched tumor exosomes, compared to the degraded normal exosomes in the main population, (2) for the detection of cancer biomarkers through the analysis of the molecular content enclosed within the enriched tumor exosomes, such as miRNAs. The presented work suggested the moslt likely mechanism of exosome thermal degradation. This mechanisim hypothesizes a relationship between the sizes of exosomes and their membrane conjugated macromolecules, mainly the decorating proteins and lipids. In order to make the presented approach applicable for more studies and clinical uses, particularly in the field of the early detection of cancer, there are three important points to consider before generalizing the above results: (1) The tested tumor (MCF-7) and normal-like (MCF-10A) exosomes are derived from cell lines. Hence, the thermal behavior of exosomes derived from body fluids, such as serum or urine, have to be investigated. (2) The tested exosomes are derived from same epithelial cells (breast cells). Thus, exosomes derived from different epithelial cells, such as liver or pancreatic cells, 91 have to be investigated. (3) The tested exosomes are derived from female cell lines. Hence, exosomes derived from both males and females (of different ages) have to be investigated. 92 Figure 3.1. Thermal stability of plasma and urine derived exosomes measured at different incubation times and at two different temperatures: room temperature and 4°C. Adapted from13 93 A B Figure 3.2. The size changes of exosomes stored at two different temperatures: (A) at 4°C. (B) at 37°C. Adapted from48 94 Figure 3.3. The concentration of urine particle including exosomes was analyzed at different temperatures (RT, 4°C, -20°C, and -80°C) and three incubation times (2 hours, 1 day or 1 week). The concentration of the baseline was measured immediately after sample collection. Abbreviations: PI, protease inhibitors; and RT, room temperature. Adapted from77 95 Figure 3.4. Schematic representation of the controllable factors (red arrows) used in the thermal treatment process of exosomes. Adapted from81 96 A B Figure 3.5. The two-factorial design of two quantitative factors: Temperature and Incubation time. (A) The 14 testing points to measure the changes of exosome concentration as a function of temperature and incubation times. All points are compared to the control (20°C, 0 hour). Symbols: the points of corners (in blue), the points inbetween (in black), one point in the outer range (in red), and one point of control (in green). (B) The minimum number of treatment combinations for the two-factorial design. These treatment combinations are the points of corners shown in (A). Abbreviations: Factor A is the temperature; Factor B is the incubation time. Adapted from81 97 Table 3.1. Incubation conditions applied to study the thermal stability of tumor (MCF-7) and normal-like (MCF-10A) exosomes based on the two-factorial design. Abbreviations: RT, room temperature. MCF-7/ MCF-10A Testing points 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Temperature/Incubation time RT (control) RT/0.5 hour 37°C/0.5 hour 47°C/0.5 hour RT/4 hours 37°C/4 hours 47°C/4 hours RT/8 hours 37°C/8 hours 47°C/8 hours RT/24 hours 37°C/24 hours 47°C/24 hours 57°C/24 hours 98 Table 3.2. Experimental procedure (sampling) applied to study the thermal stability of tumor (MCF-7) and normal-like (MCF-10A) exosomes. Abbreviations: RT, room temperature. A Main Run # Description Temp Incubation Measurements stock time 10 1 Analyze it RT 0 hour Particle ml (Vial#1 has immediately after concentration, 900 µl) dilution, and then mean size, average it with run2 mode size, and as a control. size distribution 2 Leave it at RT to be RT 0 hour (Vial#2 has analyzed after run1, 900 µl) and then average it with run1 as a control. 3 Place it in the block 37°C 0.5 hour (Vial#3 has heater immediately 900 µl) after dilution to be analyzed after run2. 4 Leave it at RT, and 47°C 0.5 hour (Vial#4 has then place it in the 900 µl) block heater after heating vial#3 to be analyzed after run3. 5 Leave it at RT, and 37°C 0.5 hour (Vial#5 has then place it in the 900 µl) block heater after heating vial#4 to be analyzed after run4. 6 Leave it at RT, and 47°C 0.5 hour (Vial#6 has then place it in the 900 µl) block heater after heating vial#5 to be analyzed after run5. 