Can Soluble Immune Checkpoint Molecules on Exosomes Mediate Inflammation?

Abstract

Immune checkpoints (ICPs) are major co-signaling pathways that trigger effector functions in immune cells, with isoforms that are either membrane bound, engaging in direct cell to cell activation locally, or soluble, acting at distant sites by circulating freely or potentially via extracellular vesicles (EVs). Exosomes are small EVs secreted by a variety of cells carrying various proteins and nucleic acids. They are distributed extensively through biological fluids and have major impacts on infectious diseases, cancer, and neuroinflammation. Similarly, ICPs play key roles in a variety of disease conditions and have been extensively utilized as a prognostic tool for various cancers. Herein, we explored if the association between exosomes and ICPs could be a significant contributor of inflammation, particularly in the setting of cancer, neuroinflammation and viral infections, wherein the up regulation in both exosomal proteins and ICPs correlate with immunosuppressive effects. The detailed literature review of existing data highlights the significance and complexity of these two important pathways in mediating cancer and potentiating neuroinflammation via modulating overall immune response.

Introduction

Understanding the role immune mediators and extracellular vesicles (EVs) play in regulating the immune system is essential for constructing innovative and effective therapies against a variety of diseases. For the immune system to function optimally, it must balance immune responses against antigenic threats while maintaining tolerance of self-antigens. Immune checkpoints (ICPs) play a critical role in implementing this balance (Nagai and Azuma 2019; Zahavi and Weiner 2019). Immune checkpoints are co-stimulatory and/or co-inhibitory signaling molecules (receptor/ligands) expressed on a variety of immune cells but most notably function in regulating T-cell activation. Upon antigen presentation by the Major Histocompatibility (MHC) complex and recognition by the T cell receptor (TCR), a second signal, either stimulatory or inhibitory, is necessary to direct the response. Once triggered, the downstream effects lead to either a potent immune response by releasing proinflammatory mediators, or curbing activation to prevent indiscriminate tissue destruction (Marin-Acevedo et al. 2018). This tightly regulated response is implemented by ICP receptors and ligands that belong to either TNFR superfamily of receptors i.e., 4-1BB, B- and T- lymphocyte attenuator (BTLA), HVEM, B7-1/2 (CD80/86), OX40, GITR, CD27 (Ward-Kavanagh et al. 2016) or Immunoglobin (Ig/CD28) superfamily including CD28, programmed death-1 receptor (PD-1), cytotoxic T lymphocyte antigen-4 (CTLA-4), lymphocyte activation gene-3 (LAG-3), T-cell immunoglobulin and mucin protein-3 (TIM-3), and T-cell immunoreceptor with Ig and ITIM domains (TIGIT) (Schmitz and Krappmann 2006). Crosstalk signaling commonly occurs between these families of receptor/ligands since many can act on the same signal cascade for activation via PI3K, NFκB, MAPK or inhibition via adaptor protein kinases like, Grb-2, SHP1/2 (Schmitz and Krappmann 2006; Lorenz 2009; So and Fruman 2012). Evident roles of ICPs have long been associated with the field of cancer immunotherapy and more recently in neuroinflammation and chronic viral infection, where dysfunctional cells that exhibit immunosuppressive abilities can be reverted via inhibition of select ICPs, which are often thwarted in disease states (Buchbinder and Desai 2016).

