XYZ summary ### Ewing's Sarcoma example: > [!Abstract]+ Extracellular Vesicles in Reprogramming of the Ewing Sarcoma Tumor Microenvironment [(Front Cell Dev) 2021 ](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/) Intercellular communication within the tumor microenvironment is emerging as a crucial mechanism for cancer cells to establish immunosuppressive and cancer-permissive environments, potentially leading to metastasis. Altering this communication within the tumor microenvironment, thereby preventing the transfer of oncogenic signals and molecules, represents a highly promising therapeutic strategy. > > To achieve this, extracellular vesicles (EVs) offer a candidate mechanism as they are actively released by tumor cells and enriched with proteins and RNAs. EVs are membrane-bound particles released by normal and tumor cells, that play pivotal roles in intercellular communication, including cross-talk between tumor, stromal fibroblast, and immune cells in the local tumor microenvironment and systemic circulation. EwS EVs, including the smaller exosomes and larger microvesicles, have the potential to reprogram a diversity of cells in the tumor microenvironment, by transferring various biomolecules in a cell-specific manner. Insights into the various biomolecules packed in EwS EVs as cargos and the molecular changes they trigger in recipient cells of the tumor microenvironment will shed light on various potential targets for therapeutic intervention in EwS. This review details EwS EVs composition, their potential role in metastasis and in the reprogramming of various cells of the tumor microenvironment, and the potential for clinical intervention. > ## Extracellular Vesicles > > Extracellular  vesicles (EVs) are membrane-bound particles in the subcellular size  range carrying cargo released from cells. There exists three main types  of EVs which are distinguished based on their size, biogenesis, and  content: exosomes, microvesicles, and apoptotic bodies ([Figure 1](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/figure/F1/); [Kerr et al., 1972](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B66); [Johnstone et al., 1987](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B62); [Heijnen et al., 1999](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B50)). Apoptotic bodies are 1–5 μm in size, produced during apoptosis of cells ([Table 1](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/table/T1/) and [Figure 1C](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/figure/F1/)). Hence, they contain cellular fragments, such as intact organelles and nuclear fractions ([Kerr et al., 1972](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B66)).  Unlike the other two types of EVs, apoptotic bodies do not have a known  function in intercellular communication under normal physiological  conditions ([Colombo et al., 2014](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B17)). > ### TABLE 1 > Characteristics of extracellular vesicles subtypes. > > |Type|Size|Markers|Process of formation| > |---|---|---|---| > |Exosomes|40 nm–140 nm|CD63, CD81, CD82, CD9, Tsg101, and Alix|Endosomal sorting| > |Microvesicles|100 nm–1 μm|Annexin A1, CD40L, Selectins (CD62), and Integrins|Budding of plasma membrane| > |Apoptotic bodies|1 μm–5 μm|Annexin V, Caspase-3, and gp96|Protrusions and blebbing of plasma membrane| > > Exosomes are the smallest of the three types of EVs, with a diameter of 40–140 nm. Exosomes are formed by the budding of the membrane of early endosomes to form intraluminal vesicles (ILVs), turning the endosomes into multivesicular bodies (MVBs) ([Table 1](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/table/T1/) and [Figure 1A](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/figure/F1/); [Hurley, 2008](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B57)). Exosomes are released from cells upon the fusion of MVBs with the plasma membrane. The endosomal sorting complexes required for transport (ESCRT) pathway is the key modulator of MVB synthesis in which the ESCRT protein complexes guide intracellular vesicles to their destinations within cells and also direct MVBs toward the plasma membrane ([Colombo et al., 2013](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B16)). > > Other mechanisms, such as the tetraspanin and ceramide pathways, are known to regulate the formation of ILVs. The tetraspanin CD63 was shown to regulate the sorting of proteins into ILVs and thus the budding of exosomes ([van Niel et al., 2011](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B137)). On the other hand, the sphingolipid ceramide triggers the budding of ILVs ([Trajkovic et al., 2008](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B133)). It was also recently found that caveolin-1, a membrane-associated scaffolding protein, regulates levels of cholesterol in MVBs and hence the biogenesis of exosomes ([Albacete-Albacete et al., 2020](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B1)). Given that ILVs are composed of membrane components, derived from the budding of endosomes via multiple pathways, these proteins and lipids may be present on the surface of exosomes and differ depending on their mechanism of formation. > > Although the tetraspanins CD63, CD81, and CD9 are not required for exosome biogenesis via the ESCRT-dependent pathway, they are enriched on the surface of exosomes ([Escola et al., 1998](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B25); [Heijnen et al., 1999](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B50); [Kowal et al., 2016](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B74)). Hence, these three tetraspanins have been traditionally used as markers for exosome identification ([Table 1](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/table/T1/); [Kowal et al., 2016](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B74)). However, the tetraspanin-enriched exosomes lack cytosolic proteins such as pyruvate kinase and enolase previously