Exosome biogenesis in arthropods: conserved mechanisms, experimental evidence, and emerging directions

Abstract

Extracellular vesicles (EVs) are nanoscale, lipid-bound structures released by cells across all domains of life. Once viewed as a means for discarding unwanted cellular components, they are now understood to be central mediators of intercellular communication. Much of what is known about EVs comes from mammalian systems, where extensive work has defined the major EV subtypes and the mechanisms that generate exosomes. These findings continue to serve as the primary reference for interpreting EV biology in other organisms. EVs have been isolated from diverse arthropods, including a few insect and tick species, yet the molecular pathways that produce them remain less characterized. Current evidence from Drosophila, mosquito cell lines, and tick systems shows that many of the core components driving exosome biogenesis in mammals, including ESCRT complexes, tetraspanins, lipid-modifying enzymes, Rab GTPases, and SNARE proteins, are present and, in several cases, experimentally validated in arthropods. These findings point to broad conservation of exosome biogenesis across taxa, while also highlighting key caveats, with most conclusions relying on only a small number of model systems. The definitions of EV subtypes remain unclear in many arthropods, and alternative EV biogenesis pathways have received little attention. Future studies that incorporate non-model species, apply rigorous EV characterization standards, and explore the roles of various EV subtypes will clarify how these pathways operate across arthropod lineages and how they differ from well-studied mammalian systems.

Introduction

Extracellular vesicles (EVs) are a heterogeneous group of nanoscale, lipid-bound structures secreted by cells into the extracellular space (Raposo and Stoorvogel 2013, van Niel et al. 2018). Upon their discovery, EVs were seen as carriers for the disposal of unwanted cellular components (Colombo et al. 2014), but they were later recognized as important mediators of intercellular communication, both within individual organisms and between different organisms (Simons and Raposo 2009). EVs appear to be broadly conserved across domains of life. Both eukaryotic and prokaryotic cells release EVs that selectively package proteins, lipids, and/or nucleic acids, thereby facilitating intercellular communication, environmental adaptation, and genetic exchange between cells, tissues, organs, and species (Schwechheimer and Kuehn 2015, Mathieu et al. 2019, Nagakubo et al. 2019).

EVs are broadly distributed across tissues and physiological systems. In mammals, EVs have been isolated from diverse biological fluids, including blood (Nieuwland and Siljander 2024), saliva (Ogawa et al. 2008, Palanisamy et al. 2010, Sun et al. 2018), urine (Pisitkun et al. 2004), amniotic fluid (Keller et al. 2007), breast milk (Admyre et al. 2007), cerebrospinal fluid (Street et al. 2012), and semen (Madison et al. 2015). Similarly, EVs have been isolated from plant apoplastic fluid (Rutter and Innes 2017), bacterial culture supernatants (Schwechheimer and Kuehn 2015), insect hemolymph (Tassetto et al. 2017, Linnemannstöns et al. 2022, Van den Brande et al. 2025), and the venom of spiders, scorpions, and parasitoid wasps, where EVs contribute to toxin delivery and host manipulation (Bala et al. 2025). Despite this broad distribution, most current knowledge of EV biology derives from studies in mammals, which provide the conceptual framework for investigating EVs in other systems. In this review, we summarize current knowledge of EV biogenesis in mammals and arthropods and evaluate the extent to which these pathways are conserved and functionally validated.

Mammalian EV Types

In mammals, EVs are often grouped by size (Fig. 1), but this criterion alone does not fully capture their heterogeneity. Instead, EVs are most commonly classified by their biogenesis pathways, as the mechanism of vesicle formation strongly influences cargo composition and biological function. However, many EV subtypes share overlapping protein and lipid markers, and although certain molecules are preferentially enriched in specific EV populations, no universal or exclusive molecular features currently exist that can distinguish EV subtypes with complete confidence (Kowal et al. 2016, Raposo and Stoorvogel 2013, Colombo et al. 2014, Théry et al. 2018, van Niel et al. 2018). This creates persistent challenges in EV terminology and categorization, even within well-characterized mammalian systems, as emphasized by the Minimal Information for Studies of Extracellular Vesicles (MISEV) 2023 guidelines (Welsh et al. 2024). These evolving mammalian definitions provide the primary framework for describing EV subtypes in arthropods, but subtype-specific classification is still emerging. In mammals, additional EV and EV-like particle classes have been reported and cataloged, including those described in community resources such as Vesiclepedia (Chitti et al. 2024). However, this review focuses on EV subtypes that have been explored in arthropods and EV-associated proteins are described as enriched or associated with particular biogenesis pathways, rather than as definitive markers of specific EV subtypes.

Fig. 1.

Schematic representation of selected extracellular vesicle (EV) and extracellular particle (EP) subtypes based on approximate size ranges. The figure includes apoptotic bodies (approximately 500–4,000 nm), microvesicles (50–2,000 nm), exosomes (approximately 30–150 nm), and the non-membranous extracellular particles exomeres (<50 nm) and supermeres (<30 nm). These particles differ in size, structural organization, cellular origin, and molecular composition, and may coexist in extracellular preparations. Figure adapted from Vesiclepedia (Chitti et al. 2024) and prepared with Biorender.

