The role of extracellular vesicles-associated proteins markers in prostate cancer: a review

Abstract

Prostate cancer (PCa) is in the top three most common cancers among the world’s male population, and its incidence has been increasing over the years. However, the primary screening and diagnostic tools still rely on serum prostate-specific antigen (PSA) and digital rectal examination (DRE), both of which have been inconclusive. Raised serum PSA and abnormal DRE findings require an invasive prostate biopsy, which has inherent surgical risk. This review aims to provide a comprehensive understanding of the biological production and properties of extracellular vesicles (EVs), which are released by all cell types, including cancer cells, and carry proteins and other biomolecules from their parent cells. EVs are becoming increasingly popular as possible biomarker candidates to improve the accuracy of diagnosing many diseases. Growing data suggest that EVs are crucial for cellular communication and tumour progression via multiple signalling pathways, including epithelial-to-mesenchymal transition (EMT), migration, and invasion, making them suitable for tracking PCa growth and metastasis. Furthermore, choosing the appropriate EV isolation method is essential to ensure accurate diagnosis. More clinically relevant, this review also identifies potential EV protein biomarkers derived from urine, serum, and tissue samples from PCa patients.

Introduction

Extracellular vesicles (EVs) were first noted in the mid-20th century, long before they were recognised for their biological significance. Among the earliest discoveries, EVs were first described in 1946, when Chargaff and West identified a platelet-derived minute particle, which they speculated might be a procoagulant in normal plasma, isolated by high-speed centrifugation [1]. This phenomenon was later described by Wolf in 1967 as “platelet dust,” which contains abundant phospholipids [2]. Shortly thereafter, an electron micrograph of the murine epiphyseal plate revealed membrane-bound vacuole-like vesicular structures within the cartilage matrix, which varied in size and shape and contained diverse materials believed to be involved in bone calcification [3].

As a consequence, additional findings on vesicles released by many cell types were reported. Among others, EVs were observed as membrane fragments and particles released from human rectal microvillous adenoma and liver cells, respectively [4]; as a “virus-like” structure detected around human primary cancers and cancer cell lines; and as “contamination” from foetal bovine serum [5]. The confusion was later clarified as vesicles of multivesicular bodies (MVBs) and EVs by Dalton (1975) [6], who also suggested that they were either part of normal cell structure or decomposition products of normal cells.

A significant advancement occurred between 1978 and 1983, when vesicles released from human prostatic cells were identified in prostatic fluid and seminal plasma. These vesicles were later termed “prostasomes” and were speculated to be exported out into the extracellular environment by exocytosis and/or diacytosis [7]. During the same period, from 1983 to 1986, a detailed study of the incorporation of the plasma membrane, or endocytosis, into multiple vesicular endosomes (MVEs) was conducted, and the vesicles were finally discharged extracellularly by exocytosis, retaining a functionally intact transferrin receptor, in sheep and rat reticulocytes during reticulocyte maturation [8, 9].

Not long after, this work led to the introduction of the term “exosome” in 1987, initially proposed as cellular waste, marking a significant step toward defining EVs as a distinct entity [10]. Since then, numerous studies have advanced EV research, contributing to the progression of various human diseases and creating biotechnological opportunities. As a consequence, multiple terminologies and nomenclatures for EVs have been used to define and accommodate individual scientific findings [11].

In recent years, research on EVs has expanded markedly, highlighting their importance in mediating cell-to-cell communication. EVs are now understood to transport a broad spectrum of biomolecules, including proteins, lipids, various RNA species, and metabolites, thereby influencing cellular functions in both healthy and diseased tissues. Increasing evidence indicates that EVs contribute to several key processes in cancer biology, such as enhancing tumour growth, supporting metastatic spread, shaping immune responses, and promoting resistance to therapy. Consequently, EVs have emerged as promising sources of disease-associated information and potential targets for novel therapeutic strategies.

Prostate cancer (PCa) is a malignant growth of prostate cells that metastasizes to nearby tissues, primarily affecting men [12]. Globally, PCa accounts for 7% of new cancer cases, with an estimated incidence of 1,414,300 men and 375,300 deaths reported in 2020 [13]. Prostate-specific antigen (PSA), which is specific for prostate tissue and semen, has been examined and recognised by several studies as a means of diagnosing prostate malignancy since the 1960 s [14]. In the 1980 s, PSA was identified as a member of the human kallikrein gene family and named human kallikrein 3 (hKLK3), located on chromosome 19 and encoding a protein involved in the detection and monitoring of PCa [15]. PSA is a serine protease synthesized by the prostate gland that is responsible for semen liquefaction [16]. Circulating total PSA (tPSA) is present in the bloodstream in several molecular forms, including complex PSA and free PSA; free PSA is implicit in 16% of tPSA and comprises three distinct PSA precursor isoforms: proenzyme PSA (proPSA), benign PSA (BPSA), and inactive PSA (iPSA) [17, 18]. However, PSA has low specificity for detecting PCa, necessitating biopsies to confirm or rule out malignancy, which increases the risk of biopsy complications or overdiagnosis in cases of a positive biopsy [19].

To compensate for the limitations of PSA assays and provide greater assurance for PCa diagnosis and treatment, alternative PCa biomarkers, such as circulating tumour cells, microRNAs, and EVs, have been assessed in recent studies [20]. It has been reported that EVs are involved in various physiological and pathological processes, including cancer, and that their interaction with vascular endothelial cells influences tumour metastasis [21]. EVs consist of multiple subtypes that differ primarily in cellular origin and size, including apoptotic bodies, shedding microvesicles, and exosomes [22].

Numerous previous studies have also investigated the function of PCa EV proteins in diagnosis and therapy. However, the role of circulating EVs in PCa tumorigenesis and cancer progression remains poorly understood. Therefore, the present review synthesises current evidence on the biological roles of circulating EVs in PCa, evaluates their potential as biomarkers for early detection and disease stratification, and highlights emerging opportunities for their therapeutic application.

Biogenesis of extracellular vesicles

EVs are divided into two main subdivisions based on size and release from parental cells: exosomes and ectosomes (microvesicles). The size of the former ranges from 30 to 150 nm and is released by exocytosis, while the size of the latter varies from 50 to 1000 nm and is expelled by outward membrane budding [23]. As a consequence, designating EVs as exosomes or ectosomes in research publications must be substantiated by biogenesis evidence, as highlighted by the MISEV2023 guidelines [24]. In general, the EV synthesis-and-release mechanism comprises major sequential events, including initiation of the endocytic pathway at the plasma membrane and formation of intraluminal vesicles (ILVs), which together establish MVBs [25]. Eventually, MVB either releases its contents extracellularly as an EV or fuses itself with a lysosome for degradation. By technical definition, an EV is generated by the inward budding of the late endosomal membrane, which is then referred to as an ILV [26]. It engulfs cytosolic fluid and its endocytosed cargo and is finally excreted into the extracellular milieu. On the other hand, EVs such as microvesicles and apoptotic bodies are directly formed by outward budding of the plasma membrane from living and dying cells, respectively, into the extracellular environment [27]. Overall, EV biogenesis is divided into several stages, which will be further clarified in detail in the following subtopics: (a) MVB formation; (b) ILV formation and cargo-sorting process: Endosomal sorting complex required for transport (ESCRT) & Non-ESCRT dependent; (c) The fate of MVB; (d) MVB translocation and transport; and (e) Multivesicular docking and ILV release.

Multivesicular bodies (MVB) formation

Endocytosis, or invagination of the plasma membrane, internalises fluids, solutes, macromolecules, and particles, including plasma membrane components, which are ultimately delivered by membrane scission, generating vesicles or vacuoles in the cytoplasm [28]. Consequently, the vesicles generated at the plasma membrane fuse with each other to form primary endocytic vesicles or transport vesicles, which then unite among themselves to form a bigger vesicle called the early endosome (EEs) (Fig. 1) [28]. This process results in EEs that contain a wide range of cargo, including solutes, receptors, lipids, and even pathogenic agents. Therefore, EEs are the first sorting stations for recycling molecules and regulating cell signalling [29]. Sorting proteins and lipids from EEs back to the trans-Golgi network (TGN) for degradation and plasma membrane for recycling marks the maturation of EEs to become late endosomes (LEs) [30]. Interestingly, multivesicular parts of EEs will also detach from EEs as endosomal carrier vesicles and mature, carrying cargo to LEs, the second sorting station [29]. Matured LEs, also known as MVB, contain numerous ILVs, which are formed through serial endocytotic activities at the limiting membrane of maturing LEs and MVB [28].

Fig. 1.

Extracellular vesicles synthesis and release. 1 Invagination of the plasma membrane and its components, forming a vesicle or endocytic vesicle. 2 Endocytic vesicles are delivered to existing early endosomes (EEs), and 3 coalesce. 4 EEs are maturing and becoming late endosomes (LEs). 5 Serial endocytic activities of LEs to form intraluminal vesicles (ILV) and multivesicular bodies (MVB). 6 MVB is maturing and 7 being transported to fuse with the plasma membrane to excrete ILVs, known as exosomes. 8 While degradative-fated, MVB is directed towards the lysosome for disposal. Meanwhile, ectosomes, or microvesicles, are excreted from cells by outward budding of the plasma membrane (insert) ESCRT: Endosomal Sorting Complex Required for Transport

Intraluminal vesicle (ILV) formation and cargo-sorting process: endosomal sorting complex required for transport (ESCRT) & non-ESCRT dependent

The ESCRT is a cytosolic protein subcomplex that is essential for ILV formation. This machinery comprises ESCRT 0, I, II, and III, along with accessory proteins such as the ATPase vacuolar protein sorting 4 (VPS4), which enables ESCRT complexes to dissociate for recycling [31]. The primary function of this machinery is to induce budding-in of the membrane and assist the scission of the neck of the invaginated membrane from the inside [32]. Additionally, the ESCRT machinery facilitates sorting of ubiquitin-labeled selective transmembrane and cytosolic proteins for lysosomal and proteasomal degradation, respectively [32].