thought to be common in exosomes, suggesting the presence of a class of non-classical exosomes devoid of these three tetraspanin markers ([Jeppesen et al., 2019](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B58)). Meanwhile, other previously identified exosomal markers, such as MHC molecules and heat shock protein 70, were refuted since they are also present in other EVs ([Kowal et al., 2016](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B74)). > > Microvesicles are a larger class of EVs with a diameter ranging from 100 nm to 1 μm ([Heijnen et al., 1999](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B50)). The synthesis of microvesicles results from the outward budding and shedding of the plasma membrane ([Table 1](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/table/T1/) and [Figure 1B](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/figure/F1/); [Heijnen et al., 1999](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B50)). Intracellular Ca2+ levels modulate the biogenesis of microvesicles by affecting the plasma membrane and cytoskeleton. Increased concentrations of cytosolic Ca2+ activate the protease calpain, leading to the proteolysis of cytoskeletal proteins ([Pasquet et al., 1996](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B99)). Following calpain proteolysis, Ca2+ also activates gelsolin, which cleaves actin filaments, enabling cytoskeleton remodeling for the release of microvesicles in lipid rafts or caveolae in the plasma membrane ([Sun et al., 1999](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B124); [del Conde et al., 2005](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B22)). The proteins on the surface of microvesicles are acquired from the plasma membrane of donor cells, and hence may serve as markers for this type of EVs, including the previously described tetraspanins ([Kowal et al., 2016](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B74)). These proteins may also provide indications of the state of donor cells, as microvesicles from activated and apoptotic cells exhibit different surface markers ([Jimenez et al., 2003](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B60)). Recently, annexin A1 has emerged as a specific protein marker for microvesicles ([Table 1](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/table/T1/)). Annexin A1 was found in EVs similar in size to microvesicles that are produced by direct shedding from the plasma membrane of donor cells, but was undetectable in the CD63, CD81, and CD9 tetraspanins-positive exosomes ([Jeppesen et al., 2019](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B58)). Annexin A2 was also discovered in non-exosomal EVs, in a population potentially overlapping with annexin A1 EVs and microvesicles ([Jeppesen et al., 2019](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B58)). Given that EVs are lipid-bound vesicles generated via specific molecular pathways, their lipid composition may serve as a marker for each class of EV ([Llorente et al., 2013](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B82); [Singhto et al., 2019](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B118)). Indeed, exosomes and microvesicles contain different sets of sphingolipids, exemplified by the validation of ceramide phosphates as a specific marker of urinary exosomes given its undetectable level in urinary microvesicles ([Singhto et al., 2019](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B118)). These novel markers may potentially be used to differentiate the two types of EVs, which is crucial for the characterization of their specific biological functions. > > In addition to membrane-associated molecules derived from their membrane of origin, both exosomes and microvesicles contain soluble proteins and nucleic acids as cargos ([Valadi et al., 2007](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B136); [Guescini et al., 2009](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B44)). Though the specific mechanisms of cargo sorting into EVs are unclear, ESCRT-dependent and ESCRT-independent pathways are believed to contribute to this process ([van Niel et al., 2018](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B138)). Selectivity for specific protein cargo is further regulated by diverse post-translational modifications, such as ubiquitination, sumoylation, and phosphorylation ([Blot et al., 2004](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B11); [Saman et al., 2012](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B107); [Villarroya-Beltri et al., 2013](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B141)). Caveolin-1 also controls sorting of exosomal protein cargo, in addition to exosome biogenesis ([Albacete-Albacete et al., 2020](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B1)). Regulation of the RNA content in EVs was shown to be achieved by RNA-binding proteins targeting RNAs to the site of EV synthesis to protect them from degradation or by recognizing specific sequence motifs to control exosomal sorting ([Villarroya-Beltri et al., 2013](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B141)). > > Meanwhile, the cargo sorting process may be altered in cancer to allow packaging of cancer-specific cargo in EVs ([Welton et al., 2010](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B149); [Ji et al., 2013](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B59); [Thakur et al., 2014](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B126)). Aberrant expression of oncogenic molecules like EWS-FLI1 and EV biogenesis molecules like caveolin-1 may deregulate various mechanisms in cells, thereby altering the repertoire of proteins, RNAs, and other biomolecules packaged in their EVs. Additionally, some pathways may potentially be manipulated during tumorigenesis to allow the predominant secretion of one type of EV, such as in hepatocellular carcinoma where long non-coding RNAs are upregulated to enhance exosome secretion ([Cao et al., 