Apoptotic bodies are the largest class of EVs (500–4,000 nm, Fig. 1) and are produced during programmed cell death (Battistelli and Falcieri 2020, El Andaloussi et al. 2013). As the cell breaks apart through blebbing and fragmentation, it releases large vesicles that can carry substantial cellular material, including organelle fragments, DNA, and even portions of the nucleus. These vesicles are often discussed in the context of packaging and sequestering cellular components, including materials that could otherwise elicit inflammatory responses. Nuclear material, including chromatin-associated components such as histones, are commonly detected in apoptotic bodies and are often used as indicative components of their nuclear origin, although they are not exclusive markers and may also be present in other extracellular vesicle populations (Crescitelli et al. 2013, Théry et al. 2018).

Microvesicles (MVs) are generally larger than exosomes (50–2,000 nm, Fig. 1), although their size ranges can overlap (Akers et al. 2013). They form by outward budding and fission of the plasma membrane. Because of their origin, MV membranes often reflect the protein and lipid compositions of the parent cell’s plasma membrane. Commonly reported microvesicle-associated protein markers include integrins, selectins, and CD40 ligand (Sedgwick and D’Souza-Schorey 2018). These proteins are well known for their roles in cell adhesion and immune signaling in their cellular contexts and have been detected on microvesicles in multiple mammalian systems, though their expression is not exclusive to microvesicles. They are broadly discussed in relation to vesicle-associated cell–cell communication and immune-related interactions (Théry et al. 2018).

Exosomes are EVs (30–150 nm, Fig. 1) generated through the endosomal pathway, although MISEV guidelines caution that biogenesis-based classification requires direct mechanistic validation (Welsh et al. 2024). Their biogenesis begins with the inward budding of the endosomal membrane to form multivesicular bodies (MVBs), which subsequently fuse with the plasma membrane, releasing intraluminal vesicles as exosomes. This process is regulated by the endosomal sorting complex required for transport (ESCRT) machinery (Baietti et al. 2012, Colombo et al. 2014) or by ESCRT-independent mechanisms involving lipids, cholesterol, and tetraspanins. Given their endosomal origin, exosomes are often enriched in endosome-associated proteins, including tetraspanins such as CD9 (Cluster of Differentiation 9), CD63 (Cluster of Differentiation 63), and CD81 (Cluster of Differentiation 81), as well as the ESCRT-I component TSG101 (Tumor Susceptibility Gene 101) and the accessory protein ALIX (ALG-2–interacting protein X). However, as with other EV subtypes, these proteins are not exclusive to exosomes and can also be detected in other EV populations. As such, they are best interpreted as commonly enriched features associated with endosomal vesicle formation rather than definitive markers of exosome identity (Andreu and Yáñez-Mó 2014, Théry et al. 2018). Functionally, exosomes serve as important mediators of intercellular communication by transferring proteins, lipids, and nucleic acids between cells, thereby influencing processes such as cell signaling, immune modulation, development, and disease progression (Andreu and Yáñez-Mó 2014).

Exomeres and supermeres are categories of extracellular particles (EPs). They are smaller than exosomes (less than 50 or 30 nm, respectively, Fig. 1), lack a surrounding membrane, and have been described as non-membranous, protein-rich particles that can be released into the extracellular space carrying selected molecular cargoes. However, the mechanisms underlying their biogenesis and release remain incompletely understood, and their classification as actively secreted entities may depend on cell type and experimental context (Jeppesen et al. 2022). Exomeres and supermeres are distinguished using sequential fractionation approaches, such as asymmetric flow field-flow fractionation, which separates extracellular nanoparticle populations based on size and density (Zhang et al. 2018, 2021). Exomeres tend to be enriched in proteins related to metabolism and signaling, such as FASN (Fatty Acid Synthase), ACLY (ATP-Citrate Lyase), and LGALS3BP (Galectin-3-Binding Protein) (Zhang et al. 2018). Supermeres contain molecules such as TGF-βI (Transforming Growth Factor Beta–Induced Protein), HSPA13 (Heat Shock Protein Family A Member 13), ENO2 (Enolase 2), AGO2 (Argonaute 2), ACE2 (Angiotensin-Converting Enzyme 2), and PCSK family proteins. These signatures suggest that exomeres are strongly connected to metabolic activity, lipid synthesis, and extracellular matrix organization, while supermeres carry cargoes linked to immune signaling, stress responses, and RNA-based gene regulation (Zhang et al. 2021). Although the exact pathways that generate these particles are still not well understood, current evidence suggests that they are not primarily derived from the classical endosomal route and may instead arise through alternative, non-membranous secretion processes (Yu et al. 2025).