The ESCRT-dependent ILV formation involves five steps. The pathway commences when ESCRT-0, which comprises two subunits, HRS (hepatocyte growth factor-regulated tyrosine kinase substrate) and STAM1/2 (signal transducing adaptor molecule 1/2), recognises, binds to, and retains ubiquitinated proteins at ubiquitin-binding regions along flat endosomal membranes [26, 33]. The HRS-subunit of ESCRT-0 binds to the endosomal lipid phosphatidylinositol 3-phosphate (PI3P) [32], a kind of lipid that is plentiful on the LE membrane (34) (Fig. 2). Then, upon recruitment of ESCRT I & II to ESCRT-0, the complex binds to the cargo and clusters them while the flat membrane starts to invaginate [31]. To be specific, ESCRT-I and ESCRT-II induce the inward budding of the ILVs into the lumen of the MVBs [35]. Among others, Tsg101 (Vps23) and Alix (Bro1) are the subunits of ESCRT-1 [31]. Meanwhile, the ESCRT-II complex consists of 4 subunits: a Y-shaped Vps22, Vps36, and two Vps25 [31].

Fig. 2.

The ESCRT-dependent ILV formation & cargo sorting. Three major events during ILV formation occur concurrently and are assisted by the endosomal sorting complex required for transport (ESCRT) protein components: at the membrane level A, ESCRT recruitment B, and cargo sorting C. ILV: Intraluminal vesicles

Vps20 of ESCRT-III has a strong affinity for Vps25 of ESCRT-II, which explains why ESCRT-II is needed to recruit ESCRT-III [31]. Next, upon ESCRT-III assembly, the deubiquitination machinery is engaged, and ubiquitin is removed from the cargo by deubiquitylating enzymes (DUBs) [36]. At the same time, the invaginated membrane is maturing and becoming a vesicle by forming a neck [31]. After that, the ESCRT 0-I-II-III complexes drive further constriction at the vesicular neck, resulting in sequestration of the cargo into the maturing vesicle [31, 37]. Finally, the narrowed and constricted neck is cleaved from the limiting membrane, forming a completely independent vesicle [31]. The ESCRT complexes are disassembled by the Vps4-Vta1 complex, also known as ESCRT-IV, for recycling, and cargo is sorted within the vesicle [38].

Alternatively, cargo sorting and ILV formation can still be achieved in the absence of ESCRT machinery. This was demonstrated by MVB productions in cells depleted of all four ESCRT subunits [39]. Instead, the process is facilitated by lipids, tetraspanins, and heat-shock proteins [40]. The budding-in of the MVB membrane, which is ESCRT-independent, involves lipid raft-based microdomains (complexes) that consist of either lysobisphosphatidic acid or ceramide (generated by sphingomyelin hydrolysis) [41]. In a study tracing proteolipid protein (PLP) movement in EVs, it was observed that depleting Hrs, Alix, and Tsg101, subunits of the ESCRT machinery, did not affect PLP inward budding or its coexistence with secreted EVs [41]. Apart from PLP, cholesterol was abundant in EV membranes [42].

The fate of multivesicular bodies (MVB)

Whether to be extracellularly excreted as EVs or fused with lysosomes for degradation, the factors that determine the fate of ILVs and MVBs, and their regulation, are still unclear. The mechanism of their formation might determine the balance between the two destinies at the very beginning. The budding-in of the MVB membrane through the ESCRT-independent mechanism involves a lipid raft-based microdomain complex that consists of either lysobisphosphatidic acid or ceramide (generated by sphingomyelin hydrolysis). Both are cone-shaped, facilitating ILV budding into the MVB. The former directs ILVs to lysosomal degradation and is absent from EVs, whereas the latter directs ILVs to EV-mediated excretion [41]. MVBs that are associated with high cholesterol are selected for secretion; on the other hand, the ones with low cholesterol are fated for degradation [43].

Multivesicular bodies (MVB) translocation and transport

Once the fate of MVB has been assigned to secretory or degradative, they will be transported to either the plasma membrane for secretion or the lysosome for degradation of their contents. In any case, the specific, more detailed molecular mechanisms of EV secretion remain unclear. For secretory-fated MVB, they are intracellularly translocated to the plasma membrane based on their communication with actin filaments (microfilaments) and microtubule cytoskeleton [44]. This is consistent with findings that knocking down cortactin, an actin-binding protein, decreased EV secretion [45].

In the same study, live-cell imaging also disclosed that this protein modulates MVB trafficking and docking to the plasma membrane [45]. The mechanism of secretory MVB mobilisation towards the plasma membrane has been widely studied for the past few years. The mobilisation and transport of MVB to the plasma membrane involve a few Ras-associated binding (RAB) protein families, for instance, Rab GTPases that are associated with EV secretion and facilitate several stages in intracellular vesicle translocation and movement along actin and tubulin, membrane trafficking, vesicle budding, and membrane docking into the targeted compartment, hence membrane fusion [27, 40].

Intriguingly, different Rab proteins take part in distinct stages of MVE translocation and docking to the plasma membrane along the endocytic pathway [40] (Fig. 3). Rab11, the first Rab GTPase reported to be involved in EV secretion [46], and Rab35 are associated with early MVEs’ translocation and docking to the plasma membrane. In contrast, Rab 27a and Rab 27b are associated with late MVBs [40]. Rab11 has been shown to be associated with EV release in human leukaemic K562 cells [47], and a decade later, its deficiency in Drosophila S2 cells decreased EV-enriched EV excretion [48]. In contrast, it does not affect EV secretion in cervical cancer HeLa cells [49]. At the same time, Rab35 depletion from the oligodendroglial cell line reduced proteolipid protein-enriched EV excretion, possibly due to the inability to dock at the plasma membrane [50]. Whereas Rab35 protein’s knock-out did not affect EV secretion in Drosophila S2 cells [48].

Fig. 3.

RAB proteins assist in MVB translocation for EVs secretion. Various RAB proteins facilitate the transport of distinct phases of multivesicular endosomes (MVE) to the plasma membrane and are also implicated in the docking of multivesicular bodies (MVB). ESCRT: endosomal sorting complex required for transport; RAB: Ras-associated binding protein

In a study that screened for targeted human Rabs, they found an association between five Rab proteins, Rab 2b, 9a, 5a, 27a, and 27b, and EV excretion [49]. By knocking down those Rabs, EV excretion was reduced, and Rab27a and Rab27b were highly associated with plasma membrane docking [49]. Previous investigations have also identified a connection between Rab27a and EV excretion [44]. Apart from EV excretion, Rab27a also mediates the excretion of soluble factors, such as the pro-metastatic factor MMP-9, in the 4T1 mammary carcinoma cell line, and pro-angiogenic placental growth factor 2, platelet-derived growth factor A, and osteopontin in the B16-F10 melanoma cells [40].

Multivesicular docking and intraluminal vesicle (ILV) release

Upon arrival and docking at the plasma membrane, the MVB must surmount an energy barrier during fusion to release its contents, specifically the ILVs. This necessitates several interactions between protein-lipid and protein-protein [27]. The soluble N-ethylmaleimide-sensitive factor attachment protein (SNAP) receptors, or SNAREs, machinery plays a significant role in the membrane fusion mechanism [51]. Intriguingly, 36 SNARE proteins are expressed in mammals [52]. There are two categories of SNARE proteins: R and Q-SNAREs. R-SNAREs protein lies on the vesicle’s membrane (the donor’s membrane) and is called vesicular SNAREs or v-SNAREs, or vesicular-associated membrane protein (VAMP). At the same time, Q-SNAREs are located on the receiving membrane or targeted membrane and called target SNAREs or t-SNAREs [51]. The combination of those two subtypes of SNARE molecules from opposing membranes leads to the establishment of a four-chain α-helical bundle complex, which is composed of one v-SNARE and three t-SNAREs that bridge the two membranes and consequently promote bilayer fusion [51, 53].

It was reported that the fusion of the secretory lysosome with the plasma membrane was regulated by Ca2 + and the SNARE protein SNAP-23 [54], located at the receiving membrane, lysosomal VAMP8 [55], and VAMP7 [56]. However, further details and precise mechanisms underlying fusion between the MVB and the plasma membrane during EV excretion have not yet been elucidated [27, 47].

Biological characteristics and contents of extracellular vesicles

EVs differ in terms of size, mechanism of liberation [27], and the state of the cells at the time of release (i.e., healthy cells vs. cells undergoing cell death) [57]. Functionally, EVs can be fundamentally divided into three main categories: exosomes, microvesicles, and apoptotic blebs [25, 58]. However, owing to their diversity, various names have been given based on the nature, origin, and characteristics of EVs. Generally, the prefix used in the names given reflects the size (macro or nanovesicles), cells or tissue of origin (prostasome), designated functions (calcifying matrix vesicles, telorosome), or intelligibly being outside the cells (exosome, ectosome) [46]. According to the International EV Society, EVs are categorised by size: small EVs, which range from 30 to 150 nm; microvesicles, whose size is broadly variable between 100 and 1000 nm; and apoptotic blebs (500–1000 nm) [23, 25].