ICPs, while often analyzed by cell surface expression, can vary in their isoforms, i.e., membrane-bound and modulating responses by direct cell to cell signaling, or in soluble form, where they can circulate and affect distant sites. Soluble and membrane-bound ICPs are transcribed from the same gene, but the mRNA is alternatively spliced, leading to two different products: membrane-bound ICP and soluble ICPs (Fig. 1) (Champion 1988; Magistrelli et al. 1999; Gu et al. 2018). The membranous isoforms are transmembrane proteins that have a cytosolic domain as well as extracellular binding domains and can vary in cell surface expression depending on cell activation or external stimulus. For example, CTLA-4 is minimally found on the cell surface and is mainly stored in organelles, but when T cells are activated, CTLA-4 gets transported to the surface of the cell where it can interact with cognate ligands (Buchbinder and Desai 2016; Gu et al. 2018; Nagai and Azuma 2019; Zahavi and Weiner 2019; Kong et al. 2020). While most ICPs and their ligands are produced via variable splicing of mRNA, some are produced by cleaving portions of the membrane bound ICPs, giving rise to soluble forms (Gu et al. 2018). These isoforms are functional and initiate similar responses on target cells. Soluble forms are shed in serum and have been associated with affecting the efficiency of the immune system and hindering treatment strategies by interacting with the binding capacity of immunotherapeutic drugs (Raza et al. 2021). sLAG-3 form was found to positively influence the immune system’s response by stimulating the antigen-presenting cells (Casati et al. 2006; Donini et al. 2018). Furthermore, sLAG-3, sTIM-3, sCTLA-4, and sPD-1 were all found to be upregulated upon the stimulation of the immune system creating a potential targeted profile for the diagnosis hepatocellular carcinoma (Dong et al. 2020; Khan et al. 2020; Odagiri et al. 2020; Raza et al. 2021).

Fig. 1.

Generation of soluble and membranous ICPs. DNA encoding an immune checkpoint receptor/ligand is transcribed into mRNA in the nucleus and is alternatively spliced, leading to soluble or membrane bound isoforms. The respective mRNAs are translated in the ER and further modified in the Golgi. The membrane bound isoforms are transported from the Golgi and fuse into the plasma membrane while the soluble isoforms can be released directly into the extracellular environment (ECM) or as cargo within or bound to an exosome. Figures were generated using BioRender software

Another contributor that influences potent immune responses are exosomes. Exosomes are membrane-bound vesicles, measuring between 40 and 150 nm in diameter, distributed extensively throughout biological fluids and released by most cells carrying diverse cargo consisting of proteins, nucleic acids and metabolites (Harding et al. 2013; Jeppesen et al. 2019; Yang et al. 2019). These vesicles have been reported to carry a narrow spectrum of molecules involved in the immune response and subsequent signal transduction. Initially, exosomes were considered to be an alternative method to eliminate waste to maintain homeostasis (Johnstone et al. 1987; Baixauli et al. 2014; Hessvik et al. 2016; Hessvik and Llorente 2018). But lately, numerous studies have highlighted their physiological and pathological functions in cell-to-cell signaling (Vacchi et al. 2020), antigen presentation (Raposo et al. 1996), promoting angiogenesis when released from cancer cells, and immune regulation (Mathivanan et al. 2010; Raposo and Stoorvogel 2013; Record et al. 2014; Robbins and Morelli 2014; Hessvik and Llorente 2018). Overall, exosomes have a significant impact on the spread of infections and immune responses.