Arthropod EV Types

In arthropods, vesicles resembling exosomes and microvesicles have been documented (Van den Brande et al. 2025). Evidence includes proteomic characterization of EVs released by a Trichoplusia ni (Hübner) (Lepidoptera: Noctuidae)-derived cell line under uninfected conditions and in response to baculovirus infection (Hausjell et al. 2023). EVs have also been extracted from the hemolymph of fifth instar nymphs of Locusta migratoria L. (Orthoptera: Acrididae) and the larvae of Helicoverpa armigera (Hübner) (Lepidoptera: Noctuidae) and Tenebrio molitor (L.) (Coleoptera: Tenebrionidae) (Van den Brande et al. 2025), hemolymph of nymphal and adult ticks Ixodes scapularis Say (Ixodida: Ixodidae) (Xu et al. 2023), larval and adult hemolymph of Allomyrina dichotoma (Coleoptera: Scarabaeidae) (Kim et al. 2015), Bombyx mori L. (Lepidoptera: Bombycidae) larval hemolymph (Kaur et al. 2025), and whole insects of Laodelphax striatellus Fallén, (Hemiptera: Delphacidae) (Lu et al. 2022). EVs have also been extracted from the salivary glands of I. scapularis (Oliva Chávez et al. 2021) and the venom of other arthropods, primarily members of Araneae (spiders) (Xun et al. 2021) and Hymenoptera (parasitoid wasps) (Wan et al. 2019). Functional studies in mosquito vectors demonstrate the release of EVs carrying viral RNA or proteins capable of infecting naïve cells, consistent with exosome-like behavior (Vora et al. 2018, Martínez-Rojas et al. 2020). Similar evidence also supports EV-mediated RNA transfer in insects, where RNAi activity has been localized to EV-enriched fractions and shown to mediate gene silencing both within and across species (Wang et al. 2025, Mingels et al. 2020, Yoon et al. 2020, Callewaert et al. 2026). However, characterization of the markers and roles of specific EV subtypes in arthropods remains limited, and further investigation is needed.

In addition to exosome- and microvesicle-like populations, “apoptotic bodies” represent an understudied component of extracellular vesicle diversity in arthropods. These vesicles are well characterized in mammals and arise during programmed cell death, containing cellular fragments, including DNA and organelles (Elmore 2007, Caruso and Poon 2018). In arthropods, the study of apoptotic bodies has received little attention. Given that many studies rely on bulk isolation approaches, these vesicles are likely present but not explicitly distinguished, which may further contribute to EV heterogeneity and classification challenges (Théry et al. 2018, van Niel et al. 2018).

More broadly, the identification of EV subtypes in arthropods is strongly influenced by the methodologies used for their isolation and validation. Common approaches, including differential ultracentrifugation and density-based centrifugation, often yield heterogeneous vesicle populations and may co-isolate non-vesicular material (Lobb et al. 2015, Li et al. 2017). In combination with the lack of exclusive molecular markers, this complicates subtype assignment and interpretation. As a result, distinctions between exosome-like and microvesicle-like vesicles in arthropods should be interpreted with caution, as they may reflect methodological limitations rather than clearly defined biological categories.

Exosome Biogenesis in Mammals

Extracellular vesicles can form through several routes, including outward budding of the plasma membrane to produce microvesicles, formation of exosomes via the endosomal pathway, fragmentation of cells during apoptosis, or through the release of non-membranous nanoparticles such as exomeres and supermeres (Saleem et al. 2025). Even though these pathways contribute to the overall diversity of EVs and EPs, the endosomal pathway that gives rise to exosomes remains the most studied, particularly because it is closely associated with selective cargo loading and regulated secretion (Saleem et al. 2025). Much of what is known about this pathway comes from mammalian systems, and these studies continue to provide the foundation for understanding how similar processes may function in arthropods.

In mammals, exosome formation typically begins with the formation of intraluminal vesicles (ILVs) within larger endosomes, resulting in an MVB (Fig. 2, Table 1). The process is typically modulated through either ESCRT-dependent or ESCRT-independent mechanisms. In the former, recognition and organization of protein cargoes on early endosomes by the ESCRT machinery initiates the formation of ILVs (Fig. 2A). ESCRT-0 components, including Hrs and STAM, identify ubiquitinated cargo (Raiborg et al. 2006), which then allows the recruitment of ESCRT-I through interactions involving VPS27 and TSG101 (Katzmann et al. 2003). ESCRT-I subsequently recruits ESCRT-II proteins such as VPS28 and VPS36 (Gill et al. 2007; Fig. 2A). Once these complexes are assembled, ESCRT-II recruits ESCRT-III subunits such as VPS20 or CHMP6 and CHMP4, which drive the inward budding events that form intraluminal vesicles (Babst et al. 2002; Fig. 2A). The final membrane scission step depends on the ATPase, VPS4, which removes ESCRT-III components from the membrane so vesicle formation can be completed (Babst et al. 1998, Fig. 2A).

Fig. 2.