Regardless of EV subtype, EVs are, by definition, membranous vesicles formed by invagination into MVBs, excreted from cells, and that transport parental cell information throughout the body via the extracellular fluid [59]. Unlike in the past, when EVs were initially thought to be homogeneous, advances in discovery have shown that EVs are heterogeneous, comprising subpopulations with distinctive protein compositions and RNA profiles that may differentially affect recipient gene expression [60].

Extracellular vesicle (EV) components: membranous structure

An EV is composed of a bilayer lipid membrane that encloses its cytosolic contents (Fig. 4). The membrane is constructed by lipid rafts such as sphingomyelin, ceramide, phosphatidylserine, and cholesterol [61]. Owing to the way an EV membrane is initially generated, first by invagination of parental cell membrane, including their array of proteins, then second invagination of late endosome membrane that selectively encapsulates specific cytosolic and membranous components of each ILV, making each of them differ from their counterparts, as well as their parents’ membrane compositions [62]. EVs secreted from a diverse range of cells, in comparison to their parental plasma membranes, have their lipid bilayer membrane highly prettified with sphingomyelin, hexosylceramides, and cholesterol, yet lower in phosphatidylcholine and phosphatidylethanolamine [63]. Nonetheless, the most functional component is the set of proteins that span the membrane, adhere to it, and are ready to serve the purpose of EVs. Structurally, there are several types of EV membrane proteins, including transmembrane proteins, which comprise tetraspanin families and signalling proteins; lipid-anchored proteins at the outer and inner surface of the membrane; and the peripheral membrane proteins at the inner side of the membrane, as well as at the surface [64].

Fig. 4.

Components of an EV. An EV is composed of membranous and cytosolic structures. The membranous structure consists of different types of lipid rafts and various proteins. Its cytosolic compartment contains a wide range of RNA and DNA subtypes, as well as soluble and bulky proteins, amino acids, metabolites, and lipids. Diagram constructed based on the concepts and information presented in several articles [61, 64, 71, 72, 243]

Transmembrane proteins from the tetraspanin family: CD9, CD37, CD53, CD63, CD81, CD82, CD151, Tspan8, would be commonly identified as an EV surface biomarker for EV verification [65]. However, tetraspanin families would only perform facilitating roles that involve target cell adhesion, motility, activation, and protein recruitment into EVs [66], for example, by assisting endothelial cell binding and uptake [67]. These undertakings require support from tetraspanin-associated partner proteins, such as protein aggregates from immunomodulatory complexes and antigen-presenting cells, as well as many other adhesion proteins [64]. Among the substantial presence of adhesion proteins in the membrane of an EV are integrins, which have been widely associated with cancer metastasis [68]. Transmembrane signalling proteins likewise project through the thickness of the bilayer membrane, serving as functional receptors and transporters that activate signalling pathways [64]. Epidermal growth factor receptor (EGFR) is a membrane-associated signalling protein expressed in PCa patients’ EVs and is involved in PCa development and metastasis [69]. Other membrane-associated proteins of EVs discovered in PCa include transmembrane lysosome-associated membrane glycoproteins (LAMP-1 and 2B), peripheral membrane proteins heat shock proteins (HSP60, 70, and 90), MHC I and II, lipid-anchored proteins Rabs, and membrane-binding proteins (annexins) [70].

Extracellular vesicle (EV) components: cytosolic compositions

Released EVs carry parental cell information in the form of various biomolecules, including DNA, RNA, transcription factors, bulky and soluble cytosolic proteins, bioactive lipids (prostaglandins, fatty acids, and leukotrienes), amino acids, and metabolites [38, 71]. EVs take in parental cells’ cytosolic DNA, including single and double-stranded DNA (ssDNA & dsDNA), genomic DNA (gDNA), mitochondrial DNA (mtDNA), reverse-transcribed complementary DNA (R-t compDNA), and viral DNA [64, 72, 73]. A diverse array of RNA types engulfed by EVs can be identified during the progression and metastasis of PCa. EV-derived microRNA (miRNA) miR-409-3p was found to be upregulated in PCa patients and has been implicated in PCa growth, EMT, and bone metastasis [74]. In another study, 198 different messenger RNA (mRNA) species were detected from a PCa cell line-derived EVs by microarray-based transcriptome analysis, which classified them into nine functional categories, such as modulation of cell growth and tumour progression, modulation of extracellular space, transporter, cytoskeleton, involved in metabolism, protein turnover, and expression, as signalling molecules and as pseudogenes [75].

Other types of RNA, long non-coding RNAs (lncRNAs), PCa-specific expression, and the EZH2-associated transcript (PCSEAT) were found to be involved in cancer cell proliferation, growth, and migration [76]. Unlike the usual linear RNA with two ends, circular RNA (circRNA) is a long, covalently closed circular structure with no ends, so it is spared exonucleases-mediated degradation and is regarded as sturdy [77]. Xia et al. (2018) [78] identified 1021 circRNA transcripts that were differentially expressed in PCa tissue and cell lines, including SLC19A1, which promote cell proliferation and invasion in PCa. Numerous other RNA subtypes were unveiled carried by EVs, including piwi RNA (piRNA), small-nuclear RNA (snRNA), small-nucleolar RNA (snoRNA), ribosomal RNA (rRNA), transfer RNA (tRNA), transfer RNA (tRNA)-derived small RNAs (tsRNA), Y-RNA, vault RNA (vtRNA), pre-mRNA, and finally repetitive element RNAs as well as fragmented RNAs [70, 79–81].

Another essential component transported by EVs is a broad range of bulk and soluble cytosolic proteins (Fig. 5). Functionally, they include enzymes, growth factors and cytokines, cytoskeletal components, EV biogenesis machinery, RNA-binding proteins, heat-shock protein families, and apoptotic signal transducers [39, 44]. Not all proteins and other cytosolic components will be consumed by all ILVs in late endosomes. To grow and sustain themselves, cancer cells must acquire specific biological capabilities, known as hallmarks, along the way. According to Hanahan & Weinberg (2011) [82], cancer cells should be able to sustain proliferative signals, escape growth suppressors, resist cell death, enable replicative immortality, induce angiogenesis, activate invasion and metastasis, reprogram energy metabolism, and evade immune destruction [82]. A tumour does not exist in isolation within the systemic environment; it is part of a complex, sophisticated network that engages in two-way, synergistic communication among transformed cells, the tumour itself, and surrounding healthy cells to grow. Within this network, apart from the tumour cells themselves, is called the tumour microenvironment (TME), which comprises cellular and acellular components, such as cancer-associated fibroblast (CAF), macrophages, angiogenesis-related cells, immune cells such as T-cell and dendritic cells, as well as extracellular matrix (ECM) and its elements, such as cytokines and proteases [83, 84]. In addition to direct cell-to-cell contact, those cells communicate through various messengers, including secreted mediators such as EVs [84]. Altogether, EV biogenesis, the components, and ultimately, how EVs play roles in cancer development and metastasis are crucially important to be enlightened.

Fig. 5.

Categories and sub-categories of proteins within an EV. In particular, EV proteins are the composition of an EV’s lipid bilayer membrane and its cytosol. The diagram was constructed based on the concepts and information presented in several articles [61, 64, 71, 72, 244]

A plethora of evidence has accumulated, indicating that EVs have been primarily involved and contributed an essential component to the crosstalk between cancer cells and their surrounding stroma. In addition to being secreted at levels more than 10-fold higher than those of normal cells, cancer cells have also been found to be directly involved in tumour initiation, growth, progression, dissemination, and cancer treatment resistance by interacting with the microenvironment through the utilisation of their released cargo [65, 85]. More importantly, the content of cancer-derived EVs would determine changes and fate in the acceptor cells [67]. Numerous experiments have demonstrated the impact of EVs on every stage of carcinogenesis, from the pre-cancerous stage to development, metastasis, and angiogenesis.

Extracellular vesicles (EVs) in pre-cancerous stage

Physiological niches consist of fibroblasts, immune cells, ECM components, endothelial and perivascular cells, numerous cytokine circuits, various growth factors, and cellular secretions [86]. For cancer to develop and progress, regardless of the initiating causes, persistent and relentless genetic mutations, as well as functional modification of the physiological niches, are essential [87]. The transformation of niches into a TME favourable for the seeding of cancerous cells is highly dependent on numerous interacting mediators, such as EVs. EVs are believed to have played diverse roles in many pre-cancerous conditions by supporting chronic inflammatory processes preceding cancer onset [88].

Chronic inflammation caused by an autoimmune disorder that marks the pathophysiology of inflammatory bowel disease (IBD) is associated with the risk of developing colorectal cancer [89]. There is a substantial number of infiltrating polymorphonuclear neutrophils (PMNs) in the intestinal lumen of IBD patients, which release myeloperoxidase (MPO) [90], which in turn produces ROS, a known source of oxidative stress, involved in many aspects of cancer development [91]. One study showed that microparticles released by PMNs, including EVs carrying miR-23a and miR-155, contributed to the accumulation of double-strand breaks (DSBs), which aggravate colonic epithelial injury in IBD [90].

Another instance of chronic inflammation-induced cancer is hepatocellular carcinoma (HCC), whereby the changes in its TME play a significant role in transforming pre-malignant fibrotic and cirrhotic hepatocytes into malignant ones [92]. Chronic inflammation of hepatocytes releases ROS and activates hepatic stellate cells (HSCs), promoting fibrotic changes that lead to cirrhosis [92]. Regulated by hypoxia-inducible factor 1-α (HIF-1α), HSC excreted EVs containing glucose transporter-1 (GLUT-1) and pyruvate kinase M2 (PKM2) that shifted the metabolism and functions of non-parenchymal liver cells, thus promoting liver fibrosis [93]. A new mechanism of Hepatitis C (HCV)-induced liver fibrosis, involving EVs as an important mediator, was discovered when HCV-replicating hepatocyte-derived EVs carrying miR-192 were found to upregulate fibrogenic markers and stimulate transforming growth factor-beta 1 (TGF-β1) in HSCs, thereby promoting their differentiation into myofibroblasts [94]. An immunosuppressive niche in the fibrotic and cirrhotic liver is another favourable situation for inducing HCC development [92].