Exosome biogenesis begins with the inward budding of the endosomal membrane, either by direct invagination of the plasma membrane or through a receptor mediated process. The endosome undergoes a maturation process where the transmembrane proteoglycan complex, syndecan-syntenin, promotes intraluminal budding of the membranes of the endosome, forming intraluminal vesicles (ILV) inside the endosome and generating a multivesicular body (MVB). Following intraluminal budding, the vesicles invaginate cytosolic proteins and RNA such as mRNA, miRNA, and non-coding RNA (ncRNA), as well as lipids and metabolites (van Niel et al. 2018; Jeppesen et al. 2019; Mathieu et al. 2019) (Fig. 2). The incorporation and concentration of these various cargos are dependent on cell type, localization signals, and the ESCRT machinery that assists with the process (Hessvik and Llorente 2018). The ESCRT machinery features four different protein complexes: ESCRT-0, -I, -II, -III, and AAA ATPase Vps4 complex (Futter et al. 2001; Alenquer and Amorim 2015). The ESCRT-0 heterodimer recognizes proteins that are ubiquitinated and recruits them to the membrane, resulting in the formation of microdomains that can sort the content of nascent exosomes (Skryabin et al. 2020). ESCRT-0 also recruits ESCRT-I and -II, which initiate invagination of membrane regions with specific cargo while the protein Alix, recruits ESRT-III, which is responsible for vesicle budding and the ultimate dissociation from the membrane (Skryabin et al. 2020). The silencing of Alix has been shown to change protein composition of exosomes (Hessvik and Llorente 2018). Biogenesis can also occur without ESCRTs as ILVs have been formed after silencing of all four complexes. Tetraspanins (such as CD63) are enriched in exosomes and are also involved in the biogenesis process (Escola et al. 1998; Stuffers et al. 2009; Hessvik and Llorente 2018) as following its knockout with CRISPR/Cas9, EV secretion was reduced (Hurwitz et al. 2016; Hessvik and Llorente 2018). In addition, the sphingomyelinase pathway that catalyzes ceramide synthesis (Trajkovic et al. 2008; Alenquer and Amorim 2015), and the phospholipase D2 and ADP ribosylation factor-6-mediated ILV budding mechanism (Ghossoub et al. 2014; Alenquer and Amorim 2015) are additional ESCRT independent methods of biogenesis. Another mechanism of biogenesis is via membrane lipid microdomains associated with modifications of endosomal membrane lipid composition, with lipid rafts mediating membrane invagination and the formation of vesicles (Skryabin et al. 2020). However, research in this process of cargo sorting is limited. Additional factors such as intracellular concentration of calcium and the pH gradient between the interior and exterior of the cell could also impact the secretion of exosomes (Savina et al. 2003; Parolini et al. 2009). Exosomal cargo are often common molecules seen among all cells but can also contain cell specific proteins based on the source of the exosome- this can include proteins only expressed in specific cell types, tumor antigens, intracellular infection and/or invading pathogen associated molecules (Mathivanan et al. 2010; D'Asti et al. 2012; Raposo and Stoorvogel 2013; Baixauli et al. 2014; Mashouri et al. 2019). Those common among all exosomes include cell adhesion molecules (CAM) and tetraspanins. (Raposo et al. 1996; Raposo and Stoorvogel 2013; van Niel et al. 2018; Mathieu et al. 2019). Additionally, the structural integrity of exosomes is critical for immune responses, as the lipid-bilayer enclosed extracellular vesicles that are exosomes must be kept intact to induce cytokine production and this response has been seen in a TLR-dependent manner in monocytes (Bretz et al. 2013). Once released, exosomes can enter recipient cells and merge into endosomes and then either mature into lysosomes or undergo transcytosis and enter adjacent cells after being released by recipient cell endosomes (Munich et al. 2012; Tian et al. 2013; Mulcahy et al. 2014; Yang et al. 2019).

Fig. 2.

Overview of exosome biogenesis. Endosomes enters the cell either by invagination or receptor mediated endocytosis. Through a sequence of intraluminal invagination and endosomal sorting, the endosome transforms into multivesicular bodies (MVB) and then matures into a late endosome. During this phase, the contents that are ultimately secreted in the exosome are added through modifications and transport complexes within the MVB, the ER and Golgi. MVBs, are then either transported to the plasma membrane and fuse with the plasma membrane or directed to lysosomes where their content is degraded (not shown). Ultimately, exosomal content is a heterogeneous mixture of cargo inside and on the surface, depending on the parent cell, including host proteins- cell specific or ubiquitous, nucleic acids, ICPs, lipids, pathogen associated molecules and other metabolites. Figures were generated using BioRender software

Immune Checkpoint Molecules and Exosomes in the Context of Various Diseases

Bioactive molecules carried by exosomes have major influences on the extracellular environment and immune system (Chen et al. 2018). As such, it comes as no surprise that exosomes have also been associated with ICPs. Investigations of exosome biology and their influence on immune responses and general pathophysiology are still being explored, but recent evidence has clearly demonstrated a significant association of exosomes with ICPs.

