Extracellular vesicles—An omics view

Abstract

Extracellular vesicles (EVs) are taking a central stage in intercellular communication, as conserved signaling mediators across species and kingdoms. In fact, there has been an emergence in the understanding and expansion of EVs in diverse fields including cell biology, biomedical sciences, immune regulation and vaccine development, biomarker discovery, and disease diagnosis/monitoring. To enhance and modify their form, function, and therapeutic utility, the field has expanded to modify EVs using various bioengineering strategies, which collectively have garnered significant clinical interest given their potential for drug delivery and therapeutic intervention. As heterogeneous, phospholipid membrane-enclosed structures. EVs affect the functions of other cells through their surface proteins, complex encapsulated cargo molecules (including proteins and RNAs), and select lipids and glycans. Moreover, EVs are a potential source of disease-associated biomarkers for diagnosis, composed of a molecular fingerprint of the releasing cell type (i.e., tumor-specific molecules), enabling a molecular analysis of practically all organs in the body.

The form and function of EVs is marked by their proteome. Proteomic studies have generated new knowledge in the EV field, with exceptional insights into cargo sorting mechanisms [1, 2], EV heterogeneity and the subtypes and sub-populations of EVs and particles [3-7], genesis [8], surfaceome [9, 10], interaction network [11], intracellular trafficking pathways [12], release (including organs [13, 14]/tissues [15]), targeting/localization [16], uptake [17] and function [18, 19], to identifying specific marker proteins [20]. Their interrogation of EVs in biofluids has also highlighted their diagnostic potential [21-23] and therapeutic targets [24], establishing the role of EVs in health and disease.

This issue reveals a new understanding of EVs that influences their diverse signaling functions. The studies outlined in this issue (26 articles, covering 16 research studies) shed light on new potential players in intercellular signaling, from cells, tissues, bacteria, and even platelets, and uncovers the functional tasks and diagnostic potential accomplished by the cargo of these extracellular membranous structures.

Platelet-derived EVs (pEVs) represent the most abundant EV type in the circulation in healthy humans. Moon et al., [25] examined the proteome dynamic of pEVs in the context of different physiological platelet agonist to induce platelet activation, uncovering the mode of platelet activation as a direct impact on the proteome landscape. Using agonists representative of the varied activation states of platelets within a thrombus, the correlation study revealed an upregulation of various classes of proteins including those involved in regulation of complement and coagulation, let alone platelet activation in platelets and their derived EVs. The study further identified that the proteome of pEVs occurs in the absence of de novo protein synthesis, indicating that pEV protein selectivity is a specific and active process. Implications for these findings could understand mechanisms regulating the formation of the pEVs potentially leveraged to modify the cargo of pEVs for therapeutic function such as anticoagulation or fibrinolysis, or as potent effectors of tissue regeneration and cell function.

EVs are important complex signaling players in the tumor microenvironment, interacting not only with diverse cell types to regulate and alter their functions, but further in their dynamic interaction with the extracellular matrix (ECM). Jiminez identified specific cells could synthesize and secrete quantities of ECM that are carried primarily by small EVs [26]. The study identified that protein cargo associated small EVs comprised the ECM factors in cancer cells enhanced adhesion and survival of stem-like, tumor propagating cells, where the ECM-carrying EVs from cancer cells induce a morphological shift from cell-cell to cell-substrate adhesion. The study identified both epithelial (e.g., laminins) and stromal (e.g., fibronectin) types of ECM are enriched on cancer EVs and function to promote stem-like cell adhesion and survival. Indeed, ECM and adhesion receptors such as integrins are frequent cargoes of small EVs in benign/non-disease and cancer cell types and have been previously implicated in driving cancer aggressiveness and metastasis.

Microglia play a crucial role in maintaining brain homeostasis and are involved in various functions essential for the CNS. Niu et al., [27] performed characterization of human microglia EVs under both pro-inflammatory and anti-inflammatory conditions to understand how EVs from microglia are involved in various brain diseases based on their signature profile. Of note, various protein subsets associated with the cell surface, fibrinolysis, and receptor binding were linked with cell-induced inflammation, suggesting these proteomic changes in expression may reflect local inflammatory states within the brain microenvironment.