Chronic Helicobacter pylori infection is highly associated with gastric adenocarcinoma and gastric mucosal-associated lymphoid tissue (MALT) lymphoma [95]. EVs isolated from the serum of patients with chronic H. pylori-induced gastritis activate the pro-inflammatory cytokine interleukin-1α (IL-1α) by upregulating IL-6 receptor expression, thereby promoting a persistent inflammatory state [96]. H. pylori-induced inflammation is the primary factor related to peptic ulcer disease and gastric cancer [97] (Fig. 6).

Fig. 6.

EVs can act as mediators in chronic inflammation, facilitating the progression to malignancy. Types of chronic inflammatory diseases that can progress to cancer that are regulated by EVs [89, 92, 97, 98, 160]. EMT: Epithelial-mesenchymal transition; HCC: Hepatocellular carcinoma; H. pylori: Helicobacter pylori; PBC: Primary biliary cholangitis; PSC: Primary sclerosing cholangitis; ROS: Reactive oxygen species

In addition to that, chronic inflammatory cholestatic diseases, such as primary biliary cholangitis (PBC) and primary sclerosing cholangitis (PSC), are closely related to both cholangiocarcinoma (bile duct cancer) and final-stage liver diseases, such as liver cirrhosis, as well as HCC [98]. In cholestatic diseases, cholangiocyte-derived EVs are abundant in the lncRNA H19, which is not typically expressed in hepatocytes and promotes HSC activation and transdifferentiation during cholestatic disease progression. This clarifies the impact of EVs excreted by cholangiocytes on hepatic cholestatic injury [98].

Extracellular vesicles (EVs) in tumour development

Constant, unremitting interactions between cancer-derived EVs and the TME, and vice versa, largely contribute to tumour growth and development at the primary site. EV cargo, such as RNAs, DNAs, proteins, and metabolites, can affect the fate of their parental cells via autocrine signalling (99). For instance, chronic myeloid leukaemia (CML)-derived EVs containing TGFβ−1 promote tumour growth via anti-apoptotic pathways by binding to its TGFβ−1 receptor to prevent cancer cell termination [100].

DNAs in EVs within the cytoplasm could also influence parental cell survival [59]. EV secretion by parental cells is essential for preserving cellular homeostasis, and if it is hampered, the accumulation of these DNAs will subsequently activate ROS production, thereby induce EV apoptosis and resulting in fragmented DNA throughout the cytoplasm [101]. To survive, cells must prevent this occurrence, which necessitates the expulsion of EVs. Cancer-derived EV cargoes also dictate the fate of neighbouring cells within the TME through paracrine mechanisms that could alter their biological phenotype. These interactions occur not just between cancer cells and normal cells but also between cancer cells and stromal cells, as well as among cancer cells themselves, and vice versa [59]. The miRNAs carried by cancer-derived EVs are more abundant and, at the same time, dysregulated in expression and are significantly tumour-specific [102].

Certain transcription factors in cancer-derived EVs were found to be upregulated, thereby initiating protein expression. For instance, the upregulation of HIF1-α enhances metastatic potential in acceptor cells [103], as well as epidermal growth factor receptor (EGFR), Fas ligand (FasL), and tumour necrosis factor-alpha (TNF-α), both directly and indirectly, thereby causing immunosuppression that enables tumour progression [104]. The upregulation of vascular endothelial growth factor (VEGF), TGFβ−1, matrix metalloproteinase 2 (MMP-2), and urokinase plasminogen activator (uPA) facilitates changes in stromal activity that support tumour-associated neovascularisation, tumour progression, and dissemination [105].

One of the significant cellular modifications within the TME is the epithelial-mesenchymal transition (EMT). The epithelial-like characteristics of the functional cells will disappear, and within no time they acquire mesenchymal-like behaviour, which is invasive, more mobile, more resistant to apoptosis, and associated with massive ECM production [106]. Transformed mesenchymal-like cells trigger a loss of cell-to-cell adherence and less-tight junctions, facilitating metastasis and ECM remodelling [107]. ECM, in general, is composed of various types of collagens, proteoglycans, and glycoproteins that interact with resident cells, dynamically shaping the tissue as a whole [108]. The ECM structure, the interstitial matrix, provides support that surrounds cells, interconnects them within the stroma, and attaches them firmly to the basement membrane [108].

However, in TME, there are a number of pro-fibrotic growth factors and inflammatory factors, such as TGF-α, TGF-β, fibroblast growth factor (FGF)−2, platelet-derived growth factor (PDGF), and epidermal growth factor (EGF), that influence the activation and differentiation of stromal cells and fibroblasts into CAFs that acquire a myofibroblast phenotype and are involved in ECM remodelling [108, 109]. A study indicated that TGF-β was carried by cancer-derived EVs and triggered fibroblast differentiation to myofibroblast with enhanced production of FGF-2 [110]. In turn, a study demonstrated that CAF-derived EVs contain sufficient metabolites and TCA-cycle intermediates to serve as nutrient sources for cancer cells, thereby supporting their survival and growth [146]. This is in line with the discovery of CAF-derived EVs that highly express TGF-β1, which induce EMT and promote invasion and migration capability in ovarian cancer [111]. CAF-derived EVs from breast cancer TME carrying various miRNAs such as miR-181d-5p, miR-500a-5p, miR-21, miR-22, miR-378e, and miR-143 have been shown to be involved in the promotion of breast cancer cell proliferation, invasion, migration abilities, and EMT, as well as inhibition of breast cancer cell apoptotic activities through many different pathways [112, 113].

Despite the immune response against the tumour initially, breast cancer-derived EVs internalised by local and distant macrophages induce the activation of macrophage nuclear factor-КB (NF-КB), resulting in the release of pro-inflammatory cytokines such as IL-6, TNFα, granulocyte-colony stimulating factor (GCSF), and chemokine (C-C motif) ligand 2 (CCL2) that perhaps may, in turn, exacerbate more inflammation that supports tumour growth and metastasis [114]. The interaction between cancer-derived EVs and macrophages occurs through palmitoylated protein ligands on the cancer-derived EVs’ membrane and through macrophage toll-like receptor 2 (TLR2) expression, resulting in the recruitment of tumour-associated macrophages (TAM) to tumour sites and causing necrosis of the solid tumour [114]. Lung carcinoma-derived EVs induce macrophage polarisation towards M2-type attributes, which promote tumour progression by repressing anti-tumour immunity and modulating T cell activation and differentiation [115].

Extracellular vesicles (EVs) in cancer metastasis

Cancer metastasis is undeniably the leading cause of cancer mortality and reflects cancer aggressiveness. Metastasis, or the dissemination of cancer cells, is a dynamic, multistep, and multifactorial process. These steps incorporate (i) local penetration of cancer cells into the neighbouring tissue, (ii) cancer cells’ intravasation, or migration of cancer cells through endothelial cells entering the vasculature, (iii) extravasation or exiting of cancer cells from the vessels, and finally, (iv) settlement and establishment of the cancer cells in a new organ, or colonisation [116]. In addition to the migration of original cancer cells, EVs released by the cancer cells, carrying parental information, could also change the behaviour of distant, surrounding normal cells, and, somehow, EVs could reach potential metastatic loci before the cancer cells do [117]. This is how cancer is preparing its pre-metastatic niche (PMN). It is akin to an individual furnishing a new residence with furniture, dishware, and cutlery before occupancy, ensuring the space is prepared for the eventual arrival of family members. Among distinctive features of PMN are (i) enhanced MMPs undertakings along with the recruitment of collagen, fibronectin, chemoattractant, and growth factors in favouring humongous ECM remodelling; (ii) eminent vascular porosity; and (iii) intensified intrusion of bone marrow-derived cells (BMDCs) and proinflammatory elements to provide an immunosuppressive condition [118, 119].

A study showed that melanoma-derived EVs efficiently prepare sentinel lymph nodes by initiating paracrine signaling that modulates ECM deposition and vascular growth, thereby attracting melanoma cells [120]. In another study, EVs excreted only from CD105 + cancer stem cells of human renal cell carcinoma, rather than the entire tumour cell population, induce angiogenesis, supporting the establishment of PMNs in the lungs [121]. Pancreatic ductal carcinoma-derived EVs triggered metabolism-related signalling pathways, which promoted metastatic behaviour in other pancreatic carcinoma cell lines, induced liver PMN, and enhanced primary tumour growth [122]. Having said that about metabolism, breast cancer-derived EVs loaded with miR-122 inhibit glucose uptake by normal niches in the brain and lungs, utilising downregulating pyruvate kinase, a glycolytic enzyme, in favour of glucose uptake by cancer cells in due course [123]. This exemplifies how distant cancer cells undergo metabolic reprogramming to create a favourable environment for their establishment. The success of PMN formation would also support the preservation and survival of cancer stem cells at the metastatic site, which is crucial for malignant transformation, maintenance, and recurrence [87]. Further development within the metastatic site would govern the survival, dormancy, or overgrowth of disseminated cancer cells [59].

