Message in a vesicle – trans-kingdom intercommunication at the vector–host interface

Vector-borne illnesses contribute to the global burden of human maladies, affecting hundreds of millions of people each year (WHO, 2017). Recently, there has been a significant increase in research dedicated to the understanding of arthropods as vectors of diseases because these ailments disproportionately impact vulnerable populations around the world. Mosquitoes, ticks and sandflies, among other arthropods, secrete salivary proteins that contribute to microbial transmission (Leitner et al., 2013; Shaw et al., 2016; Šimo et al., 2017). One immune evasion strategy used by vector-borne pathogens to promote a successful infection is through the secretion of extracellular vesicles (Atayde et al., 2015; Babatunde et al., 2018; Lovo-Martins et al., 2018; Nogueira et al., 2015; Silverman et al., 2010a,b; Sisquella et al., 2017; Trocoli Torrecilhas et al., 2009). Extracellular vesicles are involved in the exchange of molecular cargo between cells. They are present in the blood, lymph, saliva and urine, and have been investigated for their use as disease biomarkers (Cheshomi and Matin, 2018; McCarthy et al., 2016; Nair et al., 2018; Zhang et al., 2018).

Infected host cells and eukaryotic pathogens secrete vesicles containing antigens, nucleic acids and other microbial determinants that exacerbate pathogenesis and modulate host immunity (Fleming et al., 2014). The cargo within these vesicles may influence how the host responds to a pathogen, and how microbes communicate with each other (Eliaz et al., 2017). In this Review, we discuss our current understanding surrounding extracellular vesicle biogenesis in the context of vector-borne illnesses. We present the most up-to-date classification of extracellular vesicles, as well as general mechanisms of action in infectious diseases. Finally, we provide specific examples of how vesicles influence vector-transmitted infections, and how this knowledge may steer the future of arthropod vector research.

Extracellular vesicles are heterogeneous and may be classified according to their biogenesis, content and biochemical features (Théry et al., 2009). Classically, they have been categorized as microvesicles and exosomes. Herein, we will briefly discuss biogenesis, characteristics, cargo and the re-classification of exosomes as a group. For a complete review of exosome and microvesicle biogenesis, please refer to van Niel et al., (2018). Importantly, it should be noted that the mechanisms discussed here have been investigated predominantly in mammalian systems, accentuating the limited understanding of extracellular vesicle biogenesis in arthropod vectors.

Exosomes are vesicles derived from multivesicular bodies (MVBs) (van Niel et al., 2018). These vesicles are formed by two different pathways: the endosomal sorting complex required for transport (ESCRT)-dependent and the ESCRT-independent network (Fig. 1A). In the ESCRT-dependent pathway, ESCRT complexes (ESCRT-0, -I, -II and -III) are involved in the budding of the membrane internally into the MVB (Baietti et al., 2012; Colombo et al., 2013; Sahu et al., 2011; Tamai et al., 2010). The ESCRT-independent pathway involves the hydrolysis of sphingomyelin into ceramide at the MVB membrane, which is mediated by the action of sphingomyelinases, also resulting in the budding of exosomes internally into the MVB (Trajkovic et al., 2008). Transport, tethering and docking of the MVB to the plasma membrane are mediated by Rab GTPases, a large family of Ras-like small GTPases that are associated with vesicular trafficking (Ostrowski et al., 2010). Once trafficked to the plasma membrane, the MVB fuses with the plasma membrane, possibly through the action of soluble N-ethylmaleimide-sensitive factor activating protein receptors (SNAREs) (Fader et al., 2009; Gross et al., 2012; Wei et al., 2017). At this juncture, exosomes are released into the extracellular space.