Blood Cancers and Solid Tumors

Inhibitory ICPs are usually expressed by malignant cells to thwart an anti-tumor immune response favoring the expansion of malignant cancer cells. In human carcinomas of the lung, ovary, and colon as well as in melanomas, there is significant upregulation of PD-L1 (Dong et al. 2002; Chen and Han 2015; Chen et al. 2018). Notably, there is also a significant presence of PD-L1 in tumor-derived EVs. Exosomal PD-L1 has the potential to alter immune surveillance and has demonstrated clinical significance as a cancer biomarker (Li et al. 2019; Theodoraki et al. 2019; Xie et al. 2019; Raimondo et al. 2020). PD-L1 was found to be present in exosomes in mice with human melanoma xenografts that lack endogenous PD-1 (Chen et al. 2018) and there has also been a positive correlation shown with pathological-clinical features in patients and disease severity (Ludwig et al. 2017; Zhou et al. 2017; Chen et al. 2018; Ricklefs et al. 2018; Theodoraki et al. 2018). Studies also have shown that immunosuppressive pathways are driven by intrinsic means rather than by cancer cells, indicating that though cancer immunotherapy targets negative regulatory ICPs, other mechanisms contribute to drug efficacy (Spranger et al. 2013). CD8+ T-cell infiltrated tumors have shown high amounts of CD4+, CD25+, and FoxP3+cells expressing PD-L1, suggesting that it is immunosuppression that prevents the optimal functionality of T cells (Spranger et al. 2013). In metastatic melanoma, there is a high release of exosomes carrying PD-L1 on their surface and IFN-γ presence, which upregulates PD-L1, resulting in tumor growth and immune suppression of CD8 T cells (Chen et al. 2018). In cases of IFN-γ treatment for melanoma, exosomal PD-L1 secretion and binding to PD-1 are significantly increased. Furthermore, studies have shown that PD-L1 from exosomes have the same membrane topology as on the surface of cells and in vitro analysis of exosomes expressing exogenous PD-L1 from melanoma cells that do not express endogenous immunosuppressive proteins results in inhibition of proliferation, production of cytokines, and decreased cytotoxicity of CD8 T cells (Chen et al. 2018). Besides PD-L1, other extracellular vesicle proteins can also have immunosuppressive effects through different mechanisms, including Fas ligand (Chen et al. 2018).

Recent therapies utilize ICPs to either stimulate or block these pathways, improving the body’s immune activity, both cellular and humoral, against tumors. Improvements in treatments and patient outcomes have seen progress with such treatments (Gonzalez et al. 2018; Marin-Acevedo et al. 2018). In many cancers, the expression of soluble ICPs has been analyzed in addition to those found on the membrane, as these soluble checkpoints can diffuse in the serum and their plasma levels can affect the development and prognosis of cancer (Gu et al. 2018). This is important as therapies that aim to counteract immune evasion mechanisms of tumors, often seen in metastatic melanoma and extending to other tumor settings such as lung, kidney, and bladder cancers, can be effected by these circulating ICPs. For viral infection associated cancers, ICIs (immune checkpoint inhibitors) have been shown to be safe and effective in patients with hepatitis, though there is risk of secondary reactive infection (Pu et al. 2020). In essence, therapeutic strategies have demonstrated efficacy, though not all beneficial depending on the prognosis. Other problems such as resistance, autoantibodies, and disease relapse have occurred after initial clinical benefits, as checkpoint inhibitors may inhibit overall immune function, leading to undesirable clinical results (Marin-Acevedo et al. 2018). Nonetheless, ICI have shown significant promise in cancer therapy.