99 Table 3.2 Continued. It follows the experimental procedure shown in Table 3.2A to study the thermal stability of tumor (MCF-7) and normal-like (MCF-10A) exosomes. Abbreviations: RT, room temperature. B Main Run # Description Temp Incubation Measurements stock time 10 1 Analyze it RT 0 hour Particle ml (Vial#1 has immediately after concentration, 900 µl) dilution, and then mean size, average it with run2 mode size, and as a control. size distribution 2 Leave it at RT to be RT 0 hour (Vial#2 has 1 analyzed after run1, 900 µl) and then average it with run1 as a control. 3 Place it in the block 47°C 8 hours (Vial#3 has heater immediately 900 µl) after dilution to be analyzed after run2. 4 Place it in the block 47°C 8 hours (Vial#4 has heater immediately 900 µl) after dilution to be analyzed after run3. 5 Leave it at RT to be RT 8 hours (Vial#5 has analyzed after run4. 900 µl) 6 Leave it at RT to be RT 8 hours (Vial#6 has analyzed after run5. 900 µl) 7 Place it in the block 47°C 24 hours (Vial#7 has heater right after 900 µl) dilution to be analyzed after run6. 8 Place it in the block 47°C 24 hours (Vial#8 has heater immediately 900 µl) after dilution to be analyzed after run7. 9 Leave it at RT to be RT 24 hours (Vial#9 has analyzed after run8. 900 µl) 10 Leave it at RT to be RT 24 hours (Vial#10 has analyzed after run9. 900 µl) 100 Table 3.2 Continued. It follows the experimental procedure shown in Table 3.2B to study the thermal stability of tumor (MCF-7) and normal-like (MCF-10A) exosomes. Abbreviations: RT, room temperature. C Main Run # Description Temp Incubation Measurements stock time 10 1 Analyze it RT 0 hour Particle ml (Vial#1 has immediately after concentration, 900 µl) dilution, and then mean size, average it with run2 mode size, and as a control. size distribution 2 Leave it at RT to be RT 0 hour (Vial#2 has analyzed after run1, 900 µl) and then average it with run1 as a control. 3 Place it in the block 37°C 4 hours (Vial#3 has heater immediately 900 µl) after dilution to be analyzed after run2. 4 Place it in the block 37°C 4 hours (Vial#4 has heater immediately 900 µl) after dilution to be analyzed after run3. 5 Leave it at RT, to be RT 4 hours (Vial#5 has analyzed after run4. 900 µl) 6 Leave it at RT, to be RT 4 hours (Vial#6 has analyzed after run5. 900 µl) 7 Place it in the block 37°C 8 hours (Vial#7 has heater immediately 900 µl) after dilution to be analyzed after run6. 8 Place it in the block 37°C 8 hours (Vial#8 has heater immediately 900 µl) after dilution to be analyzed after run7. 9 Place it in the block 37°C 24 hours (Vial#9 has heater immediately 900 µl) after dilution to be analyzed after run8. 10 Place it in the block 37°C 24 hours (Vial#10 has heater immediately 900 µl) after dilution to be analyzed after run9. 101 Table 3.2 Continued. It follows the experimental procedure shown in Table 3.2C to study the thermal stability of tumor (MCF-7) and normal-like (MCF-10A) exosomes. Abbreviations: RT, room temperature. D Main Run # Description Temp Incubation Measurements stock time 10 1 Analyze it RT 0 hour Particle ml (Vial#1 has immediately after concentration, 900 µl) dilution, and then mean size, average it with run2 as mode size, and a control. size distribution 2 Leave it at RT to be RT 0 hour (Vial#2 has analyzed after run1, 900 µl) and then average it with run1 as a control. 3 Leave it at RT, to be RT 0.5 hour (Vial#3 has analyzed after run2. 900 µl) 4 Leave it at RT, to be RT 0.5 hour (Vial#4 has analyzed after run3. 900 µl) 5 Place it in the block 47°C 4 hours (Vial#5 has heater immediately 900 µl) after dilution to be analyzed after run4. 6 Place it in the block 47°C 4 hours (Vial#6 has heater immediately 900 µl) after dilution to be analyzed after run5. 7 Leave it at RT, to be RT 4 hours (Vial#7 has analyzed after run6. 900 µl) 8 Leave it at RT, to be RT 4 hours (Vial#8 has analyzed after run7. 900 µl) 9 Place it in the block 57°C 24 hours (Vial#9 has heater immediately 900 µl) after dilution to be analyzed after run8. 10 Place it in the block 57°C 24 hours (Vial#10 has heater immediately 900 µl) after dilution to be analyzed after run9. 