EV biogenesis in mammals. A) Diagram illustrating the canonical ESCRT-dependent endosomal pathway of exosome biogenesis in mammalian systems. Ubiquitinated cargo is recognized by ESCRT-0 components, followed by sequential recruitment of ESCRT-I, ESCRT-II, and ESCRT-III complexes that drive intraluminal vesicle (ILV) formation within multivesicular bodies (MVBs). B) Illustration of ESCRT-independent ILV formation mediated by lipid remodeling. Neutral sphingomyelinase (nSMase) catalyzes the conversion of sphingomyelin to ceramide, promoting the formation of ceramide-enriched membrane microdomains. These domains, often associated with cholesterol and tetraspanins, induce negative membrane curvature and facilitate inward budding of ILVs independent of ESCRT complexes. Figure created with BioRender.

Table 1.

Core molecular components involved in exosome biogenesis, trafficking, and release, including associated genes and protein complexes, their primary functions, and supporting evidence from mammalian and arthropod systems

Process step	Mammals key genes/proteins	Arthropod orthologs and evidence	Primary function in the pathway	Regulatory context/system-level modulation	Key references (mammals)	Key references (arthropods)
1. Early endosome sorting and cargo recognition	HRS (VPS27), STAM (ESCRT-0)	Hrs required for receptor sorting and endosomal invagination in Drosophila (genetic)	Recognizes ubiquitinated cargo and PI(3)P-rich early endosomes; recruits downstream ESCRT complexes	Largely constitutive; influenced by tissue and receptor signaling load	Raiborg et al. (2001), Lloyd et al. (2002)), Mizuno et al. (2003)	Jékely and Rørth (2003)
2. ESCRT-I recruitment and cargo handoff	TSG101 (VPS23), VPS28	Erupted/TSG101 regulates endosomal trafficking in Drosophila (genetic)	Bridges ESCRT-0 to ESCRT-II; clusters cargo for ILV formation	Modulated by pathway signaling (Wnt, Notch)	Hurley and Emr (2006)	Moberg et al. (2005)
3. ESCRT-II assembly and membrane deformation	VPS22, VPS25, VPS36	VPS25 required for exosome-related signaling in Drosophila (genetic)	Scaffold linking ESCRT-I to ESCRT-III; promotes membrane curvature	Tissue-specific signaling outputs	Hierro et al. (2004)	Vaccari and Bilder (2005), Thompson et al. (2005)
4. ESCRT-III  polymerization and  ILV scission	CHMP6, CHMP4, CHMP1/5, IST1, VPS4	Shrub/CHMP4 and VPS4 required for EV formation in Drosophila (genetic)	Drives inward budding and scission; VPS4 recycles ESCRT-III	Varies with developmental stage	Henne et al. (2011), Lata et al. (2008)	Chen et al. (2024)
5. ESCRT-independent ILV formation	nSMase2; Syndecan–Syntenin–ALIX; tetraspanins	Tsp29Fb mediates viral EV cargo loading in mosquitoes (genetic/functional)	Lipid- and tetraspanin-driven ILV budding	Strongly affected by viral infection	Baietti et al. (2012), Andreu and Yáñez-Mó (2014)	Vora et al. (2018)
6. MVB transport toward release sites	Rab27a/b, Rab35; kinesin/dynein; Myosin V	Rab11 and Myosin V required for EV trafficking in Drosophila NMJ (genetic)	Positions MVBs for secretion	Modulated during feeding and infection	Stenmark (2009), Ostrowski et al. (2010)	Karim et al. (2002), Koles et al. (2012)
7. Docking and fusion with plasma membrane	SNAREs (VAMP7, SNAP-23, Syntaxin-1)	Syntaxin 1A required for EV release in Drosophila; v-SNAREs in ticks (RNAi)	Drives MVB–plasma membrane fusion	EV release increases with mating cues and infection	Chaineau et al. (2009), Vats and Galli (2022)	Gross et al. (2012) , Oliva Chávez et al. (2021)

Beyond ESCRT, several ESCRT-independent routes also contribute to MVB formation (Fig. 2B, Table 1). Ceramide-dependent budding, driven by neutral sphingomyelinase 2 (nSMase2), involves the conversion of sphingomyelin to ceramide, which accumulates in the endosomal membrane and induces negative curvature that promotes inward budding of the membrane to generate ILVs. In addition, tetraspanin-rich microdomains, and pathways involving syndecan–syntenin–alix, flotillin or caveolin rafts, and actin- or lipid-mediated (cholesterol) curvature contribute to cargo clustering and stabilization of budding sites, collectively facilitating ILV formation in the absence of canonical ESCRT machinery (van Niel et al. 2011). These pathways highlight the existence of multiple exosome biogenesis pathways and suggest similar diversity in MVB synthesis pathways may exist in arthropods.