Regarding EV involvement in PCa metastasis, some studies have shown that EV-derived contents directly promote metastatic characteristics in other cells. EVs released by metastatic PCa cell lines PC3 and LNCaP increased proliferation, migration, and invasiveness in non-cancerous prostate epithelial cells. Integrin-alpha 3 (ITG-α3) and integrin-beta 1 (ITG-β1) proteins carried in the EVs were responsible for this change [124]. Integrins are transmembrane receptor proteins composed of an α and a β subunit and are recognised as being associated with advanced PCa and bone metastasis [125]. Another integrin family member, ανβ6 integrin, was identified in PCa-derived EVs, and its transfer to ανβ6-integrin-negative cells increased cell migration and adhesion to two common ECM proteins, latency-associated peptide-TGFβ (LAP-TGFβ) and fibronectin [126]. Previously, ανβ6 integrin was upregulated in human PCa bone metastasis and promoted MMP2 catalytic activity, thereby contributing to osteolytic bone disease [127]. Meanwhile, EV-derived ανβ3 integrin from bone-metastatic PCa cell lines enhanced adhesion and migration of non-tumorigenic recipient cells, and its higher yield in EVs from tumour-bearing mice indicates its clinical significance [128].

Extracellular vesicles (EVs) in angiogenesis

Angiogenesis under physiological conditions involves the generation of new blood vessels from pre-existing vessels to repair damaged vessels, support embryo development, or create collateral vessels to enhance organ perfusion [129]. However, in cancer, angiogenesis becomes structurally and functionally abnormal and chaotic, leading to excessive branching, increased permeability, and an imbalance between angiogenic activators and inhibitors that disrupts normal vascular homeostasis [129, 130] and is part of dynamic metastatic processes. Nevertheless, angiogenesis is the most crucial component of tumour growth and progression, and in its absence, the tumour will not survive beyond 2 mm in size [131].

Vascular endothelial growth factor (VEGF), basic fibroblast growth factor (bFGF), FGF, PDGF, transforming growth factor-β (TGF-β), TNF-α, and IL-8 are among the primary pro-angiogenic factors transported by cancer-derived EVs [87]. One of the most important angiogenic pathways involved in cancer development and progression is VEGF and its receptor (VEGFR) [131]. The pro-angiogenic cytokine VEGF is a key factor in vascular endothelial cell proliferation. In addition to binding to VEGFR-2 on endothelial cells, it also binds VEGFR-1 on macrophages, tumour cells, and fibroblasts [131]. EVs excreted by glioblastoma, multiple myeloma, and melanoma, among others, carry VEGF proteins that modulate angiogenesis in their TME [132]. Hypoxic cancer cells are key regulators of tumour angiogenesis, collaborating with tumour-associated stromal cells and their bioactive secretions [133]. Hypoxia itself, through HIF-1α, induces abundant EV secretion. In colorectal cancer, the β-catenin signalling pathway in endothelial cells is activated by the wingless-type MMTV integration site family, member 4 secreted protein (Wnt4a), which is present in EVs, leading to angiogenesis [134].

EVs secreted by HCC cells contain angiopoietin-2 (ANGPT2), which promotes angiogenesis through a Tie2-independent pathway in recipient human umbilical vein endothelial cells (HUVECs) [80]. Annexin A1 (ANXA1), a pro-angiogenic protein that belongs to the annexin superfamily, transported in pancreatic cancer-related EVs, was shown to induce tubulogenesis [135]. Apart from its pro-angiogenic properties, ANXA1 can also trigger EMT and promote cancer cell metastasis, and it has been found in glioblastoma-derived EVs [135].

As for PCa, a previous study confirmed that PCa-derived EVs contain TGF-β1, which is actively transferred to recipient cells in vitro and is involved in fibroblast differentiation into myofibroblasts [110]. Myofibroblasts are well-known sources of various growth factors, inflammatory cytokines, and chemokines that directly or indirectly modulate angiogenesis, including VEGF, PDGF, CTGF, and interleukin-8 [136]. Lung fibroblasts had been TGF-β1-dependently differentiated into myofibroblasts under the influence of PCa-derived EVs, which are then pro-angiogenic and promote tumour growth. On the other hand, TGF-β1 in PCa-derived EVs altered bone marrow mesenchymal stem cell (BM-MSC) differentiation, which would otherwise have proceeded along an ordinary adipogenic lineage, toward myofibroblastic cells [137]. As a consequence, both EV-treated BM-MSCs and BM-MSC-converted myofibroblasts secreted high levels of pro-angiogenic factors, such as VEGF-A and hepatocyte growth factor (HGF) [137].

In addition, TGF-β1 in PCa cell lines attracted BM-MSC and induced their trans-differentiation into cancer-associated fibroblasts, which, in turn, facilitated monocyte recruitment and angiogenesis-related activity [138]. Some experiments have shown that cancer-derived EVs can polarise macrophages towards the M2 phenotype, which secretes angiogenic factors [139] (Fig. 7).

Fig. 7.

The roles of EVs in tumour growth, cancer progression, and metastasis. EVs secreted by cancerous cells cause many changes in neighbouring healthy cells and supporting stromal cells, transforming them into cancer-related cells. These cells, in turn, secrete mutant-containing EVs and reciprocally influence one another, thereby promoting metastasis. CAF: Cancer-associated fibroblast; ECM: Extracellular matrix; EMT: Epithelial-mesenchymal transition; PMN: Pre-metastatic niche; TAM: Tumour-associated macrophage

Extracellular vesicles (EVs) in hormonal and chemotherapy resistance

EVs are also widely acknowledged as both indicators of advanced PCa and active facilitators of disease advancement and therapy resistance [140–143]. By delivering bioactive cargo that promotes tumour cell survival and adaptation under treatment pressure, EVs may help castration-resistant prostate cancer (CRPC) develop resistance to both AR signalling inhibitors and taxane-based chemotherapy [140–143]. EVs have been linked to the preservation of AR signalling in the context of hormonal therapy in spite of androgen deprivation. The transfer of AR and constitutively active splice variants, such as AR-V7, via EVs may enable recipient cells to sustain AR-driven transcription in low-androgen environments [141]. This process is particularly pertinent in CRPC, where sustained AR activity remains a primary driver of disease progression, despite therapy with AR-targeted drugs like enzalutamide and abiraterone. The identification of AR-V7 in plasma-derived EV RNA is clinically significant and correlates with resistance to hormonal therapy and adverse outcomes in patients with metastatic CRPC [140].

The functional significance of EV trafficking in treatment adaptation is further supported by evidence that increased EV secretion is linked to enzalutamide resistance and that inhibiting Syntaxin 6-mediated EV release decreases the viability of resistant PCa cells [143]. Additionally, resistance to taxane treatment, especially docetaxel, has been linked to EVs. Docetaxel-resistant PCa cells can produce EVs that transfer resistance-associated features to previously sensitive recipient cells, suggesting that EVs may actively spread chemoresistance among tumour cells [142]. One potential mechanism entails the transfer of multidrug resistance-related molecules, such as P-glycoprotein (P-gp/MDR1), which may decrease intracellular drug concentration and hence limit chemotherapy effectiveness [142]. In addition to protein cargo, EVs may transport RNAs and other modulatory elements that support cell survival, stress adaptability, and resistance-related signalling [142]. Collectively, these data underscore the dual importance of EVs in CRPC as they serve as attractive candidates for liquid biopsies and as active contributors to therapy resistance. An enhanced understanding of EV-mediated resistance could refine therapeutic stratification and facilitate the development of more efficacious treatment options for advanced PCa.

Extracellular vesicles (EV)-derived proteins as cancer biomarkers in prostate cancer (PCa)

EVs carry a broad spectrum of molecular constituents, including proteins, various RNA species, DNA fragments, and other regulatory nucleic acids. Advances in next-generation sequencing have enabled comprehensive profiling of EV-derived nucleic acids in PCa, demonstrating their potential utility as liquid biopsy biomarkers for disease detection and prognosis. However, in this review, the discussion is primarily centred on EV-associated proteins, as proteomic studies have provided important insights into PCa pathophysiology, tumour progression, and mechanisms contributing to therapeutic resistance.

In current practice, abnormal blood PSA levels and/or DRE findings require a prostate biopsy to confirm the diagnosis. However, elevated blood PSA levels are observed not only in PCa, but also in many other benign prostatic conditions. Therefore, there is an urgent need for more precise PCa biomarkers with minimally invasive techniques to avoid unnecessary prostate biopsy and its associated surgical complications. Over the past decade, the role of EVs as potential biomarkers for PCa has been intensively investigated. EVs have been recognised as optimal candidates for biomarker applications owing to their availability in various body fluids, abundance in the discharge from cancerous cells, the resemblance of their contents to the parental cells, resilience against enzymatic degradation that safeguards their contents, and the feasibility of employing straightforward techniques for their isolation and subsequent laboratory analysis. A recent study detailing the discovery of EV proteins as biomarkers for PCa in urine, serum, and tissue samples from PCa patients, mainly resourced from Overbye et al. (2015) [144], Welton et al. (2016) [145], Turay et al. (2016) [146], and Ronquist et al. (2010) [147], is presented in Tables 1, 2 and 3.

Table 1.