102 Table 3.3 The experimental procedure (sampling) applied for the blind, randomized experiment of tumor (MCF-7) and normal-like (MCF-10A) exosomes at 57°C/24 hours. Abbreviations: RT, room temperature. Main stock 10ml for each type of exosome Description Run Temp Two vials (900 µl each), for each type of exosomes were analyzed by NTA and then averaged as a control 15 vials (900 µl each) • 7 for MCF-7 • 7 for MCF-10A • 1 mixed sample 50/50 Analyze it immediately after dilution. RT Immediately after dilution, the 15 vials were first marked from 1 to 15 and then randomized as blind samples for the experimenter. Then, the 15 vials were placed in the block heater immediately after randomization. After 24h, all vials were removed from the heater at once, and analyzed subsequently by NTA (1 to 15). 57°C Incubation Measurements time 0 hour Particle concentration, mean size, mode size, and size distribution 24 hours 103 1.80# MCF$7&exosomes& Normalized#ConcentraAon# 1.60# 1.40# 0.5 h MCF$10A&exosomes&& 4h 8h 24 h 1.20# 2%#1%# 1.00# 9%# 11%# 11%# 13%#8%# 15%# 0.80# 10%# 23%# 14%# 26%# 22%# 28%# 24%# 31%# 36%# 34%# 34%# 36%# 35%# 0.60# 50%# 40%# 42%# 58%# 0.40# 67%# 0.20# 62 4h # 57 °C 62 4h # 47 °C 62 4h # 4h # 37 °C 62 RT 68 h# 47 °C h# 68 h# 37 °C 6#8 RT 64 h# 47 °C h# 64 h# 37 °C 64 RT 63 0# 47 °C # in 63 0# 37 °C 0#m 63 RT co nt ro l# 0.00# Figure 3.6. The profiles of the thermal stability of tumor (MCF-7) and normal-like (MCF-10A) exosomes: Compared to controls, the figure shows percentages of degradation in the normalized concentration of MCF-7 (red) and MCF-10A (blue) exosomes incubated at four different times (30 minutes, 4 hours, 8 hours, and 24 hours) and four temperatures (RT, 37°C, 47°C, and 57°C). Abbreviations: RT, room temperature; h, hour(s). 104 A B MCF,7#exosomes# MCF,10A#exosomes# 30# 25# 66# 64# 65# 50# 58# 42# 72# 78# 69# 76# 64# 66# 87# 92# 77# 90# 74# 86# 98# 99# 89# 91# 85# 89# 60# 33# Incuba3on#Time,#h# 20# 15# 10# 5# 0# 100# 100# ,5# ,5# 0# 5# 10# 15# 20# 25# 30# Temperature,#°C# 35# 40# 45# 50# 55# 60# 65# Figure 3.7. For the 14 testing points of the two-factorial design, Figure (A) shows the difference in concentration measurements (difference in thermal stabilities) between tumor (MCF-7) and normal-like (MCF-10A) exosomes. (B) Displays the values of the normalized concentration (bubble sizes) of MCF-7 (red) and MCF-10A (blue) exosomes. All bubbles are compared to controls located at (x=0, y=0) 105 A 200.00# Mode%MCF(7%exosomes% Mode%MCF(10A%exosomes% 180.00# 160.00# 140.00# Size,#nm# 120.00# 100.00# 80.00# 60.00# 40.00# 20.00# 12 4h # 57 °C 12 4h # 47 °C # 12 4h # 4h 37 °C RT 12 18 h# 47 °C 18 h# h# 37 °C RT 1#8 14 h# h# 14 h# 47 °C 37 °C RT 14 co nt ro l# RT 13 0# m in # 37 °C 13 0# m in # 47 °C 13 0# m in # 0.00# B Mean%MCF(7%exosomes% 200.00# Mean%MCF(10A%exosomes% 180.00# 160.00# 140.00# Size,#nm# 120.00# 100.00# 80.00# 60.00# 40.00# 20.00# 12 4h # 57 °C 12 4h # 47 °C 12 4h # # 37 °C 4h 12 RT 18 h# 47 °C 18 h# 37 °C h# 1#8 RT 14 h# 14 h# 47 °C 37 °C h# 14 RT co nt ro l# RT 13 0# m in # 37 °C 13 0# 47 °C 13 0# 0.00# Figure 3.8. Compared to the control (RT-0 hour), each figure of mode (A) and mean (B) shows the size changes of MCF-7 (red) and MCF-10A (blue) exosomes during the thermal treatment process at different temperatures and incubation times. 106 2.00E+09$ MCF$7&exosomes& 1.90E+09$ MCF$10A&exosomes& 1.80E+09$ 1.70E+09$ 1.60E+09$ 1.50E+09$ 1.40E+09$ Concentra=on$(#/ml)$ 1.30E+09$ 1.20E+09$ 1.10E+09$ 32%$ 1.00E+09$ 9.00E+08$ 38%$ 37%$ 34%$ 36%$ 35%$ 32%$ 41%,$48%$ 8.00E+08$ 7.00E+08$ 64%$ 67%$ 6.00E+08$ 67%$ 5.00E+08$ 69%$ 70%$ 68%$ 73%$ 4.00E+08$ 3.00E+08$ 2.00E+08$ 1.00E+08$ 15 $ 13 $ 12 $ 10 $ 8$ 6$ 3$ 14 $ (5 0/ 50 )$ 11 $ 9$ 7$ 4$ 2$ 1$ Co nt ro l$ Co nt ro l$ 0.00E+00$ Figure 3.9. The thermal profiles of exosomes treated according to the blind, randomized experiment at 57°C for 24 hours. The y-axis shows exosome concentration and the x-axis shows the sample numbers. Compared to controls, the figure shows the percentages of concentration reduction (degradation levels) of tumor MCF-7 exosomes (seven samples in red), normal-like MCF-10A exosomes (seven samples in blue), and the 50/50 mixed sample (one sample in green). The 50/50 mixed sample has two reduction rates of 41% (in red) and 48% (in blue), if it is compared to the controls of MCF-7 and MCF-10A, respectively. 