Once formed, MVBs must be transported to the plasma membrane for fusion and release (Fig. 3, Table 1). The trafficking and release of MVBs depend on microtubule-based motor proteins, including dynein and kinesin, which move endosomal compartments toward or away from the plasma membrane (Raposo and Stoorvogel 2013). Additional regulation is provided by Rab GTPases such as Rab27a/b and Rab35, which help position and dock MVBs at the plasma membrane prior to fusion (Stenmark 2009, Ostrowski et al. 2010). The final fusion step requires SNARE proteins. In mammals, VAMP7 is located on the MVB membrane, while SNAP-23 and Syntaxin-1 reside at the plasma membrane. When an MVB reaches the cell surface, these SNARE partners engage to drive membrane fusion, allowing the intraluminal vesicles to be released as exosomes. Without this step, ILVs would remain within the cell where they are typically redirected toward lysosomal fusion and degradation, underscoring the essential role of the SNARE system in exosome secretion (Chaineau et al. 2009, Vats and Galli 2022).

Fig. 3.

Model of MVB transport and exosome secretion in mammalian cells. MVBs are transported along microtubules by kinesin and dynein motors and dock at the plasma membrane through Rab GTPase-dependent mechanisms. Membrane fusion is mediated by SNARE complexes involving VAMP7, SNAP-25, and Syntaxin-1A, resulting in exosome release into the extracellular space. Figure created with BioRender.

Exosome Biogenesis in Arthropods

In arthropods, most knowledge of exosome biogenesis comes from studies in Drosophila melanogaster Meigen (Diptera: Drosophilidae) and from arthropod-derived cell lines. These studies show that many of the core exosome biogenesis genes described in mammals are present and active in arthropods, suggesting that endosomal pathways are broadly conserved. The arthropod components of the endosomal pathways are listed in Table 1 alongside their mammalian orthologs.

Understanding whether insects use orthologous exosome biogenesis pathways to those described in mammals begins with examining the proteins that drive these processes. Many core ESCRT components exhibit strong domain conservation between insects and mammals (Table 1), suggesting that key steps in cargo recognition, membrane remodeling, and intraluminal vesicle formation likely operate similarly. At the same time, several proteins show only partial similarity or remain untested in arthropods, leaving important gaps in our understanding.

ESCRT Machinery: Strong Conservation With Extensive Functional Evidence

The ESCRT pathway is the most strongly conserved and experimentally supported machinery for exosome biogenesis in arthropods (Fig. 2A; Table 1). Core ESCRT-0 components involved in cargo recognition exhibit clear domain conservation. Both mammalian and Drosophila Hrs contain VHS, UIM, and FYVE domains that mediate binding to ubiquitinated cargo and recognition of PI(3)P-enriched endosomal membranes. Functional validation in Drosophila using FLP/FRT-generated mutants demonstrates that loss of Hrs leads to enlarged endosomes and cargo accumulation, confirming the conservation of its early endosomal sorting function (Jékely and Rørth 2003).

Similarly, ESCRT-I and ESCRT-II components show both structural conservation and functional importance. VPS23, the insect ortholog of TSG101, retains the UEV domain required for PTAP motif recognition, whereas VPS25 preserves the winged-helix domains necessary for assembly of the ESCRT-II complex and interaction with VPS22 and VPS36. In Drosophila, RNAi-mediated knockdown of ESCRT-I (VPS23) components using GAL4–UAS systems impairs intraluminal vesicle (ILV) formation and reduces exosome release, particularly at neuromuscular synapses, thereby disrupting the distribution of signaling molecules such as Wnt, Notch, and Dpp (Sánchez-López et al. 2022, Chen et al. 2024).

Downstream ESCRT-III components also show strong conservation and functional validation. Proteins such as VPS20/CHMP6, SHRUB/CHMP4, CHMP1, CHMP5, and IST1 retain conserved helical core regions required for membrane remodeling and VPS4 recruitment. In Drosophila, RNAi knockdown of CHMP1, CHMP5, or IST1 in male secondary cells reduces ILV formation in recycling endosomes and decreases exosome secretion, whereas CRISPR-TRiM knockout of SHRUB/CHMP4 at neuromuscular junctions results in marked reductions in synapse-derived exosomes and abnormal bouton growth (Chen et al. 2024).VPS4, which catalyzes the final scission step of ESCRT-mediated membrane remodeling, maintains its MIT domain and AAA+ ATPase activity in insects. Disruption via dominant-negative constructs, deletion alleles, or RNAi blocks cargo sorting and exosome release in Drosophila, mirroring its role in mammalian systems (Rodahl et al. 2009).

ESCRT-Independent Pathways: Conserved Components With Targeted Validation

In parallel with ESCRT-mediated membrane remodeling, components of ESCRT-independent exosome biogenesis pathways are also conserved between insects and mammals, although functional validation is more limited (Fig. 2B). Tetraspanins provide the clearest example. Mammalian CD63 and the Drosophila tetraspanin TSP29Fb both contain 4 transmembrane helices and a cysteine-rich EC2 loop involved in organizing membrane microdomains for cargo sorting (Moulin et al. 2023). Additional functional evidence from Aedes aegypti L. (Diptera: Culicidae) also demonstrates that TSP29Fb behaves similarly to mammalian CD63. RNAi-mediated knockdown or antibody-mediated blocking of TSP29Fb in dengue-infected cells reduces the incorporation of viral proteins into exosomes, indicating a conserved role in vesicle-associated cargo organization (Vora et al. 2018). However, whether tetraspanins broadly regulate ILV formation or primarily influence selective cargo loading in arthropods remains unresolved.