Understanding what subset of cytoplasmic proteins selectively traffic to different EV populations, especially endosomally-derived exosomes, in the context of different biological systems, and how EV cargo impacts on recipient cell functionality is a key area of research in the EV field. In this issue, Simpson and colleagues reported a comprehensive comparative label-free MS-based protein analysis of distinct membrane-encapsulated EV populations—exosomes, microparticles, and midbody remnants—released from human cancer cells using a combination of differential ultracentrifugation and isopycnic iodixanol density centrifugation [28]. The EV analytic pipeline identified 39 human cancer-associated genes in EVs. Subtype analysis revealed selective composition packaging of known EV protein markers in exosomes, metabolic enzymes in microparticles, while centralspindlin complex proteins, nucleoproteins, splicing factors, RNA granule proteins, translation-initiation factors, and mitochondrial proteins as cargo in midbody remnants. Further, Suwakulsiri, Simpson, and colleagues performed a comparative transcript profiling analysis of the same sample preparations used for protein profiling [29], identifying new knowledge in RNA transcripts (protein-coding, lncRNA, pseudogene transcripts) and fusion genes selectively enriched in these different EV classes. In the context of EV biology, this study identifies RNA and fusion gene compositions within specific types of purified EV subtypes (e.g., (CDK6-ATIP1B2, ADD3-PARN), that could impact on oncogenic EV function, and detection and monitoring using EV-based RNA/ fusion gene candidates.

The study by Barman et al., [30] explored the protein repertoire of a specific subset of small EVs generated at endoplasmic reticulum (ER) membrane contact sites (MCSs) and that are highly enriched in RNAs. Using a combination of differential centrifugation and cushion density gradient purification, this study identified that VAP-A, a vesicle associated protein, regulates a subpopulation of EVs originating at ER MCS and identify protein machineries that are preferentially sorted. The study identified proteins from several classes of RNA machineries, including spliceosomes, lowly expressed in cell-derived EVs genetically modified to limit the expression of VAP-A. These data suggest that dynamic regulation of these protein machineries at ER MCS are involved in the sorting of RNA-RNA binding protein (RBP) complexes into EVs. Indeed, this study focused on a small EV type which, comprised RNA-RBP complexes which may be derived from RNA processing machineries in the cell that include many of the small RNAs found in small EVs, while for larger RNAs, especially full-length mRNAs > 1 kb, often identified predominantly in larger sized EVs [31].

Cho and colleagues [32] provided an in-depth review on EVs from programmed cell death, covering key concepts of EV heterogeneity associated with cell death, proteomic profiling pipelines including cell source and associated protein markers, as well as opinion on proteomics on cell function and regulation during programmed cell death. Su et al., [33] provided a detailed review covering the importance of lipids in EV-related omic studies. Lipidomics is an omics approach to comprehensively study lipid profiles in biological samples, such as plasma, serum, urine, and tissue specimens. This review covers the structural importance of lipid types/classes in EV biology. Moreover, lipidomic analyses are useful for identifying novel lipid biomarkers, especially for various metabolic and malignant diseases in humans. Su et al., discuss the brain-derived EV lipid profile as an approach to understand the progression and pathogenesis of various brain-related disorders from tissue and biofluid source, and raise challenges that are present in current pipelines for lipidomic discovery studies and clinical applications.