107 A 200" Mode,&MCF)10A& Mode,&MCF)7& 180" 160" Size,"nm" 140" 120" 100" 80" 60" 40" 20" 0" Control" (RT10h)" 1" 2" 3" 4" 5" 6" 7" (50/50)" B Mean,&MCF)10A& 200" Mean,&MCF)7& 180" 160" Size,"nm" 140" 120" 100" 80" 60" 40" 20" 0" Control" (RT10h)" 1" 2" 3" 4" 5" 6" 7" (50/50)" Figure 3.10. Compared to the controls (RT-0 hour), the figure shows the mode (A) and mean (B) size changes for the 15 blind, randomized samples (MCF-7 in red, MCF-10A in blue, and the 50/50) all treated at 57°C for 24 hours. The sizes of mode (A) and mean (B), each match at the 50/50 mixed sample. 108 2.00E+08% MCF17,%control% MCF110A,%control% 1.80E+08% Treated%MCF17%(Avgerage%of%7%samples)% Treated%MCF110A%(Average%of%7%samples)% 1.60E+08% Treated%mixed%sample%(50/50)% Concentra)on*(#/ml)* 1.40E+08% 1.20E+08% 1.00E+08% 8.00E+07% 6.00E+07% 4.00E+07% 2.00E+07% 0.00E+00% 50% 100% 150% Size*(nm)* 200% 250% 300% Figure 3.11. Compared to the controls, the figure shows the standard distribution of MCF-7 exosomes (average of seven samples in light red), MCF-10A exosomes (average of seven samples in light blue), and the 50/50 mixed sample (in green) all treated at 57°C for 24 hours. The inserted table shows mode and mean sizes of the plotted standard distributions. 109 2.00E+09% 1.80E+09% Concentra1on$(#/ml)$ 1.60E+09% 1.40E+09% 27%$ 1.20E+09% 1.00E+09% 8.00E+08% 50%$ 6.00E+08% 46%$ 4.00E+08% 2.00E+08% 0.00E+00% 5x.0h% 5x.24h% 10x.0h% 10x.24h% 20x.0h% 20x.24h% Exosome$concentra1on$fatctors$(5x,10x,$and$20x),$each$at$two$incuba1on$1mes$(0h$and$24h)$ Figure 3.12. Compared to three controls (5x-0 hour, 10x-0 hour, 20x-0 hour) of three different exosome concentrations (three concentration factors 5x, 10x, and 20x), if MCF7 exosomes are incubated at room temperature for 24 hours (5x-24 hours, 10x-24 hours, 20x-24 hours) they produce different concentration reductions (46%, 50%, and 27% respectively). 110 Figure 3.13. The two possible mechanisms of exosome thermal degradation: (1) the degradation in exosome surface proteins; (2) the degradation in exosome lipid composition. 111 Figure 3.14. A descriptive illustration showing the abundance of exosome surface composition: Normal-like (MCF-10A) exosomes, with their larger size, are more enriched with surface proteins (long/thin blue lines) than tumor (MCF-7) exosomes. In contrast, tumor (MCF-7) exosomes, with their smaller size, are more abundant with cholesterols (short/thick yellow lines) than normal (MCF-10A) exosomes. 112 CHAPTER 4 GENERAL CONCLUSIONS AND FUTURE WORK Today, there is a very intense competition between many research centers around the world to develop simpler, cheaper, and more reliable exosome-based tools with easy protocol for the early detection of cancer before reaching an advanced-untreatable deadly stage. Several articles published in prestigious journals agreed that the exosomes play a key role in cancer progression/metastasis.85,86 Hence, exosomes have a very promising future, not only as novel biomarkers for the early detection of cancer, but also as therapeutics to overcome many current limitations in the treatment of cancer, as well as to raise the global survival rates of patients. To accelerate and stimulate research in the above direction, we have presented two new methods for exosome characterization. In these methods, exosomes were characterized simultaneously through a relationship established between: (1) exosomal biophysical properties (exosomal size, mass, density, and thermal stability) and (2) the corresponding molecular makeup (the enclosed level of one cancer biomarker, miR21). Although it is not an easy task to characterize biological membrane vesicles on a nano-scale, such as exosomes, our achieved results indicate clearly that the idea of exploiting the dual power of exosome biophysical and molecular properties has provided 113 novel characterization methods to differentiate between tumor and normal exosomes. However, to make these methods more attractive and reliable to cover different medical applications, further research studies are still needed to critique, develop, and explore the presented methods, especially since we know that there is an absence of research in the development of an accurate experimental protocol using the above idea. To help in this direction, here are some recommendations for future work. 4.1 Recommendations for the First Approach (presented in Chapter 2) 1- Apply the developed method to fully explore the heterogeneity of other types of tumor exosomes rather than MCF-7 exosomes, and provide measurements for their biophysical properties (size, mass, density). 