Trafficking and Accessory Proteins: Structural Conservation With Partial Functional Evidence

Several trafficking and accessory proteins involved in exosome release are conserved across arthropods but remain less well characterized functionally (Fig. 3; Table 1). Rab GTPases implicated in vesicle transport in mammals, including Rab11, Rab27, and Rab35, retain conserved structural features required for membrane association and vesicle positioning (Calero et al. 2003). Functional studies in ticks also provide indirect support for conserved roles in extracellular vesicle biology. In I. scapularis, Rab-dependent trafficking has been implicated in secretory pathways in salivary glands, and extracellular vesicles isolated from tick salivary glands and saliva display Rab-associated protein signatures (Karim et al. 2002, Xu et al. 2023). Although these findings suggest functional overlap, direct tests of Rab-specific roles in ILV formation or MVB maturation in arthropod systems remain limited.

Membrane Fusion Machinery: Conserved Structure With Emerging Functional Validation

Proteins involved in membrane fusion exhibit strong structural conservation across insects and mammals and growing functional evidence supports their role in arthropod exosome secretion. Mammalian SNAREs such as Syntaxin-1 have well-defined orthologs in Drosophila that retain conserved longin domains, SNARE helices, and membrane anchors (Kloepper et al. 2007) (Fig. 3; Table 1). Ykt6, a SNARE involved in regulated secretion, also retains its longin domain and prenylation motif in insects (Szenci et al. 2024).

Functional studies in Drosophila demonstrate that exosome release at the neuromuscular junction depends on coordinated transport, docking, and fusion of multivesicular bodies. Rab11 and the actin-based motor protein Myosin V are required to traffic MVBs to secretion sites, while Syntaxin 1A functions at the plasma membrane to mediate fusion and vesicle release. Reduction of Rab11, Myosin V, or Syntaxin 1A transcripts disrupts exosome secretion and impairs long-range Wnt signaling (Koles et al. 2012). Additional evidence from ticks further supports the evolutionary conservation of SNARE-mediated exosome release. In I. scapularis, RNAi silencing of Synaptobrevin-2 (VAMP2) and VAMP33 reduces exosome secretion from salivary gland cells, alters vesicle size distribution, and affects blood feeding and pathogen transmission, indicating that v-SNARE-dependent MVB fusion is a conserved feature of arthropod extracellular vesicle biology (Oliva Chávez et al. 2021).

Conserved Candidate Genes With Indirect Evidence Supporting Exosome Biogenesis Roles

Additional proteins appear to participate in arthropod exosome biogenesis, but their functions have not yet been confirmed through direct assays. Rab27A/B are known to dock multivesicular bodies at the plasma membrane in mammals (Ostrowski et al. 2010), and recent work in ticks shows that Rab27 influences EV size and transport (Butler et al. 2024). Reducing nSMase activity with the inhibitor GW4869 lowers EV secretion in I. scapularis, Drosophila, and mosquito cells and, in some cases, alters pathogen transmission dynamics (Shibata et al. 2017, Vora et al. 2018, Sultana et al. 2024). Rab35, another trafficking regulator, has been detected in exosome-relevant compartments and plays a role in vesicle trafficking in Drosophila, but its direct involvement in exosome release has not yet been demonstrated experimentally (Koles et al. 2012). Because Rab35 supports MVB–plasma membrane fusion in mammals, its presence in arthropods suggests a potentially conserved role that remains to be tested. Additional experimentation is necessary to determine whether these proteins fulfill the roles outlined in mammalian systems.

Uncharacterized Orthologs With Known Roles in Mammalian Exosome Pathways

Several genes with well-defined roles in mammalian exosome biogenesis remain largely untested in insects. VPS28 and VPS36, 2 ESCRT-II subunits that bridge ESCRT-I to ESCRT-III during ILV formation, localize to endosomal membranes (Moberg et al. 2005). Their presence suggests that similar membrane-remodeling steps may occur. Whereas functional studies in select systems, such as Drosophila and ticks, support roles for specific SNAREs in EV secretion, direct exosome-specific functional validation across a broader range of arthropods remains limited. This is particularly true for SNARE proteins such as VAMP7 and SNAP-23, whose roles in exosome biogenesis and release have not yet been comprehensively tested across taxa. These genes, therefore, represent important targets for future work aimed at defining how exosome biogenesis diverges or remains conserved across arthropods.