Exosomes contain specific cargo, including the molecules CD63, flotillin-1 and flotillin-2, G proteins and peroxidases (van Niel et al., 2018). They also carry lipids, DNA and RNA (van Niel et al., 2018). Sorting of proteins into exosomes appears to be partly regulated by ESCRT-dependent and ESCRT-independent mechanisms (van Niel et al., 2018). However, not much is understood about this process. Similarly, it remains elusive how the sorting of the genomic cargo into these nanovesicles occurs. Several of the micro (mi)RNAs identified in exosomes are not always detected in donor cells (Lunavat et al., 2015), indicating that specific small RNAs may be selected for packing into exosomes. Indeed, the sumoylated protein heterogeneous nuclear ribonucleoprotein (hnRNP) A2B1 (SUMO–hnRNPA2B1) and synaptotagmin-binding cytoplasmic RNA-interacting protein (SYNCRIP) specifically recognize so-called ‘Exo motifs’ in miRNAs for their sorting into exosomes (Santangelo et al., 2016; Villarroya-Beltri et al., 2013). Similarly, Y-box protein 1 (YBX1) specifically recognizes exosomal miRNAs for sorting into vesicles; however, no Exo motifs that YBX1 binds to have been identified to date (Shurtleff et al., 2016).

Exosomes are commonly categorized as being within the size range of 50 to 150 nm (van Niel et al., 2018). Importantly, their classification is currently evolving, and recent studies have identified distinct subpopulations (Willms et al., 2016; Zhang et al., 2018). We will focus on the recent description of exosomal sub-populations as proposed by Zhang et al. (2018).

A recent study has divided ‘exosomes’ into two distinct subpopulations based on their size: (i) Exo-S, which range from 60 to 80 nm, and (ii) Exo-L, which span between 90 and 120 nm in size (Zhang et al., 2018). Both Exo-S and Exo-L exosomes appear to be enriched for membrane- and ESCRT-associated proteins, integrins and Rabs. Proteins involved in G-protein and STAT signaling are overrepresented in Exo-L subpopulations. By contrast, Exo-S vesicles contain molecules that are often detected in the endosome (Zhang et al., 2018). The authors suggested that Exo-S represent ‘bona fide exosomes’ and proposed the use of flotillin-1 and flotillin-2 as markers for their identification. The lipid content of Exo-L and Exo-S nanovesicles varies depending on the cell of origin. Both subpopulations are enriched for small RNAs, miRNAs and tRNAs; however, only Exo-L vesicles contain a RNA peak at 315 base pairs, which could represent a population of small RNAs that is unique to this vesicle type. Exo-S and Exo-L are enriched with DNA between 2 kb to 4 kb. Finally, Exo-L vesicles have been described to be the only subpopulation that show a tropism for lymph nodes, indicating that this type of nanovesicles may facilitate tumor dissemination in mammalian systems (Zhang et al., 2018).

Exomeres are a recently described type of vesicle that lack an external membrane and have sizes smaller than 50 nm (Zhang et al., 2018). As such, the knowledge about these small vesicles remains obscure. They are enriched in proteins involved in metabolism, cytoskeleton and glycolysis. A potential protein marker for this subpopulation is the heat shock protein (Hsp90)-β. Exomeres have been described to carry higher DNA content, but fewer lipids and less RNA material when compared to Exo-S and Exo-L vesicles (Zhang et al., 2018).

Microvesicle secretion is stimulated by the accumulation of Ca²⁺ in the cytoplasm and is a product of direct budding of the plasma membrane (Sedgwick and D'Souza-Schorey, 2018; Théry et al., 2009) (Fig. 1B). Flippases and floppases use Ca²⁺ to reorganize lipids within the cell membrane, inducing membrane budding through the accumulation of ceramide, phosphatidylserine (PS) and phosphatidylethanolamine (PE) in the outer layer (Beer et al., 2018; Théry et al., 2009; van Niel et al., 2018) (Fig. 1B). In addition to lipid rearrangements, cytoskeleton dynamics appear to play a role in the shedding of microvesicles. In cancer cells, microvesicle shedding is triggered by the small GTPase RhoA, which leads to the phosphorylation of cofilin 1 (Li et al., 2012). The ESCRT-III protein charged multivesicular body protein 4B (CHMP4B) appears to be associated with microvesicle release from cardiomyocytes (Xu et al., 2017). In addition, the tumor susceptibility gene (TSG101), an ESCRT-I protein, and the vacuolar protein sorting-associated protein 4 (VPS4, which has VPS4A and VPS4B forms in mammals), an ESCRT-III associated molecule, have also been implicated in microvesicle budding (Choi et al., 2018; Nabhan et al., 2012).