2- Apply the developed method to characterize different subpopulations of exosomes (normal or diseased) derived from cell lines and body fluids. 3- Apply the developed method to enrich for different standard cancer biomarkers under the family of miRNAs (beside miR21). 4- Apply the developed method to conduct a comparative characterization between healthy and tumor exosomes in order to establish a correlation between their biophysical and their molecular properties. 5- Measure the effectiveness of the developed method on isolating different subpopulations of EVs, or different exosome subtypes. 114 4.2 Recommendation for the Second Approach (presented in Chapter 3) 1- Apply the developed method to characterize the thermal behavior of different subpopulations of exosomes (normal or diseased) derived from cell lines and body fluids. 2- Apply the developed method using the best experimental conditions (57°C/24 hours) to produce a general thermal profile for exosomes derived from cells of the top-10 deadliest cancers (with the lowest 5-year survival rates, according to the National Cancer Institute, NCI, and the American Cancer Society, ACS). 3- Since temperature is the leading factor in our two-factorial design, it is recommended to study the effectiveness of the presented method, on MCF-7 and MCF-10A exosomes, but at higher temperatures (≥100°C) and lower incubation times (≤30 minutes). If such investigation succeeds, it will allow us to develop a simple point-of-care device for tumor exosome detection (a half an hour-thermal detector). 4- Apply the developed method to characterize exosomes using different heating sources such as Sunlight, Laser, Microwaves, Radiation sources, etc. This could provide better solutions to increase efficiency and reduce time and cost that are experienced with the process of conventional heating. 5- Apply the developed method to characterize exosomes, simultaneously, with thermal behavior of their molecular makeup (i.e., thermal-molecular method). This can help to characterize exosomes more precisely through a correlation between their thermal stability and their molecular properties. This coincident 115 molecular analysis of thermally treated exosomes can be performed on their surface cargo (proteins and lipid compositions) or luminal cargo (miRNAs and DNA). 6- Apply the developed method(s) to characterize exosomes after genetic transfection.87 Accomplishing these recommendations could be important to answer many questions such as: (1) Do all kinds of tumor exosomes have higher thermal stabilities than normal exosomes? (2) Why do cancer cells release more exosomes than normal cells? (3) Why do solid tumors have a higher temperature (higher metabolic activity) than the surrounding normal tissues? (4) Is it true that that tumor cells release more thermally stable exosomes to protect their signals needed for tumor metastasis? Getting a precise and correct answer for these questions and others can be done through additional research and exploration in the exosome field. In the coming years, exosomes are believed to provide major progress in the field of early detection of cancer, with more reliable and cost-effective solutions. Having a diagnostic exosome-based tool could replace the current conventional screening methods, which in many cases fail to detect cancer until a late stage. 116 REFERENCES 1. Vlassov, A. V., Magdaleno, S., Setterquist, R. & Conrad, R. Exosomes: current knowledge of their composition, biological functions, and diagnostic and therapeutic potentials. Biochim. Biophys. Acta 1820, 940–8 (2012). 2. www. carislifesciences.eu/mechanism-of-action. 3. Trams, E. G., Lauter, C. J., Salem Jr., N. & Heine, U. exfoliation of membrane ecto-enzymes in the form of micro-vesicles. Biochim Biophys Acta 645, 63–70 (1981). 4. Harding, C., Heuser, J. & Stahl, P. Receptor-mediated endocytosis of transferrin and recycling of the transferrin receptor in rat reticulocytes. J. Cell Biol. 97, 329– 39 (1983). 5. Pan, B. T., Teng, K., Wu, C., Adam, M. & Johnstone, R. M. Electron microscopic evidence for externalization of the transferrin receptor in vesicular form in sheep reticulocytes. J. Cell Biol. 101, 942–8 (1985). 6. Johnstone, R. M., Adam, M., Hammond, J. R., Orr, L. & Turbide, C. Vesicle formation during reticulocyte maturation. Association of plasma membrane activities with released vesicles (exosomes). J. Biol. Chem. 262, 9412–20 (1987). 