Discussion

This review highlights the core features of exosome biogenesis shared between mammals and arthropods. ESCRT components, including Hrs, Tsg101/VPS23, VPS25, ESCRT-III subunits, and VPS4, are present in both taxa and retain similar domain architectures, suggesting that the early steps of cargo recognition, ILV budding, and membrane scission are broadly conserved (Raiborg et al. 2001). Functional studies in Drosophila support this conclusion. Loss of ESCRT components leads to enlarged endosomes (Jékely and Rørth 2003), reduced ILV formation, impaired exosome release (Sánchez-López et al. 2022, Chen et al. 2024), and disrupted signaling by pathways such as Wnt, Notch, and Dpp (Thompson et al. 2005, Chen et al. 2024). Similar conservation is observed in the membrane fusion machinery. In Drosophila, Syntaxin 1A mediates fusion of multivesicular bodies with the plasma membrane, and its disruption impairs exosome secretion and Wnt signaling at the neuromuscular junction (Koles et al. 2012). Comparable roles for v-SNAREs have been demonstrated in ticks, where Synaptobrevin/VAMP proteins are required for exosome release from salivary gland cells (Oliva Chávez et al. 2021), highlighting the conservation of SNARE-dependent fusion across arthropods.

At the same time, there are clear caveats. Many arthropod studies, particularly earlier work, relied on indirect criteria such as vesicle size, density, or selected marker enrichment, reflecting evolving standards for EV characterization before the MISEV guidelines were established (Witwer et al. 2013, Raposo and Stoorvogel 2013, Welsh et al. 2024). As a result, several reports refer to “EVs” as a mixed population without clarifying whether the vesicles are exosomes, microvesicles, or a combination of both. This issue is further complicated by the likely presence of additional EVs or EPs, such as exomeres and supermeres, which have been described in mammalian systems but have not yet been systematically separated or characterized in arthropods. Consequently, many arthropod EV preparations likely contain heterogeneous mixtures of vesicles and nanoparticles, making it difficult to attribute observed phenotypes, such as antiviral protection in hemolymph or modulation of host responses by tick saliva, to specific biogenesis pathways or particle types (Tassetto et al. 2017, Van den Brande et al. 2025). In addition, most mechanistic work in arthropods has focused on only a few systems, mainly Drosophila tissues, cell lines, or vector species such as mosquitoes and ticks. This creates a strong bias toward certain developmental stages, tissues, and experimental conditions, complicating the assessment of how general these pathways are across the highly diverse arthropod phylum. Even where orthologs are present, it is not always clear that they operate in the same way. Rab GTPases, tetraspanins, and SNARE proteins often show strong sequence and domain conservation, but only a subset have been tested directly for roles in arthropod exosome biogenesis. For example, Rab11 and Myosin V have clear functions in MVB transport and exosome-mediated Wnt trafficking at the Drosophila neuromuscular junction (Koles et al. 2012), and Tsp29Fb has been linked to viral cargo loading in mosquito cells (Vora et al. 2018). In contrast, Rab27, Rab35, VPS28, VPS36, VAMP7, SNAP-23, and Syntaxin-1 are well established in mammalian exosome pathways but remain candidates rather than confirmed regulators in arthropods. This mixture of direct and indirect evidence means that insects and mammals appear broadly similar at the level of the pathway map, but the strength of experimental support is still uneven across individual components.

From an evolutionary perspective, the conservation of core components of exosome biogenesis suggest that these pathways originated early in eukaryotic evolution (Hurley 2015, van Niel et al. 2018, Callewaert et al. 2026). EVs isolated from coleopteran, dipteran, and lepidopteran cell lines display conserved size distributions and morphology (Callewaert et al. 2026). Further, as established earlier, arthropod versions of these EV proteins are often highly structurally similar to those from mammalian systems. Together, these findings provide direct experimental support for the conservation of EV-associated molecular components across invertebrate and vertebrate taxa. However, differences in physiological context, tissue specialization, and ecological interactions suggest that these conserved pathways will also include unique components or regulatory mechanisms to meet their unique biological needs, highlighting the need for broader comparative studies to distinguish conserved features from lineage-specific adaptations.

Another important caveat is the biological context. Most mammalian work has focused on EVs in cancer, immunity, and tissue repair (Colombo et al. 2014, Mathieu et al. 2019), whereas arthropod studies often emphasize antiviral defense, vector competence, or developmental signaling. As a result, our current understanding of EV biogenesis in arthropods is shaped by these research priorities. For example, EV-mediated transport of viral RNA and proteins has been extensively investigated in mosquitoes and ticks, which play key roles as disease vectors, highlighting the involvement of tetraspanins and lipid-associated pathways in these systems (Vora et al. 2018, Zhou et al. 2018). However, it remains unclear whether these mechanisms are unique to arthropods or whether the roles of these proteins are conserved across other taxa and other biological functions.

Most mechanistic insights currently available come from Drosophila and a limited number of arthropod-derived cell lines. These systems are powerful because of their genetic tools, but they represent only a small slice of arthropod diversity. Recent work has expanded to ticks, planthoppers, silkworms, beetles, and other species, showing that EVs are present in hemolymph (Kim et al. 2015, Lu et al. 2022, Xu et al. 2023, Kaur et al. 2025), salivary glands (Oliva Chávez et al. 2021), and venom, and that they can influence host–pathogen interactions, immunity, and development (Coakley et al. 2015; Szempruch et al. 2016; Van den Brande et al. 2025). However, most of these studies lacked functional testing of individual genes and their effects on EV biogenesis.