2019](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B13); [Yang et al., 2019](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B156)). Once released by donor cell, the uptake of EVs by a recipient cell may occur through two processes: endocytosis or direct fusion with the plasma membrane of a recipient cell ([Morelli et al., 2004](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B91); [Parolini et al., 2009](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B98)). Internalization of EVs releases the cargo of donor cells into recipient cells, enabling intercellular communication. In recipient cells, EVs may induce changes in cell phenotype by the transfer of morphogens, such as Hedgehog proteins in humans and Wingless proteins in _Drosophila_, to establish a gradient in tissue, whereby signals state cell maintenance or differentiation ([Greco et al., 2001](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B40); [Martínez et al., 2006](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B86)). EVs may also transfer receptors such as B1-receptor to allow stimulation of an inflammatory response by agonists ([Kahn et al., 2017](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B64)). In tumorigenesis, EVs may transfer oncogenic proteins as cargo, as depicted by the transmission of EGFRvIII receptor from glioma cells to induce oncogenic activity in other cells ([Al-Nedawi et al., 2008](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B2)). The intercellular communication functions of EVs are also exploited by cancer cells to modify the tumor microenvironment (TME) and normal cells within it. > > Despite extensive research on EwS genetics, the EwS cell of origin is still unknown. This is mainly due to difficulties in identifying an appropriate model to study the disease. Indeed, a permissive cellular model for the expression of EwS oncogenic chimeric proteins, mainly EWS-FLI and EWS-ERG, is mandatory given that their expression in primary cell lines result in cell death/growth arrest, whereas their expression in primitive cell lines results in incomplete differentiation ([Kovar, 2005](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B73)). > > However, multiple hypotheses do exist, such as the discovery that cell surface antigens in EwS may have a neuroectodermal lineage, suggesting the presence of a neural crest origin for EwS cells ([Lipinski et al., 1986](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B79)). This finding was supported by the high expression in EwS cells of genes found in neural and fetal tissues ([Staege et al., 2004](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B122)). Evidence also suggests a mesenchymal stem cell (MSC) origin for EwS cells as EWS-FLI1 fusion silencing causes the convergence of the EwS transcriptome toward that of MSCs ([Tirode et al., 2007](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B129)). Unfortunately, there are no known precursor lesions, preventing the study of early stage precancerous cells ([Toomey et al., 2010](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B131)). Studies using genetically modified mouse models were also unsuccessful to identify the molecular events of EwS tumorigenesis since genomic distribution of GGAA repeats is not conserved between human and mouse ([Torchia et al., 2003](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B132); [Gangwal et al., 2008](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B33); [Lin et al., 2008](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B78)). > > Novel therapeutic interventions for EwS are needed to increase this historically low survival rate seen in patients with metastatic disease ([Lawlor and Sorensen, 2015](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B75)). Therapies targeting EwS EVs could shift the current EwS treatment paradigm by abrogating intercellular communications within the EwS TME, thereby representing a highly innovative therapeutic strategy in EwS. In addition, a deep examination of EVs and their content could also be used as a tool to assess therapeutic response, making them very attractive for multiple clinical applications. > > ## Ewing Sarcoma-Derived Extracellular Vesicles > > Research  into EVs in the context of EwS is fairly recent, as EwS EVs were first  discovered in 2013 by the detection of exosome-associated markers, the  tetraspanins CD63 and CD81 shed by EwS cell lines _in vitro_ ([Miller et al., 2013](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B89)).  Analysis of EwS cell line derived exosomal RNAs revealed a wide range  of RNA species, with small RNAs accounting for a large proportion,  resulting in an RNA profile distinct from that of donor cells ([Miller et al., 2013](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B89)). > > [Miller et al. (2013)](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B89) identified the _EWS-FLI1_ transcript in EwS exosomes, in addition to 11 transcripts overexpressed in EwS tissue compared to normal tissue, such as transcripts for genes involved in signaling pathways and pluripotency maintenance ([Kawano and Kypta, 2003](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B65); [Khalfallah et al., 2009](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B68); [Richter et al., 2013](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B104)). Similarly, _EWS-FLI1_ mRNA was detected in microvesicles along with the CD63 marker secreted from three EwS cell lines _in vitro_ as well as secreted into the blood of mice harboring EwS xenografts ([Tsugita et al., 2013](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B135)). Evidence was also provided that through microvesicles, fusion transcripts could be transferred between different EwS cell lines ([Tsugita et al., 2013](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B135)). Although there is no evidence of direct involvement of EWS-FLI1 in regulating the biogenesis of exosomes, it was shown that expression of caveolin-1, a