7. Valadi, H. et al. Exosome-mediated transfer of mRNAs and microRNAs is a novel mechanism of genetic exchange between cells. Nat Cell Biol 9, 654–659 (2007). 8. Chevillet, J. R. et al. Quantitative and stoichiometric analysis of the microRNA content of exosomes. Proc. Natl. Acad. Sci. U. S. A. 111, 14888–14893 (2014). 9. Duijvesz, D., Luider, T., Bangma, C. H. & Jenster, G. Exosomes as biomarker treasure chests for prostate cancer. European urology 59, 823–31 (2011). 10. Raposo, G. & Stoorvogel, W. Extracellular vesicles: exosomes, microvesicles, and friends. J Cell Biol 200, 373–383 (2013). 11. Waldenström, A. & Ronquist, G. Role of exosomes in myocardial remodeling. Circ. Res. 114, 315–324 (2014). 117 12. Sahoo, S. et al. Exosomes from human CD34+ stem cells mediate their proangiogenic paracrine activity. Circ. Res. 109, 724–728 (2011). 13. Malvern Note, Measuring exosome stability with nanoparticle tracking analysis, (2014). 14. Chernyshev, V. et al. Size and shape characterization of hydrated and desiccated exosomes. Anal. Bioanal. Chem. (2015). doi:10.1007/s00216-015-8535-3 15. Théry, C. et al. Membrane vesicles as conveyors of immune responses. Nat Rev Immunol 9, 581–593 (2009). 16. Hu, G., Drescher, K. M. & Chen, X.-M. Exosomal miRNAs: biological properties and therapeutic potential. Front. Genet. 3, 1–9 (2012). 17. Silverman, J. M. & Reiner, N. E. Exosomes and other microvesicles in infection biology: organelles with unanticipated phenotypes. Cell. Microbiol. 13, 1–9 (2011). 18. Schorey, J. S. & Bhatnagar, S. Exosome function: from tumor immunology to pathogen biology. Traffic 9, 871–81 (2008). 19. Colombo, M., Raposo, G. & Théry, C. Biogenesis, secretion, and intercellular interactions of exosomes and other extracellular vesicles. Annu. Rev. Cell Dev. Biol. 30, 255–289 (2014). 20. Silverman, J. M. & Reiner, N. E. Exosomes and other microvesicles in infection biology: organelles with unanticipated phenotypes. Cell. Microbiol. 13, 1–9 (2011). 21. Conde-Vancells, J. et al. Characterization and comprehensive proteome profiling of exosomes secreted by hepatocytes. J. Proteome Res. 7, 5157–5166 (2008). 22. Subra, C. et al. Exosomes account for vesicle-mediated transcellular transport of activatable phospholipases and prostaglandins. J. Lipid Res. 51, 2105–20 (2010). 23. Mathivanan, S. & Simpson, R. J. ExoCarta: A compendium of exosomal proteins and RNA. Proteomics 9, 4997–5000 (2009). 24. Gonzales, P. a et al. Large-scale proteomics and phosphoproteomics of urinary exosomes. J. Am. Soc. Nephrol. 20, 363–79 (2009). 25. Mittelbrunn, M. et al. Unidirectional transfer of microRNA-loaded exosomes from T cells to antigen-presenting cells. Nat. Commun. 2, 282 (2011). 26. Skog, J. et al. Glioblastoma microvesicles transport RNA and proteins that 118 promote tumour growth and provide diagnostic biomarkers. Nat Cell Biol 10, 1470–1476 (2008). 27. Zomer, A. et al. Exosomes: Fit to deliver small RNA. Commun. Integr. Biol. 3, 447–50 (2010). 28. Rabinowits, G. et al. Exosomal microRNA: a diagnostic marker for lung cancer. Clin. Lung Cancer 10, 42–6 (2009). 29. Taylor, D. D. & Gercel-Taylor, C. MicroRNA signatures of tumor-derived exosomes as diagnostic biomarkers of ovarian cancer. Gynecol Oncol 110, 13–21 (2008). 30. Raposo, G. et al. B lymphocytes secrete antigen-presenting vesicles. J. Exp. Med. 183, 1161–72 (1996). 31. Caby, M. P., Lankar, D., Vincendeau-Scherrer, C., Raposo, G. & Bonnerot, C. Exosomal-like vesicles are present in human blood plasma. Int Immunol 17, 879– 887 (2005). 32. Théry, C., Amigorena, S., Raposo, G. & Clayton, A. Isolation and characterization of exosomes from cell culture supernatants and biological fluids. Curr. Protoc. Cell Biol. Chapter 3, Unit 3.22 (2006). 33. Yakimchuk, K. Exosomes: isolation methods and specific markers. Mater. Methods (2015). 34. Lötvall, J. et al. Minimal experimental requirements for definition of extracellular vesicles and their functions: a position statement from the International Society for Extracellular Vesicles. Journal of Extracellular Vesicles 3, (2014). 35. Morelli, A. E. et al. Endocytosis, intracellular sorting, and processing of exosomes by dendritic cells. Blood 104, 3257–3266 (2004). 