Microvesicles are commonly described as being within the range of 100 nm to 1 µm in size (Théry et al., 2009). Microvesicles contain proteins, lipids and genetic material, which are delivered to recipient cells. Their cargo is selectively sorted by specific proteins and depend on the secreting cell (van Niel et al., 2018). For example, vesicle-associated membrane protein 3 (VAMP-3) is involved in the packaging of membrane-type 1 matrix metalloprotease (MT1-MMP; also known as MMP14) into microvesicles (Clancy et al., 2015). Likewise, microvesicles have been shown to contain selectively sorted RNA cargo, including the enrichment of mRNAs containing a zip code sequence of 25 nucleotides in microvesicles (Bolukbasi et al., 2012). These findings suggest that cells may recognize and target certain mRNAs for packaging. Thus, microvesicles are thought to be involved in infectious diseases (Clancy et al., 2015; Hoshino et al., 2015; Moroishi et al., 2016).

Extracellular vesicles have an important role in the establishment of infectious diseases (Schorey et al., 2015). As this area of research is an emerging field, we choose to focus on specific models and relate general concepts of microbial invasion, pathogenesis, host immunity and immune evasion. For more information, please refer to the following reviews: Altan-Bonnet (2016); Marti and Johnson (2016); Raab-Traub and Dittmer (2017); and Schorey and Harding (2016).

Extracellular vesicles are involved in immune activation and antigen presentation during Mycobacterium tuberculosis infection. M. tuberculosis manipulates the secretion of extracellular vesicles in mammalian cells by increasing the abundance of immune-related proteins (Diaz et al., 2016), and packaging bacterial molecules within host-derived exosomes (Giri et al., 2010; Kruh-Garcia et al., 2014). M. tuberculosis proteins within these exosomes activate the host immune response in the recipient cell (Giri et al., 2010) (Fig. 2A). Interestingly, exosome release increases during M. tuberculosis infection of bone marrow-derived macrophages (BMDMs) and wild-type or Rab27a^−/− mice (Smith et al., 2017). The importance of exosome secretion during M. tuberculosis infection was determined using Rab27a^−/− mice, as Rab27 is necessary for docking and tethering of MVBs (Ostrowski et al., 2010). Infection of Rab27a^−/− mice with M. tuberculosis results in higher bacterial burden and diminished CD4⁺ T cell activation than for wild-type mice (Smith et al., 2017). Furthermore, treatment of bone marrow-derived macrophages with exosomes purified from infected Rab27a^−/− mice leads to diminished secretion of chemokine ligand 1 (CCL1), interleukin 1a (IL-1a), chemokine ligand 2 (CXCL2), CCL5 and tumor necrosis factor (TNF) when compared to the control treatment. The reduced secretion of these pro-inflammatory factors results in inefficient immune activation and bacterial clearance (Smith et al., 2017). Taken together, these findings indicate that exosome biogenesis, and specifically Rab27a, is necessary for an immune response against M. tuberculosis.