In-depth, reproducible, and robust quantification of proteins is critically important for the field of EVs, including EV-based proteomic studies. Cross et al. [34], presented a rapid profiling workflow of small EVs from ultralow samples. This study overcame a key challenge in the field with relation to proteomic investigation of EVs—namely the limitation in sample availability, requiring upscaled EV production from cell culture or biofluids, limiting applicability to lower yield EV sources. This comparative proteomic pipeline optimized various metrics associated with the EV proteome—namely resolving sample preparation methods, short chromatography lengths, and data-independent acquisition (DIA). This refined DIA-based MS approach combined robust single-pot, solid-phase-enhanced sample preparation with temporally optimized enzymatic digestion and short chromatography gradients. This allowed a 20–40-fold less sample input needed to resolve the EV proteome (<1 µg starting EV protein), with sample preparation/quantification/acquisition within a single day. This ease-of-use workflow addresses the pressing need to capture precise and comprehensive proteomes of EVs from ultralow sample quantities, without compromising depth and accuracy. This adaptability facilitates the characterization of EVs from different cell sources to determine known EV biology, particularly where sample availability is constrained.

In recent years there has been considerable interest in the potential application of EVs from blood and other biofluid and tissue sources as a liquid biopsy to understand physiology and disease. Recently the fields capacity to derive EVs from extracellular fluid within an organ has been shown, including the heart [13] and various tissues [35]. This novel capacity to isolate EVs from the extracellular space within an organ provides both an enriched source of EVs originating from that organ and insights regarding alternate EV populations that the organ may be exposed to. In this issue, Useckaite [36] performed proteomic assessment of tissue derived EVs from liver tissue sections (liver tissue extracellular space), showing distinct proteome relative to source tissue and intestinal tissue. Such findings present a pipeline to derive organ enriched EV markers directly from circulation originating from the liver to function as a minimally invasive liquid biopsy, let alone the impact of proteins and factors not derived from the source tissue. Importantly, such an approach employed both targeted and untargeted MS as tools to determine the relative abundance of protein markers of EVs, let alone their comprehensive proteome coverage of each sample, respectively.

Lane et al., [37] explored cellular and EV-associated proteome and transcript landscapes in the context of breast cancer utilizing various established cell lines. The core finding of this study was that key molecular features of various cell lines were associated with their derived EV composition—that is, ER and HER2 signaling pathway components in the originating cell reflected EV-associated transcripts and/or proteins based on cell origin. This study does factor in several advances in the field of EV-associated diagnostics and the reflection in cell and EV composition, where RNA sequencing and proteome analysis of livers from healthy/tumor models, revealed not only transcriptional/proteomic changes from liver, but further derived EVs [38].

In the context of tissue derived EVs, and overcoming the inherent challenges in understanding tumor heterogeneity, Blenkiron et al., (pmic.202300055) performed proteome profiling of EVs isolated directly from frozen biobanked endometrial cancers. Although from frozen tissues, the study addressed a pipeline that allowed a tissue tumor microenvironment to be profiled from source tissue, providing impact of obesity and associated comorbidities on the proteome landscape of EVs in the endometrioid subtype, and endometrial cancer development. Combining in depth EV-based characterization and SWATH-based acquisition, this study revealed likely co-isolation of other non-vesicular particles in the EV enriched fraction, likely attributed to perturbation in cell membranes during storage.

EVs hold great promise as therapeutic modalities due to their endogenous characteristics, that offer stability and facilitate crossing of biological barriers for delivery of molecular cargo to cells, acting as a form of intercellular communication to regulate function and phenotype. Despite remarkable utility of native/biological EVs, their properties and therapeutic efficacy can be improved through various bioengineering approaches for therapeutic potential [39]. Here, EVs can be engineered to harbor specific pharmaceutical content, enhance their stability, and modify surface epitopes for improved tropism and targeting to cells and tissues in vivo. In this issue, a technical challenge associated with scalable generation of native EVs (including those formed through either endosomal sorting or derived from the plasma membrane) to overcome their impedance with clinical utility was addressed. Here, Poh and colleagues [40] employed a rapid, sequential membrane extrusion-based methodology to generate artificial EVs (termed nanovesicles, NVs) from cells from early-stage human embryos. The study compared various metrics in generation/yield, as well as proteome composition; NVs retain features of their cell source, including the enrichment of pro-implantation and embryotropic factors and regulators of endometrial receptivity. Further, like natural EVs, these NVs carry known pro-implantation factors, reprogram the endometrium to enhance hTSC spheroid attachment, and promote embryo adhesion and outgrowth on fibronectin matrix. This study highlights the functional potential of NVs in enhancing embryo implantation and given their rapid and scalable generation, amenable to clinical utility.