There is a clear need to develop advanced genetic and imaging techniques for non-model species. These could include CRISPR knockouts, tissue-specific RNAi, endogenously tagged EV proteins, and live imaging to track MVB dynamics in medically and agriculturally important arthropods. Such approaches would help determine whether components such as Rab27, nSMase, and specific tetraspanins play conserved roles in exosome release across vectors and pests, as well as identify lineage-specific mechanisms that may have evolved. This is especially important for species where EVs are implicated in pathogen transmission, such as ticks and mosquito vectors, since targeting exosome pathways could offer new strategies for disease control (Fasae et al. 2022, Oliva Chávez et al. 2021, Butler et al. 2024).

Non-model systems also provide opportunities to study exosome biogenesis in tissues that are anatomically unique to arthropods and absent in mammals, including midgut epithelium, Malpighian tubules, and reproductive tissues that interface directly with symbionts, parasites, and environmental stressors. Comparing how exosome pathways operate across these tissues and species may reveal arthropod-specific adaptations that are not apparent in mammalian cell culture or in Drosophila alone.

A more complete view of EV diversity in arthropods is likely to change how EV biogenesis pathways are interpreted. In particular, greater resolution of EV subtypes may reveal that molecular signals currently attributed to exosome-specific pathways instead arise from a mixture of vesicle populations produced through distinct biogenetic routes. For example, signals currently attributed to exosomes may instead reflect contributions from microvesicles or apoptotic bodies, which can share overlapping size ranges and molecular markers, as extensively documented in mammalian systems and highlighted in EV consensus guidelines (Raposo and Stoorvogel 2013, Colombo et al. 2014, Théry et al. 2018, Mathieu et al. 2019, Welsh et al. 2024). Likewise, particles that resemble exomeres or supermeres might play roles in metabolism, stress responses, or RNA transport in insects and ticks, similar to their proposed functions in mammals (Zhang et al. 2018, 2021). Distinguishing these populations will be important for correctly linking mechanistic pathways to biological outcomes.

In addition to highlighting similarities between arthropod and mammalian exosome biogenesis pathways, this review also highlights several priorities for future work. First, there is a need for systematic functional testing of candidate genes in arthropods predicted to participate in EV biogenesis based on domain conservation and mammalian studies, but for which direct experimental evidence in arthropods is still lacking. This includes ESCRT-II components such as VPS28 and VPS36, Rab27 and Rab35, and SNAREs like VAMP7, SNAP-23, Syntaxin-1, and Vamp33. Second, future work should place exosome biogenesis within the broader landscape of EV pathways by explicitly separating exosomes, microvesicles, apoptotic bodies, and non-vesicular nanoparticles in arthropod samples, using combined biochemical, imaging, and proteomic approaches. Third, extending mechanistic studies beyond Drosophila into non-model insects and other arthropods will be essential for understanding how conserved these pathways are across ecological contexts and for identifying features that might be unique to vectors, pests, or beneficial species.

Finally, arthropod exosome pathways hold strong potential for applied research. A better understanding of how EVs are formed, loaded, and released in insects and arthropods could inform strategies to enhance RNA interference, block pathogen transmission, or manipulate host–vector interactions, building on the growing evidence that EVs can carry viral and RNA cargo between cells (Tassetto et al. 2017, Vora et al. 2018, Sultana and Neelakanta 2020, Mingels et al. 2020, Yoon et al. 2020). As tools for genetic manipulation, live imaging, and single-particle analysis continue to improve, it should become possible to connect specific biogenesis mechanisms to defined functional outcomes in vivo. Bridging this gap between molecular pathways and organismal biology will be an important next step in understanding how exosome biogenesis is conserved, adapted, and diversified across arthropods and mammals.

Author Contributions

Festus K. Ajibefun (Conceptualization [equal], Writing—original draft [lead], Writing—review & editing [equal]), Marija Milosevic (Conceptualization [equal], Writing—original draft [supporting], Writing—review & editing [equal]), Xiufeng Zhang (Conceptualization [equal], Supervision [equal], Writing—review & editing [equal]), Ana M. Vélez (Conceptualization [equal], Supervision [equal], Writing—review & editing [equal], Funding acquisition [equal], Project administration [equal]), and Kristopher Silver (Conceptualization [equal], Supervision [lead], Writing—review &editing [equal], Funding acquisition [equal], Project administration [equal])

Funding

This research was supported by the USDA National Institute of Food and Agriculture (Grant# 2021-08507), the National Institute of Food and Agriculture Hatch Multistate Project (NE2443), and USDA Agricultural Research Service Cooperative Agreement (58-3022-2-025). This manuscript is contribution No. 26-120-J from the Kansas Agricultural Experiment Station, Kansas State University, Manhattan, KS, United States.

Conflicts of Interest

None declared.