Several parasites use extracellular vesicles to diminish host immune responses (Buck et al., 2014; Eichenberger et al., 2018; Singh et al., 2011). Heligmosomoides polygyrus, a gastrointestinal nematode, secretes exosomes that are structurally distinct in their lipid signature when compared to host exosomes (Simbari et al., 2016). These parasitic exosomes contain miRNAs, which are similar to host small RNAs. Additionally, they carry Y RNAs and argonaute, a protein required for RNA-mediated gene silencing. Treatment of mammalian cells with H. polygyrus-derived exosomes results in downregulation of mitogen-activated protein kinase 1 (Mkp1) and interleukin 1 receptor-like 1 (Il1rl1) gene transcription (Buck et al., 2014). H. polygyrus-derived exosomes also mitigate eosinophil migration through the reduced secretion of IL-5 and IL-13 (Buck et al., 2014) (Fig. 2B). Interestingly, H. polygyrus-derived vesicles are preferentially internalized by macrophages activated through the alternative pathway (M2; AAMФ), which are important during anti-parasite immune responses. Treatment of pre-AAMФ and AAMФ macrophages with these vesicles results in decreased secretion of resistin-like molecule α (RELMα), Ym1 (also known as CHIL3), IL-10 and the expression of the mannose receptor CD206 (also known as MRC1) and Il1rl1 (Coakley et al., 2017). These findings demonstrate that nematode vesicles reduce anti-parasite immune responses. In addition, extracellular vesicles from another parasitic nematode, Nippostrongylus brasiliensis, mitigate host defenses (Eichenberger et al., 2018). Thus, an emerging concept in the field is that extracellular vesicles may have anti-inflammatory features.

A recent study described a role for vesicles coated with the ALG-2-interacting protein X (ALIX; also known as PDCD6IP), which is associated with the ESCRT-III complex, in the packaging of the non-enveloped hepatitis A virus (HAV). This vesicle-bound form of HAV is referred to as ‘enveloped hepatitis A virus’ (eHAV) (Fig. 2C); eHAVs are infectious, contain a capsid antigen and are resistant to neutralizing antibodies (Feng et al., 2013). Interestingly, reducing the expression of VSP4 and ALIX affects the release of eHAVs, but not other ESCRT components (Feng et al., 2013). Moreover, the human immunodeficiency virus (HIV) and herpes simplex virus 1 (HSV-1) exploit extracellular vesicles to enhance their spread into uninfected cells (Arakelyan et al., 2017). Conceptually, an emerging paradigm is that viruses may hijack the extracellular vesicle machinery to invade mammalian cells.

Acanthamoeba castellanii is a free-living amoeba that causes the eye infection Acanthamoeba keratitis. The disease pathology is primarily associated with metalloproteases and serine proteases at the site of infection, which are secreted by A. castellanii within nanovesicles and absorbed by mammalian cells to promote apoptosis (Gonçalves et al., 2018) (Fig. 2D). Although the exact mechanism that leads to cell apoptosis is unknown, it was determined that the toxicity of vesicles to mammals was dependent on the activity of the proteases within their cargo. Cell death at the infectious site by A. castellanii through extracellular vesicles is likely connected with the pathology of ocular keratitis (Gonçalves et al., 2018). Other examples of how vesicles promote pathogenesis during infectious diseases are discussed below.

The discipline of parasitology is at the forefront of research for extracellular vesicles and vector-borne diseases. Once inside the host or the arthropod, some parasites appear to hijack the extracellular vesicle machinery to tailor the mammalian immune response. Here, we will elaborate on how some parasites manipulate the host machinery, before discussing how cell-to-cell communication between the arthropod vector, the mammalian host and the microbial pathogen affects molecular signaling in order to promote or hinder infection.

Malaria is a deadly vector-borne illness caused by Plasmodium spp. These parasites are deposited into the skin by mosquitoes and eventually spread to the liver. Following an incubation period in the liver, parasites are released into the bloodstream and infect red blood cells. Much of the research surrounding malaria extracellular vesicles has been focused at the blood stage, specifically on characterizing vesicles secreted by infected red blood cells (iRBCs) (Fig. 3A). Subsequent to infection with malaria parasites, iRBCs secrete significantly more extracellular vesicles than uninfected cells (Mantel et al., 2016; Regev-Rudzki et al., 2013). Extracellular vesicles derived from iRBCs contain many factors, including antigens, virulence factors, mammalian- and parasite-derived nucleic acids, and the RNAi machinery (Babatunde et al., 2018; Mantel et al., 2016; Sampaio et al., 2018; Wang et al., 2017).