central regulator of exosome biogenesis and sorting of exosomal cargo, is directly regulated by EWS-FLI1 ([Tirado et al., 2006](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B128); [Albacete-Albacete et al., 2020](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B1)). > > Accordingly, EWS-FLI1 may play a vital role in the biogenesis of exosomes and cargo sorting. However, in a study where EWS-FLI1 was expressed in mesenchymal cells, no increase in exosome yield was reported from these cells, compared to controls ([Ventura et al., 2016](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B140)). This could mean that EWS-FLI1 may only be involved in sorting exosomal cargo but not in the regulation of exosome biogenesis. _EWS-FLI1_ transcripts were also detected in microvesicles, which suggests that the fusion protein may also assist in cargo sorting of microvesicles ([Tsugita et al., 2013](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B135)). Nonetheless, the role of EWS-FLI1 in EV biogenesis and cargo sorting needs to be thoroughly investigated. > > Similar biomolecules are found as cargo of both exosomes and microvesicles, such as _EWS-FLI1_ transcripts, making it difficult to identify which type of EV is associated to a specific event of TME reprogramming based on cargo content. This could result from the widespread use of non-specific EV purification methods, leading to co-purification of multiple types of EVs, and from the lack of confirmation of EV type by specific markers. This emphasizes the importance of using a purification method specific to the studied EV type in addition to characterizing the purified EVs using specific markers. This will allow a deeper characterization of the roles of each EV type and their associated cargo in TME reprogramming. > > Since monolayer cultures of EwS cell lines are not representative of the TME, a bioengineered tumor model was developed to better simulate the TME to study EwS EVs ([Villasante et al., 2016](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B142)). It was found that both the 3-dimensionality and extracellular matrix (ECM) composition of these bioengineered tumors influenced the size of EwS exosomes. Indeed, exosomes derived from tissue-engineered tumor had similar size distribution as those in EwS patients’ plasma and were significantly smaller than those detected in monolayer cultures, confirming the influence of the TME on EVs properties ([Villasante et al., 2016](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B142)). > > Consistent with previous studies, the mRNA of enhancer of zeste homolog 2 (EZH2) was detected in bioengineered tumor-derived exosomes, at a level similar to the one of plasma samples from EwS patients, but at a higher level than in exosomes secreted from monolayer cell lines ([Miller et al., 2013](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B89); [Villasante et al., 2016](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B142)). EZH2 is a histone methyltransferase trimethylating histone H3 at Lys 27, and a transcriptional target of EWS-FLI1, which can form the Polycomb repressive complex 2 associated with the transcriptional repression of tumor suppressors such as p14ARF and p16INK4a ([Varambally et al., 2002](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B139); [Viré et al., 2006](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B143)). > > Moreover, contents of exosomes from the bioengineered tumors were successfully transferred to human MSCs, osteoblasts, and osteoclasts, leading to an increase, no change, or a decrease in their EZH2 mRNA levels, respectively ([Villasante et al., 2016](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B142)). As osteoblasts and osteoclasts are present in bone microenvironment, their uptake of EwS EVs may potentially occur _in vivo_. Hence, although EwS-derived exosomes were shown to potentially mediate the interactions between EwS cells and the TME, the functional consequences of these interactions were only investigated recently. > > Apart from mRNAs and proteins, miRNAs have also been found in EwS EVs ([Table 2](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/table/T2/)). It was found that CD99, a surface marker of EwS cancer cells, could be released through exosomes ([Llombart-Bosch et al., 2009](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B81); [Ventura et al., 2016](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B140)). CD99 is involved in both physiological and pathological functions such as cell adhesion, migration, and differentiation ([Schenkel et al., 2002](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B110); [Cerisano et al., 2004](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B15); [Ventura et al., 2016](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B140)). When CD99 was silenced in EwS cell lines, the secreted exosomes contained increased levels of miR-34a, whose transfer repressed the Notch pathway in EwS recipient cells, inhibiting NF-κB transcriptional activity, and causing their neural differentiation similar to direct CD99 silencing ([Ventura et al., 2016](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B140)). > > A following study found that CD99-positive exosomes derived from CD99-expressing EwS donor cells led to increased proliferation, increased migration, and poor neural differentiation in EwS recipient cells compared to uptake of exosomes from CD99-silenced EwS donor cells ([De Feo et al., 2019](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B20)). CD99-silenced EwS cells exhibited reduced migratory ability, and their CD99-negative exosomes contained miR-199a-3p, thereby inhibiting growth and migration of recipient cells as seen in other cancers ([Fornari et al., 2010](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B28); [Henry et al., 2010](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B52); [De Feo et al., 2019](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484747/#B20)).