36. Denzer, K. et al. Follicular dendritic cells carry MHC class II-expressing microvesicles at their surface. J. Immunol. 165, 1259–65 (2000). 37. Clayton, A. et al. Adhesion and signaling by B cell-derived exosomes: the role of integrins. FASEB J. 18, 977–9 (2004). 38. van Niel, G., Porto-Carreiro, I., Simoes, S. & Raposo, G. Exosomes: a common pathway for a specialized function. J. Biochem. 140, 13–21 (2006). 39. Lakkaraju, A. & Rodriguez-Boulan, E. Itinerant exosomes: emerging roles in cell and tissue polarity. Trends Cell Biol. 18, 199–209 (2008). 119 40. Johnstone, R. M. Exosomes biological significance: a concise review. Blood Cells. Mol. Dis. 36, 315–21 41. Giddings, J. C., Yang, F. J. & Myers, M. N. Flow-field-flow fractionation: a versatile new separation method. Science (80-. ). 193, 1244–1245 (1976). 42. Havard, T. integration of postnova field flow fractionation systems. (2014). Available at: www.postnova.com. (Accessed: 10th March 2015) 43. Bott, J. & Tadjiki, S. Field-flow Ffractionation supporting consumer safety evaluation of silver nanoparticle applications in food packaging polymers American Laboratory. (2013). Available at: http://www.americanlaboratory.com/913-Technical-Articles/134639-Field-Flow Fractionation-Supporting-Consumer-Safety-Evaluation-of-Silver-NanoparticleApplications-in-Food-Packaging-Polymers/. (Accessed: 10th March 2015) 44. Dragovic, R. a. et al. Sizing and phenotyping of cellular vesicles using nanoparticle tracking analysis. Nanomedicine 7, 780–8 (2011). 45. Boyd, R. D., Pichaimuthu, S. K. & Cuenat, A. New approach to inter-technique comparisons for nanoparticle size measurements; using atomic force microscopy, nanoparticle tracking analysis and dynamic light scattering. Colloids Surfaces A Physicochem. Eng. Asp. 387, 35–42 (2011). 46. Nanoparticle Tracking Analysis (NTA): Nanosight. Available http://www.nanosight.com/technology/nanoparticle-tracking-analysis-nta. (Accessed: 10th March 2015) 47. Inge Johanna Stijleman. Exosomal non-coding RNAs for monitoring breast Cancer patients. (University of Utah, 2016). 48. Sokolova, V. et al. Characterisation of exosomes derived from human cells by nanoparticle tracking analysis and scanning electron microscopy. Colloids Surf. B. Biointerfaces 87, 146–50 (2011). 49. Mathivanan, S., Fahner, C. J., Reid, G. E. & Simpson, R. J. ExoCarta 2012: database of exosomal proteins, RNA and lipids. Nucleic Acids Res. 40, D1241-4 (2012). 50. Kalra, H. et al. Vesiclepedia: a compendium for extracellular vesicles with continuous community annotation. PLoS Biol. 10, e1001450 (2012). 51. Chen, X. et al. Characterization of microRNAs in serum: a novel class of biomarkers for diagnosis of cancer and other diseases. Cell Res 18, 997–1006 (2008). at: 120 52. Lee, Y. S. & Dutta, A. MicroRNAs in cancer. Annu Rev Pathol 4, 199–227 (2009). 53. King, H. W., Michael, M. Z. & Gleadle, J. M. Hypoxic enhancement of exosome release by breast cancer cells. BMC Cancer 12, 421 (2012). 54. Kucharzewska, P. & Belting, M. Emerging roles of extracellular vesicles in the adaptive response of tumour cells to microenvironmental stress. J. Extracell. vesicles 2, 20304 (2013). 55. Laulagnier, K. et al. Characterization of exosome subpopulations from RBL-2H3 cells using fluorescent lipids. Blood Cells. Mol. Dis. 35, 116–21 (2005). 56. Smith, Z. J. et al. Single exosome study reveals subpopulations distributed among cell lines with variability related to membrane content. Journal of Extracellular Vesicles 4, (2015). 57. Oksvold, M. P. et al. Expression of B-Cell surface antigens in subpopulations of exosomes released from B-Cell lymphoma cells. Clin. Ther. 36, 847–862.e1 (2014). 58. Ferguson, S. W. & Nguyen, J. Exosomes as therapeutics: the implications of molecular composition and exosomal heterogeneity. J. Control. Release 228, 179– 90 (2016). 59. Bobrie, A., Colombo, M., Krumeich, S., Raposo, G. & Théry, C. Diverse subpopulations of vesicles secreted by different intracellular mechanisms are present in exosome preparations obtained by differential ultracentrifugation. Journal of Extracellular Vesicles 1, (2012). 60. Kapanidis, A. N. et al. Fluorescence-aided molecule sorting: analysis of structure and interactions by alternating-laser excitation of single molecules. Proc. Natl. Acad. Sci. U. S. A. 101, 8936–41 (2004). 