Extracellular vesicles from Plasmodium iRBCs contribute to the severity of disease, specifically to vascular dysfunction during cerebral malaria (El-Assaad et al., 2014). Recently, it has been reported that extracellular vesicles from iRBCs contain complexes of miRNAs and argonaute-2 (Mantel et al., 2016). Delivery of vesicle cargo miRNAs, specifically human miR-451a, contributes to the downregulation of genes involved in endothelial barrier function (Fig. 3A). Indeed, miRNA delivery by vesicles has been shown to increase permeability of the endothelium, indicating a potential means for parasites to breach the blood–brain barrier (Mantel et al., 2016). Moreover, P. falciparum erythrocyte membrane protein 1 (PfEMP1), which is involved in iRBC adherence and sequestration in blood vessels, was identified in extracellular vesicles (Sampaio et al., 2018). Regulation of host gene expression and increased endothelial adherence suggests a mechanism of enhanced disease severity that is mediated by extracellular vesicles.

iRBCs secrete vesicles that contain parasite nucleic acids (Babatunde et al., 2018; Sisquella et al., 2017), including Plasmodium genomic DNA, which may activate cytosolic DNA sensors in monocytes (Sisquella et al., 2017). Delivery of parasite DNA also promoted type I interferon and chemokine production, which is indicative of a stimulator of interferon genes (STING) response (Fig. 3A) (Sisquella et al., 2017). The transport of parasitic cargo between cells demonstrates the important role that vesicles play in engaging host immune responses.

Leishmania spp. are a group of over twenty protozoan parasites that are transmitted by various species of sandfly. Leishmaniasis manifests into three clinical forms: cutaneous, mucocutaneous and visceral leishmaniasis. Leishmania spp. were first shown to secrete exosomes by Silverman and colleagues, who observed that extracellular vesicle release increases at the mammalian host temperature (37°C) and specific cargo is enriched at acidic pH, representing conditions during an early infection (Silverman et al., 2010a). Unlike what is seen with other intracellular pathogens, infection with Leishmania parasites alone does not increase extracellular vesicle secretion (Cronemberger-Andrade et al., 2014). Leishmania promastigote exosomes carry heat-shock proteins and virulence factors, including the metalloprotease GP63 and the protein tyrosine phosphatase L. major (Lm)-PRL-1, which contribute to parasite survival in the mammalian host (Leitherer et al., 2017; Silverman et al., 2010a). Leishmania exosomes isolated from amastigotes contain siRNAs and tRNA-derived small RNAs (Lambertz et al., 2015). The RNA content of Leishmania spp. has a surprisingly high degree of similarity between the types found in the Old and New World despite them being transmitted by different vectors. Collectively, these results therefore suggest a conserved mechanism for parasite RNA packaging.

Exosomes derived from Leishmania parasites and infected macrophages aid in parasite survival. Indeed, it was demonstrated that pretreating monocytes with exosomes derived from L. donovani upregulates the production of IL-10 and decreases that of IL-8 and TNF, suggesting that they have an immunosuppressive role (Silverman et al., 2010b) (Fig. 3B). L. major exosomes appear to promote T helper 2 (T_h2) cell polarization, which is an adaptive immune response permissive to Leishmania infection (Silverman et al., 2010b). Specifically, L. major exosomes downregulate the expression of inducible nitric oxide synthase (iNOS, also known as NOS3) and promote activation of protein tyrosine phosphatases (PTPs), resulting in impaired parasite killing by macrophages. This immunomodulation was shown to be dependent on the expression of Leishmania GP63 (Hassani et al., 2014). Taken together, Leishmania exosomes drive an immune response that favors parasite infection by limiting inflammation.

Trypanosomes are unicellular flagellate protozoa divided into two major classes: salivarian and stercorarian (Kaufer et al., 2017). T. cruzi is a stercorarian trypanosome that is transmitted by triatomine bugs and causes Chagas disease. Even by 1991, the shedding of surface antigens by membrane vesicles from T. cruzi had been observed (Gonçalves et al., 1991). Trypomastigote vesicles carry molecules involved in nucleic acid binding, heat-shock proteins and mucins, as well as parasite surface components (Bayer-Santos et al., 2013; Gonçalves et al., 1991; Nogueira et al., 2015). Because T. cruzi comprises several groups and subgroups, there are strain-dependent differences in vesicle protein content, pathogenesis and immune evasion (Neves et al., 2014; Ribeiro et al., 2018; Wyllie and Ramirez, 2017).