Subsequently, a further study by Poh expanded this approach further—to load a signaling growth factor (HB-EGF) into trophectodermal cell-derived nanovesicles (NV^HBEGF) [41]. HB-EGF is a critical signaling mediator during embryo implantation to the maternal endometrium. Using a combination of functional studies with phosphoproteomics and proteomics analyses, this study revealed that NV^HBEGF short-term signaling and long-term reprogramming capabilities on target endometrial cells. This two-phase analysis therefore not only derived a scalable therapeutic strategy to derive nanovesicles directly from cells, but through an efficient bioengineering approach, could modify nanovesicles to induce multifaceted signaling and cellular events in target cells. NVs thus represent a feasible and adaptable method of large-scale generation of therapeutic vesicles for tuning endometrial phenotype and function.

EVs hold the potential for potent immunomodulation, both in eliciting and suppressing immune response. Bacterial outer membrane vesicles (OMV) are spherical lipid bilayer nanostructure naturally secreted by bacteria, which comprise diverse functions including intracellular and extracellular communication, horizontal gene transfer, molecular transfer to host cells, and eliciting an immune response in host cells. In this issue, Dhital et al., and Johnston et al., revealed new insights into the composition and function of OMVs from distinct source. Dhital et al., [42] identified that the different OMVs from N. gonorrhoeae  (derived from different sites of infection: pharynx, cervix, vagina, and urethra) shared a conserved set of proteins and lipids, including major outer membrane proteins that have been previously been identified to important during infections. Further, differences in proteome composition from these clinical isolates impacted altered immune-related function observed in cell death signaling and immune responses in macrophages.

Johnston et al., [43] revealed the dynamic production, composition, and proteome impacted by growth conditions of H. pylori OMVs. Of the  751 proteins common to both bacteria and OMVs, 9 unique outer membrane proteins to OMVs. Of note, 84% of proteins identified across different conditions were conserved in OMV proteome. These findings provide new knowledge on the cargo packaging and selectivity of the OMV proteome during acidic and neutral pH growth conditions and how H. pylori may contribute to its survival, expansion and pathogenesis in a low pH gastric environment.

The field of EVs is a rapidly growing area of basic, applied, and biomedical research, with the identification of new types of EVs, their biology and functions, as well as the development of novel approaches to purify, characterize, and understand EVs. This research topic has covered a number of cutting-edge discoveries in this field—and importantly, how proteomics is advancing key questions and reporting guidelines [44] in the field. The field has several challenges to address in the context of EV biology, function, and clinical translation, including heterogeneity (in sample and EV), contamination from non-EV particles and factors, low-abundance proteins (i.e., tissue leakage and signaling factors), analysis and understanding of protein modifications (e.g., glycome [45]), interaction of mutli-omics and interpretation (i.e., lipid:protein interactions [46], protein:metabolite interactions [47]), and data analysis/informatics.

Undoubtedly, the future direction of EV proteomics encompasses several exciting areas such as unravelling the spatial and temporal characteristics of proteins, single-cell proteomics to understand the secreted content in a dynamic state, high-throughput protein profiling, and achieving subcellular-level resolution in proteomics. We predict an innovative and bright future for expanding the application of mass spectrometry proteomics to research in the EV biology community. We thank all contributors to this issue research topic and the referees for their prompt and in-depth reviews.

Kind regards,

David W. Greening, PhD

Head, Molecular Proteomics

Baker Heart and Diabetes Institute, Australia

Alin Rai, PhD

Group Leader, Molecular Proteomics

Baker Heart and Diabetes Institute, Australia

Richard J. Simpson, PhD, FATSE

Distinguished Professor

Department of Biochemistry and Chemistry

La Trobe Institute for Molecular Science

La Trobe University, Australia