61. Paulaitis, M., Guzman, N., Agarwal, K. & Saji, M. Abstract P3-09-04: Exosomespecific microRNA signatures in combination with characteristic surface markers on the circulating exosomes themselves provide new insights into the EMT. Cancer Res. 70, P3-9-4-P3-9–4 (2011). 62. Petersen, K. E. et al. A review of exosome separation techniques and characterization of B16-F10 mouse melanoma exosomes with AF4-UV-MALSQELS-DLS-TEM. Anal. Bioanal. Chem. submitted, (2014). 63. Oh, S. et al. Miniaturized asymmetrical flow field-flow fractionation: application to biological vesicles. J. Sep. Sci. 30, 1082–7 (2007). 121 64. Kang, D., Oh, S., Ahn, S.-M., Lee, B.-H. & Moon, M. H. Proteomic analysis of exosomes from human neural stem cells by flow field-flow fractionation and nanoflow liquid chromatography-tandem mass spectrometry. J. Proteome Res. 7, 3475–80 (2008). 65. Ashby, J. et al. Distribution profiling of circulating microRNAs in serum. Anal. Chem. 86, 9343–9 (2014). 66. Sitar, S. et al. Size characterization and quantification of exosomes by asymmetrical-flow field-flow fractionation. Anal. Chem. 87, 9225–9233 (2015). 67. Chernyshev, V. S. Determination of biophysical and chemical properties of shortchain molecules and biological nanoparticles. (University of Utah, 2015). 68. Martin E. Schimpf, Karin Caldwell, J. C. G. Field-flow fractionation handbook. (Wiley-Interscience, 2000). 69. Shi, J. & Jian. Considering exosomal miR-21 as a biomarker for cancer. J. Clin. Med. 5, 42 (2016). 70. Wu, K., Li, L. & Li, S. Circulating microRNA-21 as a biomarker for the detection of various carcinomas: an updated meta-analysis based on 36 studies. Tumor Biol. 36, 1973–1981 (2015). 71. Creemers, E. E., Tijsen, A. J. & Pinto, Y. M. Circulating microRNAs: novel biomarkers and extracellular communicators in cardiovascular disease? Circ. Res. 110, 483–495 (2012). 72. Hergenreider, E. et al. Atheroprotective communication between endothelial cells and smooth muscle cells through miRNAs. Nat. Cell Biol. 14, 249–256 (2012). 73. Vickers, K. C., Palmisano, B. T., Shoucri, B. M., Shamburek, R. D. & Remaley, A. T. MicroRNAs are transported in plasma and delivered to recipient cells by high-density lipoproteins. Nat. Cell Biol. 13, 423–433 (2011). 74. Melnik, B. C. et al. MiR-21: an environmental driver of malignant melanoma? J. Transl. Med. 13, 202 (2015). 75. Li, L. et al. Exosomes derived from hypoxic oral squamous cell carcinoma cells deliver miR-21 to normoxic cells to elicit a prometastatic phenotype. Cancer Res. 76, 1770–1780 (2016). 76. Hannafon, B. N. et al. Plasma exosome microRNAs are indicative of breast cancer. Breast Cancer Res. 18, 90 (2016). 77. Oosthuyzen, W. et al. Quantification of human urinary exosomes by nanoparticle 122 tracking analysis. J. Physiol. 591, 5833–42 (2013). 78. DeBerardinis, R. J. Is cancer a disease of abnormal cellular metabolism? New angles on an old idea. Genet. Med. 10, 767–77 (2008). 79. Garbett, N. C. et al. Detection of cervical cancer biomarker patterns in blood plasma and urine by differential scanning calorimetry and mass spectrometry. PLoS One 9, e84710 (2014). 80. Zapf, I. et al. DSC analysis of human plasma in breast cancer patients. Thermochim. Acta 524, 88–91 (2011). 81. Douglas C. Montgomery. Design and analysis of experiments. (John Wiley, 2005). 82. Osteikoetxea, X. et al. Improved characterization of EV preparations based on protein to lipid ratio and lipid properties. PLoS One 10, e0121184 (2015). 83. Lee, M., Ban, J.-J., Im, W. & Kim, M. influence of storage condition on exosome recovery. biotechnol. bioprocess Eng. 21, 299–304 (2016). 84. Zhou, H. et al. Collection, storage, preservation, and normalization of human urinary exosomes for biomarker discovery. Kidney Int. 69, 1471–6 (2006). 85. Steinbichler, T. B., Dudás, J., Riechelmann, H. & Skvortsova, I.-I. The role of exosomes in cancer metastasis. Semin. Cancer Biol. (2017). doi:10.1016/j.semcancer.2017.02.006 86. Ziaei, P., Geruntho, J. J., Marin-Flores, O. G., Berkman, C. E. & Grant Norton, M. Silica nanostructured platform for affinity capture of tumor-derived exosomes. J. Mater. Sci. 1–10 (2017). doi:10.1007/s10853-017-0905-0 87. Trivedi, M., Talekar, M., Shah, P., Ouyang, Q. & Amiji, M. Modification of tumor cell exosome content by transfection with wt-p53 and microRNA-125b expressing plasmid DNA and its effect on macrophage polarization. Oncogenesis 5, e250 (2016).