Extracellular vesicles from Trypanosoma spp. possess immunomodulatory properties that contribute to parasite survival. Once T. cruzi enters the host through a wound and penetrates the mucous membranes, it arrives in the bloodstream and becomes a target of the complement system. As a mechanism of host immune evasion, both T. cruzi trypomastigotes and the host blood cells that interact with them secrete vesicles that promote resistance to the lectin pathway of complement-mediated lysis (Wyllie and Ramirez, 2017) (Fig. 3C). Although the exact mechanism is unknown, it was speculated that this could be due to the inhibition of the C3 convertase, which is a key component during the complement activation (Lidani et al., 2017; Wyllie and Ramirez, 2017). Interestingly, vesicles from one strain do not confer complement resistance to trypanosomes from a different group (Wyllie and Ramirez, 2017). This strain-dependent resistance highlights the level of specificity in the ability of the parasite to protect itself from the immune system of its host.

Extracellular vesicles derived from T. cruzi also promote an immunosuppressive response through the induction of IL-4 and IL-10 in the heart tissue, and a reduction of iNOS and TNF production by macrophages (Lovo-Martins et al., 2018; Trocoli Torrecilhas et al., 2009). Accordingly, priming of macrophages with extracellular vesicles from T. cruzi results in fewer macrophages that are positive for MHC class II molecules, suggesting that vesicles can suppress macrophage stimulation and dampen T cell activation (Lovo-Martins et al., 2018). Such an immunosuppression is further indicated by the decreased expression of CD86 (B7-2), a co-stimulatory molecule involved in T cell activation (Lovo-Martins et al., 2018).

T. cruzi-derived vesicles also contain phosphatases that increase parasite adhesion to host macrophages (Neves et al., 2014). Pretreatment with vesicles from T. cruzi has been shown to contribute to increased invasion of heart tissues by the parasite, with exacerbated lesions and higher mortality (Trocoli Torrecilhas et al., 2009). As Chagas disease is primarily a disease of the heart and the nervous system, this enhanced pathology emphasizes the important role of extracellular vesicles in promoting disease.

Another example of how nanovesicles are involved in the pathogenesis of a trypanosome is the case of T. brucei, a salivarian trypanosome that causes African sleeping sickness (African trypanosomiasis). These parasites produce filamentous structures termed ‘membrane nanotubes’ that release extracellular vesicles (Szempruch et al., 2016). These vesicles rapidly fuse with the membranes of mammalian red blood cells and alter their composition, promoting erythrophagocytosis and resulting in a smaller number of circulating red blood cells (Szempruch et al., 2016). Decreased number of red blood cells provokes the development of anemia, which is the primary cause of morbidity associated with African trypanosomiasis. These results indicate the importance of T. brucei nanovesicles in mediating disease pathology.

Brugia malayi, the causative agent of lymphatic filariasis, is a parasitic roundworm spread by mosquitoes. During a blood meal, mosquitoes deposit larvae into the skin, which then develop as adult worms within the lymphatics. Brugia malayi eventually reaches the bloodstream as microfilariae. Extracellular vesicles from Brugia spp. are internalized by macrophages (Zamanian et al., 2015) and secreted during all intra-mammalian life cycle stages (Harischandra et al., 2018). The protein content of these vesicles differs depending on the life cycle stage and the sex of the worms (Harischandra et al., 2018). For instance, vesicles secreted by female Brugia malayi are enriched in immunomodulatory molecules, such as bioactive microRNAs (Zamanian et al., 2015). Interestingly, RNAseq analysis has identified miRNAs in Brugia spp. vesicles that are homologous to host miRNAs, including miRNAs that are either identical or similar to let-7. The let-7 miRNA family targets genes involved in innate immune pathways, cell proliferation and macrophage polarization (Banerjee et al., 2013; Teng et al., 2013), suggesting a possible mechanism for how this extracellular parasite modulates the host immune response.