Proposed potential EV proteins derived from the urine of PCa patients for PCa biomarkers

No	Entry name	Protein full name	Gene	Function	References
1.	ADIRF	Adipogenesis regulatory factor	ADIRF	Induce PPARG expression to promote adipocyte differentiation.	[148]
2.	AMACR	a-Methylacyl-CoA racemase	UE-A	Facilitates the transformation of (R)-a-methylbranched-chain fatty acyl-CoA esters into their (S)-stereoisomers.	[149]
3.	ARL8B	ADP-ribosylation factor-like protein 8B	ARL8B	Lysosomal transport.	[150]
4.	CALM	Calmodulin	CALM	The interaction of CALM and AR promotes the proliferation of LNCaP cells.	[144, 151]
5.	CLD10	Claudin-10	CLDN10	Transforming growth factor-β (TGF-β)- or WNT/b-catenin-induced EMT affects the progress of ovarian cancer.	[144, 152]
6.	CLD2	Claudin-2	CLDN2	EMT, tumour initiation, and chemotherapy resistance.	[144, 153]
7.	CLD3	Claudin-3	CLDN3	Enhance cell motility and survival by activating MMP-2 and suppressing EMT.	[144, 153]
8.	CTND	D-catenin	CTNND	Interacts with E-cadherin to inhibit tumour migration.	[154]
9.	DHRS6	3-hydroxybutyrate dehydrogenase type 2/Dehydrogenase/reductase SDR family member 6	BDH2	Induce apoptosis.	[144, 155]
10.	FLOT2	Flotillin-2	FLOT2	Molecules involved in signal transduction, adhesion, and ECM remodelling.	[144]
11.	GMDS	GDP-mannose 4.6 dehydratase	GMDS	Regulate TRAIL-induced apoptosis and increase natural killer (NK) cell-mediated tumour surveillance.	[144, 156]
12.	GNPI1	Glucosamine-6-phosphate isomerase 1	GNPDA1	Promote metabolism and inhibit apoptosis.	[157]
13.	ITA3, ITB1	Integrin alpha-3, Integrin beta-1	ITGA3, ITGB1	Activate oncogenic signalling pathway.	[158]
14.	LEG3	Galectin-3 binding protein	LGALS3	These include the inhibition of apoptosis, the promotion of cell growth, and the regulation of T-cell receptor signal transduction, which, in turn, promotes angiogenesis.	[144, 159]
15.	LTOR1	Regulator complex protein LAMTOR1	LAMTOR1	Affect lysosomal localisation.	[160]
16.	MELPH	Melanophilin	MLPH	Accelerate EMT to promote tumour metastasis.	[144]
17.	MFS12	Major facilitator superfamily domain-containing protein 12	MFSD12	Promote the G1 phase.	[144, 161]
18.	MYCT	Proton myo-inositol cotransporter	SLC2A13	Regulate Hif-1a to promote tumour cell hypoxia.	[162]
19.	P2RX4	P2X purinoceptor 4	P2RX4	Induction of immunosuppression and angiogenesis activates antitumor response.	[144, 163]
20.	PARK7	Protein DJ-1 (Parkinson’s disease protein 7)	PARK7	Inhibit phosphatase and tensin homolog (PTEN) tumour suppressor.	[164]
21.	PLSL	Plastin-2	LCP1	Regulate integrin-mediated tumour cell adhesion.	[165]
22.	PSMA	PSA, PSMA	PSMA	Related to angiogenesis.	[166]
23.	PSMA	Prostate-specific membrane antigen	PSMA	Urine-derived exosomal PSMA is a promising diagnostic biomarker for detecting PCa on initial biopsy.	[167]
24.	RAB2A	Ras-related protein Rab-2 A	RAB2A	Activate Erk signal to promote breast cancer stem cells and tumourigenesis.	[144, 168]
25.	Rab-35	Ras-related protein Rab-35	RAB35	Induced EMT, intracellular signalling, apicobasal polarity, cytokinesis, and cell migration. Promote differentiation and tumour cell proliferation.	[169]
26.	RAB3B	Ras-related protein Rab-3B	RAB3B	Inhibit apoptosis and maintain cancer cell survival.	[144, 170]
27.	RAB3D	Ras-related protein Rab-3D	RAB3D	Induces cytoskeleton remodelling, enhances cancer cell movement, induces EMT, regulates Hsp90a secretion, and promotes tumour cell invasion.	[144, 171]
28.	RAB7A	Ras-related protein Rab-7a	RAB7A	Prevent HGF-induced lysosomal trafficking, cathepsin B secretion, and cell invasion.	[144, 172]
30.	S10A6	Protein S100-A6	S100-A6	S100A6 interacts with annexin 2 and promotes cancer cell motility.	[144, 173]
31.	TM256	Transmembrane protein 256	TMEM256	Induce tumour formation.	[144]

Table 2.

Proposed potential EV proteins derived from the serum of PCa patients for PCa biomarkers

No	Entry name	Protein full name	Gene	Function	References
1.	A2GL	Leucine-rich alpha-2-glycoprotein	LRG1	Promote angiogenesis.	[174]
2.	ACTN4	Alpha-actinin-4	ACTN4	Heighten cells’ proliferation through the growth signalling pathway and invasion capability.	[175]
3.	APOE2	Apolipoprotein E (isoform E2)	APOE2	Inhibit mitochondrial apoptosis, hence evading cellular apoptosis.	[145, 176]
4.	BIRC5	Survivin/baculoviral IAP repeat-containing protein 5	BIRC5	Inhibit apoptosis.	[177]
5.	C1Q	Complement C1q subcomponent	C1Q	It entices endothelial cell movement and promotes tubal formation.	[145, 178]
6.	C1R	Complement C1r subcomponent	C1R	Support tumour vascularisation and tumour growth in cutaneous squamous cell carcinoma.	[145, 179]
7.	CO3	C3a anaphylatoxin des Arginine	C3	Inhibit neutrophils and CD4+T cell response to the tumour.	[145, 180]
8.	CYP17A	Steroid 17-alpha-hydroxylase/17,20 lyases	CYP17A1	Involves androgen production to induce PCa growth.	[181]
9.	EXOS5	CML28/Exosome complex component RRP46/hRrp46p	EXOSC5	Supports mRNA turnover in RNA transcripts to maintain the cell’s highly proliferative state, hence supporting tumour growth.	[182]
10.	FHL1	Four and a Half LIM Domain 3	FHL1	Interact with p21, c-myc, cyclin D1 & B1, SOX4, and SOX2 proteins to retard cell growth, and reduce cell proliferation and self-renewal by inducing cell-cycle stagnation in breast, liver, and glioma stem cell cancer.	[146, 183]
11.	GSTO2	Glutathione S transferase Omega 2	GSTO2	Catalyse carcinogen detoxification and metabolism of xenobiotics and carcinogens. Impedes cell multiplication, the chance of metastasis, and mitochondria function through p38 signalling.	[146, 184]
12.	IRX5	Iroquois homeobox protein 5	IRX5	Boosts proliferation, migration, invasion, and EMT of oral colorectal carcinoma cells.	[146, 158]
13.	ITB3	avb3 integrin	ITGB3	Promote cell adhesion and migration on vitronectin, an ανβ3 ligand.	[128]
14.	ITIH3	Inter-alpha-trypsin inhibitor heavy chain H3	ITIH3	Chemotherapy resistance.	[185]
15.	LRRC4	Leucine-rich zipper containing 4/Leucine-rich repeat containing 4	LRRC4	Inhibit migration and invasion.	[146, 186]
16.	MCM5	Minichromosome maintenance complex component 5/DNA replication licensing factor MCM5	MCM5	Involved in the initial phase of DNA replication & cell cycle regulation.	[146, 187]
17.	MELK	Maternal Embryonic Leucine Zipper Kinase	MELK	Promotes cell proliferation.	[146, 188]
18.	MTUS1	Mitochondrial tumour suppressor 1 isoform 4	MTUS1	Retards cancer cell proliferation by delaying mitosis succession.	[146, 189]
19.	NEUR3	Sialidase-3	NEU3	Immunosuppression	[190]
20.	P-gp	p-glycoprotein	MDR1	Chemotherapy resistance.	[191]
21.	PIF1	DNA helicase homologue PIF1/ATP-dependent DNA helicase PIF1	PIF1	Preserves tumour cells from apoptosis, hence supporting tumour growth.	[146, 192]
22.	PTEN	Phosphatidylinositol 3,4,5-trisphosphate 3-phosphatase and dual-specificity protein phosphatase PTEN	PTEN	Inhibits tumour progression by maintaining a minimal level of cellular PIP3, thereby suppressing the PI3K-AKT pathway and leading to cell death.	[148, 193]
23.	TNRC6B	Trinucleotide repeat containing 6B Isoform 3	TNRC6B	Mediate cell proliferation and apoptosis in head & neck cancer.	[148, 194]
24.	UHRF1	Ubiquitin-like with PHD and ring finger domains	UHRF1	Induce cell growth by expressing DNA methyltransferase 1 (DNMT1).	[148, 195]

Table 3.