Mosquitoes, kissing bugs and sandflies spread many pathogens. Therefore, the role of extracellular vesicles in the transmission of pathogens through their vectors has become an attractive area of research. In this section, we will discuss some of these guiding principles.

During the Leishmania life cycle, parasites infect the midgut of sandflies before being transmitted to a mammalian host. Leishmania parasites constitutively release exosome-like vesicles directly into the sandfly midgut (Atayde et al., 2015) (Fig. 4A). These vesicles are found in the sandfly inoculum and are subsequently co-egested with parasites during the blood meal. Extracellular vesicles derived from midgut-isolated Leishmania also carry virulence and immunological factors, including GP63 and Hsp70 (Atayde et al., 2015). Importantly, these extracellular vesicles appear to worsen the disease pathology. Co-inoculation of L. major parasites with purified exosomes increased the lesion size and footpad swelling in mice. Mice injected with both exosomes and Leishmania also had higher parasite loads in non-necrotic lesions and increased secretion of a number of cytokines, including IL-2, IL-4, IFN-γ, IL-17a, IL-23 and IL-10, than those injected with Leishmania alone (Atayde et al., 2015). Hence, these findings demonstrate that extracellular vesicles are relevant in the arthropod vector and clearly influence the mammalian immune response.

Extracellular vesicles secreted within tsetse flies have an important role in the communication between trypanosomes, particularly during social motility (Eliaz et al., 2017). T. brucei move through semi-solid surfaces in swarms. ‘Scouts’ sense the environmental conditions at the edges and send signals to other individuals within the swarm to guide their movement, in a process defined as ‘social motility’ (Oberholzer et al., 2010). Furthermore, it has been postulated that trypanosomes send stress-condition signals into the environment to halt parasite movement (Eliaz et al., 2017). In experiments with T. brucei, it has been demonstrated that the procyclic forms typically found in the midgut of the tsetse fly are capable of secreting extracellular vesicles when trans-splicing of mRNAs is disrupted (Eliaz et al., 2017). During disrupted trans-splicing, the spliced leader (SL) RNA begins accumulating in the parasite, which leads to the formation of granules (Fig. 4B). These SL RNA-containing granules are secreted by the affected trypanosome in the form of exosomes and are then endocytosed by neighboring parasites. Intriguingly, secreted exosomes significantly enhance the repulsion observed between parasites. Repulsion signals from unfit parasites could be a method of promoting a successful colonization of the tsetse fly (Eliaz et al., 2017), suggesting another mechanism of vesicle-mediated communication to ensure an infection.

Extracellular vesicles originating from arthropod vectors are becoming apparent as an important strategy for immune evasion during microbial transmission. During infection with Langat virus (LGTV), the tick cell line ISE6 (Oliver et al., 2015) secretes exosomes that contain cargo from both the virus and the vector (Zhou et al., 2018). Indeed, tick cells infected with LGTV release higher numbers of exosomes than uninfected cells, and these exosomes can transmit the virus to an uninfected organism. Interestingly, virus purified from infected exosomes are also capable of colonizing human HaCaT keratinocytes (Zhou et al., 2018). These findings reveal an intriguing interaction between vector-derived vesicles and mammalian host cells. Similarly, other viruses hijack extracellular vesicles from the arthropod vector for their transmission. For instance, the Dengue virus serotype 2 (DENV2) uses extracellular vesicles derived from mosquitoes to infect mammalian cells (Vora et al., 2018). During Dengue virus infection, mosquito-derived vesicles carry viral proteins and a full-length DENV2 genome. Importantly, DENV2 transmission is dependent on vesicle secretion by the vector, which may occur through the interaction between the tetraspanin domain-containing glycoprotein Tsp29Fb, a mosquito homolog of the exosomal marker CD63, and the viral E protein (Vora et al., 2018). Although these findings were primarily obtained in vitro, they clearly suggest that arthropod extracellular vesicles may transmit viruses. Future efforts are needed to fully elucidate the contribution of vector-derived extracellular vesicles in pathogen transmission at the vector–host interface.