Proposed potential EV proteins derived from the tissue samples of PCa patients for PCa biomarkers

No	Entry name	Protein full name	Gene	Function	References
1.	1433 S	14-3-3 Protein sigma	SFN	Regulates the inhibition of cell proliferation, migration, and tumorigenesis, and modulates cell cycle retardation and cancer cell apoptosis.	[147]
2.	ACBP	Acyl-CoA-binding protein	DBI	Binds to acyl-CoAs to regulate long-chain fatty acids for their oxidation in mitochondria, assisting cancer cell proliferation and, hence, tumorigenesis.	[147, 196]
3.	ACTB	Actin, cytoplasmic 1	ACTB	Induces cytoskeletal alterations to facilitate tumour advancement.	[197]
4.	ADH1A	Alcohol dehydrogenase [NADP+]	ADH1A	Its high expression increases the risk of various types of cancer.	[147, 198]
5.	AMPL	Cytosol aminopeptidase/Leucine aminopeptidase 3	LAP3	Suppresses cancer cell invasion by regulating fascin and MMP2/9.	[147, 199]
6.	ANXA1	Annexin A1	ANXA1	Promote invasion & migration of pancreatic & laryngeal cancer cells. Annexin A5 upregulates MMP2 & MMP9, which causes EMT remodelling. Regulate apoptotic activities	[147, 200]
7.	ANXA3	Annexin A3	ANXA3	Modulate cell proliferation, migration, and apoptosis.	[147, 201]
8.	ANXA5	Annexin A5	ANXA5	Promote invasion & migration of pancreatic & laryngeal cancer cells. Annexin A5 upregulates MMP2 & MMP9, which causes EMT remodelling. Regulate apoptotic activities.	[147, 200]
9.	CAZA1	F-Actin-capping protein subunit alpha-1	CAPZA1	Decreases cancer cell invasion and migration by inhibiting EMT.	[202]
10.	DDAH1	N(G), N(G)-Dimethylarginine dimethylaminohydrolase 1	DDAH1	Stop angiogenesis by modulating the DDAH/ADMA/NO pathway.	[147, 203]
11.	DHSO	Sorbitol dehydrogenase	SORD	Inhibits tumour growth and cancer cell stemness by modulating the necroptosis signal in hepatocellular carcinoma.	[147, 204]
12.	ENOA	Alpha-enolase	ENO1	Promotes tumour glycolysis, therefore increasing cancer cell proliferation and drug resistance.	[147, 205]
13.	GSHB	Glutathione synthetase	GSS	Retards tumour progression by suppressing ROS and lessening drug resistance.	[147]
14.	HINT1	Histidine triad nucleotide-binding protein 1/Adenosine 5’-monophosphoramidase HINT1	HINT1	Deranged (CD)4+ T cells differentiation, affiliated with increased tumour stemness and depressed stromal cells’ immunity in breast cancer.	[147, 206]
15.	IDHC	Isocitrate dehydrogenase 1 (NADP+), soluble/Isocitrate dehydrogenase [NADP] cytoplasmic	IDH1	Reduces the cellular NADPH/NADP+ ratio, thereby decreasing protection against oxidative damage.	[147]
16.	K1C10	Keratin, type I cytoskeletal 10	KRT10	Suppress cell proliferation and tumour formation in skin cancer.	[147, 207]
17.	PEBP1	Phosphatidylethanolamine-binding protein 1/Raf Kinase Inhibitory Protein (RKIP)	PEBP1	Suppress metastasis, inhibit kinases, inhibit NF-қB & PI3K/Akt/mTOR signalling pathway.	[147, 208]
18.	PIP	Prolactin-inducible protein	PIP	Increase NK cells and decrease type-2 T-helper cells in the TME, which consequently delays tumour onset and reduces tumour development.	[147, 209]
19.	PRDX6	Peroxiredoxin-6	PRDX6	Modulate EMT signalling pathway via p38 phosphorylation in colon cancer, increasing cell proliferation and tumour growth.	[147, 210]
20.	S10A9	Protein S100-A9	S100A9	Activate the NF-κB pathway to regulate the pro-inflammatory response in the TME, thereby promoting tumour growth.	[147, 211]
21.	S10AB	Protein S100-A11	S100A11	Activates EMT, then promotes migration and invasion of cancer cells via Wnt/β-catenin signalling pathways in ovarian cancer.	[212]
22.	SDCB1	Syntenin-1	SDCBP	Promote tubular formation by endothelial cells, thereby supporting angiogenesis in the TME.	[147, 213]
23.	SEMG1	Semenogelin-1	SEMG1	Increases glycolysis and respiration in cancer cells, thereby boosting their energy metabolism.	[147, 214]
24.	SERA	D-3-phosphoglycerate dehydrogenase	PHGDH	Increases serine production to promote cell proliferation.	[147, 215]
25.	SODC	Superoxide dismutase [Cu-Zn]	SOD1	Unites with copper to prevent angiogenesis. Prevent cellular oxidative stress response.	[212, 216]
26.	TPIS	Triosephosphate isomerase	TPI1	It is involved in energy metabolism through glycolysis, gluconeogenesis, the pentose phosphate pathway, and fatty acid biosynthesis, enabling cancer cells to meet the high demands of metastasis.	[147, 217]
27.	UBE2N	Ubiquitin-conjugating enzyme E2 N	UBE2N	Activate NF-қB and p38 signalling pathway, therefore inducing pro-inflammatory conditions in TME to support tumour development & metastasis.	[147, 218]

EV isolation techniques

Increasing research highlights the critical role of EVs across various physiological and pathological contexts. It remains necessary to extract and isolate the body’s circulating EVs for research and laboratory experiments that support practical clinical applications, such as diagnostics and disease monitoring. To obtain pure results, EVs released from cells in cell culture must be purified. Currently, there are various methods for EV isolation, each with its own pros and cons. To ensure the isolated EVs are of high quality, it is crucial to use the appropriate technique for each sample. Fundamentally, EV isolation techniques documented and demonstrated in many experimental papers to produce EVs include ultracentrifugation (UC)-based [124], precipitation-based [219], immunoaffinity separation [220], size-based isolation [25], and, most recently, microfluidic-based [221].

Ultracentrifugation

The advancement of EV isolation techniques has led researchers to use terms such as traditional and conventional methods versus those employing up-to-date technology, UC, and non-centrifugation-based methods [222]. UC has been acknowledged as the prototype, gold standard [223], traditional way, and the most commonly used method for EV isolation [221, 224]. Since the advent of the UC machine by Svedberg in the 1920 s, there has been substantial development of other EV separation methods, mainly to address the drawbacks associated with UC [225]. Li et al. (2017) [65] classified UC into two categories: analytical and preparative. The former analyses the physicochemical properties of particulate materials, while the latter investigates molecular-level interactions in polymeric materials [65]. On the other hand, preparative UC, which is more applicable to the separation or compartmentalisation of small bio-particles, is further divided into differential and density-gradient UC [65]. Differential UC involves consecutive centrifugation steps with increasing gravitational force and duration; the greater the force applied and the longer the duration, the smaller the particles isolated [226]. Technically, to isolate EVs, a 10-min of 300 to 400 g-force is meant to get rid of intact cells, a 10 to 15-min of 2000 x g-force, and a 30-min of 10,000 x g-force is intended to clear out dead cells and cell debris, respectively, while much greater g-force of 100,000 to 200,000 in 70 to 90 min is meant to root out contaminating proteins [224]. The downside of UC, however, has led to the development of other alternative EV isolation methods. UC, indeed, is a time-consuming procedure that requires a heavy and space-occupying machine, a trained and dedicated machine operator, a large sample to start with, and is undeniably costly to maintain. Therefore, it is not clinically applicable in an under-resourced hospital laboratory [221].

Precipitation technique

There are generally two types of precipitation isolation techniques: polymeric precipitation and lectin-induced agglutination [225]. The water-excluding polymer, polyethylene glycol (PEG), behaves like a “polymer net” that traps only EVs within a specific size range (60–150 nm), which falls within the small EV or exosome size range [225, 227]. The precipitated EV pellet would then be retrieved with low centrifugation. ExoQuick-^TC reagent (Systemic Bioscience (SBI)), Exosome Isolation kit (Life Technologies), Exospin (Cell Guidance System), and Total Exosome Isolation Kit (Thermo Fisher Scientific) are examples of ready-to-use kits that use precipitation [221, 225, 227]. The advantages of this method include a simple protocol, no need for sophisticated or complex equipment, no expertise required, and a benchtop procedure [65, 228]. Despite its sensibility for clinical application, however, there are still some issues that need to be addressed, such as the manufacturers’ unwillingness to disclose the chemical recipe and the ingredients’ mode of action, which could affect EVs’ biological functions, and last but not least, possible contamination with protein-like extracellular argonaute 2 complexes [221, 229].

Immunoaffinity separation

This is an immunologic technique that employs surface proteins on EV’s membrane as antigens towards the selected antibodies used either to tackle desired or to catch unwanted EVs, and thus, this technique secures highly specific EVs. Common surface protein biomarkers used to isolate EVs are CD81, CD63, and CD9 [222]. However, owing to a clear-cut, definite antibody-antigen selection for capturing EVs, a heterogeneous EV population that does not express a similar surface protein is at risk of being left out [224].

Size-based technique

EV isolation techniques based on EV size include ultrafiltration [230], size-exclusion chromatography (SEC) [231], flow-field-flow fractionation (FFFF) [225], and hydrostatic filtration dialysis (HFD) [25]. Ultrafiltration utilises porous membranes of varying pore sizes to separate and isolate EVs, thereby yielding much better purification than other techniques [232]. Both ultrafiltration and SEC use a solid matrix, such as a polyethersulfone nanomembrane concentrator, which is semipermeable [227]. Although the SEC promises higher purity isolation without requiring complex equipment and is very suitable for Mass Spectrometry analysis, it liquefies them. Therefore, combining those two methods unravels the misery [232]. The disadvantages of these methods include proteins that stick to the nanomembrane, which are difficult to remove and impede EV isolation. During pre-concentration of the sample, forces applied to EVs as they pass through the filters can cause damage and rupture, ultimately reducing their concentration [224].

Microfluidic-based technology

Devices that utilise microfluidic technology manipulate fluid volumes at the micro- or nanolitre level through channels and structures downscaled to a miniature scale, a technology known as “lab-on-a-chip” [233]. The characteristics of the device utilising this technology include a large surface-to-volume ratio, shorter analysis time, laminar flow, simplicity, multiple detection with higher recovery rates, and purity preparation [234]. Every microfluidic device employs methods tailored to the physical characteristics of EVs [235]. There are several approaches used in microfluidic devices in EV isolation, including immunoaffinity capture, where antigens, EV membrane proteins, and monoclonal antibodies interact immune-affinitively [235], sieving or capturing EV by size-filtering them through a nanoporous membrane directly from a biological fluid such as whole blood driven by pressure or electrophoresis [221], and entrapping EVs through porous or ciliated micropillar structures like nanowire, which picking up particles of 40 to 100 nm size [236]. Fast and effortless isolation, economical [221], fewer reagents and biological samples are required while maintaining its isolation efficiency at a quicker rate of particle separation and identification [237], simultaneous high-throughput processing, self-regulating and streamlined operation, and scrupulous control of overflow conditions [233], are among the advantages of this technology over the conventional EV isolation methods.