Within the past decade, research on extracellular vesicles has increased exponentially in the field of infectious diseases. This is partially due to the better-appreciated role these vesicles have in directing host immunity. Although we do not fully understand how arthropod vectors aid in the establishment of illnesses, an emerging principle is that pathogens exploit extracellular vesicles secreted by their vectors. Indeed, independent groups have shown that vector-borne microbes can change the protein and miRNA content of vector and host vesicles by adding pathogen-derived components that assist in microbial spreading (Sampaio et al., 2018; Vora et al., 2018; Zhou et al., 2018). Dissection of these underlying mechanisms is clearly a priority, and there are outstanding questions that should be explored. For instance, do microbial pathogens inhibit inflammation by modifying the oxidative state of arthropod vesicles? Do the skin microbiome and metabolites present at the bite site influence the content of arthropod vesicles? Furthermore, what is the contribution of different molecular entities (e.g. nucleic acids, carbohydrates, lipids or proteins) to an immune response against a microbial pathogen and/or an arthropod vector?

Blood-feeding arthropods secrete a vast arsenal of immunomodulatory molecules (Leitner et al., 2013; Shaw et al., 2016; Šimo et al., 2017). How these effector molecules affect immune signaling is another active area of investigation. Experimental evidence suggests that SNAREs are important for vector feeding (Browning and Karim, 2013; Karim et al., 2005; Villarreal et al., 2013), indicating that these effectors may be secreted inside nanovesicles. Saliva from different arthropod species carry classical markers of extracellular vesicles (Díaz-Martín et al., 2013; Tirloni et al., 2015, 2014), as well as anti-inflammatory miRNAs (Hackenberg and Kotsyfakis, 2018). These observations suggest that nanovesicles may modulate the host immune response at the bite site, resulting in successful feeding. How does vesicle secretion assist arthropod vectors during feeding? Does arthropod feeding show a different dynamic state when abiotic factors (e.g. temperature) and biotic factors (e.g. commensals) are present? Moreover, how does prolonged versus short arthropod feeding affect trans-kingdom intercommunication? Are secretory mechanisms of extracellular vesicles conserved between vertebrate and non-vertebrate animals?

If the underlying mechanics of vesicle secretion are evolutionarily conserved among blood-feeding arthropods, it becomes paramount to identify cellular targets. Equally important will be to define the transduction networks that are affected by vesicles during pathogen transmission. For example, it remains unclear whether vector-secreted vesicles interact with mammalian cells through endocytosis or receptor-mediated processes. Similarly, we do not yet know whether ligands present on the vesicle membranes dictate the organotropism of arthropod vesicles. For example, integrins direct the organotropism of cancer cell-derived exosomes (Hoshino et al., 2015). Thus, it is possible that integrins on the membranes of vector-derived vesicles may govern the interaction of exosomes with specific cells and organs. In this regard, another interesting question is whether the effect of arthropod vesicles is limited to the regulation of immune responses, or whether they also affect metabolism and physiology? Does the extent of co-evolutionary history between vectors and pathogens affect vesicle secretion?

In this Review, we attempted to summarize the current knowledge of extracellular vesicles in vector-borne diseases. We also highlighted the most critical questions that would fundamentally advance the field of arthropod-borne illnesses. Such an undertaking is monumental because genetic manipulation is unavailable for most arthropod vectors. Similarly, specific tools that completely abolish vesicle secretion in vivo have not been developed. However, with the advent of genome editing and the power of systems biology, mechanistic analysis focusing on the arthropod vector is becoming more realistic.

We would like to thank Erin McClure Carroll for editing this manuscript. We also acknowledge Holly Hammond for assistance with the illustrations.