Challenges in extracellular vesicle (EV) isolation from biofluids for cancer biomarker discovery

Urine and blood are among the widely used biofluids for biomarker discovery, including in PCa, due to their accessibility and minimally invasive collection procedures. However, isolating EVs from these biofluids remains technically challenging. In addition to EVs carrying potential PCa-associated biomarkers, urine and blood also contain various circulating components, including intact cells, cell fragments, protein aggregates, lipoproteins, and viral particles. The coexistence of these structures complicates EV purification and may introduce contaminants that co-isolate with EV fractions, thereby reducing their purity. Furthermore, the high abundance of soluble proteins in blood, together with other non-vesicular particles present in both biofluids, can mask or dilute low-abundance EV-derived protein biomarkers, thereby limiting the sensitivity and accuracy of downstream proteomic analyses. To minimise this interference, several strategies have been employed, including the use of a spin column [238]. In a different study, combining a spin column with SEC revealed that depletion of abundant plasma proteins, particularly albumin, enhanced EV recovery and improved EV proteome characterisation, thereby enabling the identification of more proteins in mass spectrometry-based analyses. However, EVs isolated from albumin-depleted plasma exhibited a slightly reduced detection of EV markers compared with EVs obtained from non-depleted plasma [239].

In urine, the presence of Tamm–Horsfall protein (THP), also known as uromodulin, has posed a significant challenge to the isolation of urinary EVs [240, 241]. THP is a glycan-rich glycoprotein normally present in urine and secreted by epithelial cells in the kidney, and it readily forms polymers at low temperatures [241]. These ‘net-like structures’ can entrap EVs or co-sediment during centrifugation, thereby contaminating EV preparations and interfering with downstream analyses. To minimise THP-associated interference, several pre-analytical strategies have been proposed. For instance, a combination of salt precipitation and differential low to high-speed centrifugation (3000 g to 17000 g) has been used to remove the THP complex prior to EV isolation using ultracentrifugation [242]. Mild treatment with reducing agents such as dithiothreitol (DTT) has been used to disrupt THP polymer structures and release EVs trapped within the THP matrix while preserving EVs’ integrity [240]. Successfully removing these abundant proteins is paramount for guaranteeing the reliability of downstream EV biomarker assays. Consequently, careful evaluation of the strengths and limitations of different EV isolation and purification approaches is necessary. The choice of isolation method should therefore be tailored to the specific aims of the study, the characteristics of the biofluid, and the requirements of subsequent analytical techniques.

Summary of future perspectives

The inconclusiveness of current diagnostic tests for PCa has led to the emergence of alternative biomarkers, including EVs. EVs serve as key mediators of intercellular communication, transferring molecular cargos such as RNAs and proteins to recipient cells and thereby influencing a wide range of physiological and pathological processes, including cancer. Owing to their abundance, stability, and specificity, EVs contain rich biological information and have emerged as highly valuable specimens for liquid biopsy applications. Their protein and nucleic acid contents have been increasingly explored as novel diagnostic and prognostic biomarkers across multiple cancer types, including PCa. EV-associated proteins have been extensively investigated for their potential roles as disease-related biomarkers, which may facilitate the identification of novel biomarkers for early PCa diagnosis and more reliable prognostic assessment. Accumulating evidence indicates that EV-proteins regulate EV-driven tumour migration, invasion, metastasis, and angiogenesis, highlighting their strong potential for translation into clinically relevant applications. Although significant progress has been made, substantial work remains before EV-based biomarkers can transform cancer diagnostics. Bridging experimental insights with clinical practice will pave the way for personalised diagnostics and precision medicine, offering new therapeutic possibilities and a more hopeful outlook for PCa patients.

Nevertheless, the clinical utility of EV-derived biomarkers is significantly hindered by two core challenges: the lack of procedural standardisation and the instability of biomarkers in patient biofluids. Current isolation methods (e.g., ultracentrifugation, precipitation, chromatography) produce heterogeneous EV populations with variable quality and quantity, complicating comparisons across studies. Therefore, moving beyond reporting guidelines such as MISEV2023 to harmonise the specific isolation and quantification protocols is a crucial step toward regulatory acceptance. Furthermore, the stability of the molecular contents of the EV itself is vulnerable to pre-analytical variables such as sample collection, processing delays, storage conditions, and freeze-thaw cycles, all of which can compromise EV integrity and the concentration of their protein and nucleic acid cargo, thereby diminishing diagnostic accuracy. Resolving this lack of uniformity and inherent instability is essential for fulfilling the potential of EV liquid biopsies in PCa management.

Abbreviations

ANGPT2

Angiopoietin-2

ANXA1

Annexin A1

bFGF

Basic fibroblast growth factor

BMDCs

Bone marrow-derived cells

BM-MSC

Bone marrow mesenchymal stem cell

BPSA

Benign PSA

CAF

Cancer-associated fibroblast

CCL2

Chemokine (C-C motif) ligand 2

circRNA

Circular RNA

CML

Chronic myeloid leukaemia

CRPC

Castration-resistant prostate cancer

DRE

Digital rectal examination

DSBs

Double-strand breaks

dsDNA

Double-stranded DNA

DUBs

Deubiquitylating enzymes

ECM

Extracellular matrix

EEs

Early endosomes

EGF

Epidermal growth factor

EGFR

Epidermal growth factor receptor

EMT

Epithelial-to-mesenchymal transition

ESCRT

Endosomal sorting complex required for transport

EVs

Extracellular vesicles

FasL

Fas ligand

FGF

Fibroblast growth factor

GCSF

Granulocyte-colony stimulating factor

gDNA

Genomic DNA

GLUT-1

Glucose transporter-1

HCC

Hepatocellular carcinoma

HCV

Hepatitis C

HGF

Hepatocyte growth factor

HIF-1α

Hypoxia-inducible factor 1-α

hKLK3

Human kallikrein 3

HRS

Hepatocyte growth factor-regulated tyrosine kinase substrate

HSC

Hepatic stellate cells

HSP

Heat shock proteins

HUVEC

Human umbilical vein endothelial cells

IBD

Inflammatory bowel disease

IL-1α

Interleukin 1α

ILVs

Intraluminal vesicles

iPSA

Inactive PSA

ITG-α3

Integrin-alpha 3

ITG-β1

Integrin-beta 1

LAMP-1/2B

Lysosome-associated membrane glycoproteins

LAP-TGFβ

Latency-associated peptide-TGFβ

LEs

Late endosomes

lncRNAs

Long non-coding RNAs

MALT

Mucosal-associated lymphoid tissue

miRNA

MicroRNA

MMP-2

Matrix metalloprotease 2

MPO

Myeloperoxidase

mRNA

Messenger RNA

mtDNA

Mitochondrial DNA

MVB

Multivesicular bodies

NF-КB

Nuclear factor-КB

PBC

Primary biliary cholangitis

PCa

Prostate cancer

PCSEAT

PCa-specific expression, and the EZH2-associated transcript

DGF

Platelet-derived growth factor

PI3P

Phosphatidylinositol 3-phosphate

piRNA

Piwi RNA

PKM2

Pyruvate kinase M2

PLP

Proteolipid protein

PMN

Pre-metastatic niche

PMNs

Polymorphonuclear neutrophils

proPSA

Proenzyme PSA

PSC

Primary sclerosing cholangitis

PSA

Prostatic-specific antigen

RAB

Ras-associated binding

R-t compDNA

Reverse-transcribed complementary DNA

ROS

Reactive oxygen species

rRNA

Ribosomal RNA

SNAREs

SNAP receptors

SNAP

Soluble N-ethylmaleimide-sensitive factor attachment protein

snRNA

Small-nuclear RNA

snoRNA

Small-nucleolar RNA

ssDNA

Single-stranded DNA

STAM1/2

Signal transducing adaptor molecule ½

TAM

Tumour-associated macrophages

tPSA

Total PSA

t-SNAREs

Target SNAREs

TGF-β1

Transforming growth factor-beta 1

TGN

Trans-Golgi network

THP

Tamm–Horsfall protein

TLR2

Toll-like receptor 2

TME

Tumour microenvironment

TNF-α

Tumour necrosis factor-alpha

tRNA

Transfer RNA

tsRNA

Transfer RNA (tRNA)-derived small RNAs

uPA

Urokinase plasminogen activator

VAMP

Vesicular-associated membrane protein

VEGF

Vascular endothelial growth factor

VEGFR

Vascular endothelial growth factor receptor

v-SNAREs

Vesicular SNAREs

VPS4

ATPase vacuolar-protein-associated sorting-4

vtRNA

Vault RNA

Author contributions

DSS, MNHA, and AN were involved in designing the topic, writing, and revising the manuscript. All authors read and approved the final manuscript prior to submission.

Funding

This study was supported by the Minister of Higher Education’s research grant, under the Fundamental Research Grant Scheme with the vote number of FRGS/1/2019/SKK06/UPM/02/7 (Special Graduate Research Allowance (SGRA) Scheme), with additional support from Universiti Putra Malaysia, under the IPM research grant with the vote number. GP-IPM/2018/9640100.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.