The most widely researched checkpoints are the inhibitory pathways of CTLA-4, PD-1, and their corresponding ligands (Seidel et al. 2018). CTLA-4 works primarily in the lymph nodes, regulating T cell proliferation early in the immune response and PD-1 works primarily in the peripheral tissues, suppressing T cells at later stages in the immune response (Buchbinder and Desai 2016). PD-L1, which is expressed by cancer cells, works to inhibit the functional activity of T cells. The presence of CD8+ T cells in the tumor environment is correlated with favorable anti-tumor responses. However, immunosuppression caused by the complex tumor cell network impacts the activity of CD8+T cells through inhibition of its function or through induction of programmed cell death. sPD-1, sPD-L1, sPD-L2, sCTLA-4, sCD80, sCD86, sB7-H3, and sCD137 each have been shown to have increasing relevance in the prognosis of cancer (Gu et al. 2018). Furthermore, in the case of soluble ICPs, these checkpoints can diffuse in serum and alter plasma levels and negatively or positively impact immune regulation, as well as the development and prognosis of cancers (Gu et al. 2018). In order to keep T cells functioning as anti-tumor molecules, CTLA-4 and PD-1/PD-L1 blocking agents have been developed (Quezada and Peggs 2013). The clinical application of ICIs is based on monoclonal antibodies, which block the checkpoint CTLA-4 or the PD-1/PD-L1 axis, aiming to counteract immune evasion mechanisms that tumors often utilize, (Table 1) (Heinzerling et al. 2019). The clinical significance of these antibodies was initially restricted to metastatic melanoma but is currently extending to other tumor settings, including lung, kidney, and bladder cancers (Donini et al. 2018). The rationale behind these treatments is to release an antitumor adaptive immune response that is normally inhibited by expedients that are tumor adaptive-resistant. CTLA-4 and PD-1/PD-L1 inhibition has led to survival improvement in patients with various malignancies, thus revolutionizing oncological care. ICIs along with chemotherapy, have also been used as an initial treatment for lung cancer. However, reliable results still remain elusive pending ongoing data collection and analysis (Shen et al. 2018).

Table 1.

Current status of immune checkpoint therapy in cancer

ICI Drug	Target	Indications	Status
Atezolizumab	PD-L1; IgG1 human antibody	Non-small cell lung cancer, urothelial carcinoma, triple-negative breast cancer, small cell lung cancer	Approved in the EU
Avelumab	PD-L1; IgG1 human antibody	Merkel cell carcinoma, urothelial carcinoma, renal cell carcinoma	Approved in the EU
Cemiplimab	PD-1; Human monoclonal antibody	Squamous-cell carcinoma	Approved in the USA
Durvalumab	PD-L1; IgG1 human antibody	Urothelial carcinoma, non-small cell lung cancer, bladder cancer	Approved in the USA
Ipilimumab	CTLA-4; IgG1 human antibody	Unrestrictable metastatic melanoma, renal cell carcinoma, colorectal cancer	Approved in the EU
Nivolumab	PD-1; IgG4 human antibody	Unresectable or metastatic melanoma, non-small cell lung cancer, renal cell carcinoma, urothelial carcinoma, classic Hodgkin’s lymphoma, squamous-cell carcinoma	Approved in the EU
Pembrolizumab	PD-1; IgG4 human antibody	Advanced or unresectable melanoma, non-small cell lung cancer, urothelial carcinoma, Hodgkin’s lymphoma, endometrial carcinoma, Merkel cell carcinoma, gastric cancer, cervical cancer	Approved in the EU
Relatlimab	LAG-3; humanized IgG4	Melanoma, glioblastoma, pancreatic carcinoma, lymphoma	Not yet approved
Tremelimumab	CTLA-4; IgG2 human antibody	Melanoma, pleural mesothelioma	Not yet approved
Urelumab	CD137/4-1BB; humanized IgG4, human IgG2	Glioblastoma, lymphoma	Not yet approved

It is also important to note that not all patients benefit from these therapeutic strategies. Several problems still exist for a considerable number of patients despite initial clinical benefits such as resistance and disease relapse (Marin-Acevedo et al. 2018). Numerous side effects can emerge from these therapies as checkpoint inhibitors function to globally inhibit immune function in general, leading to undesirable clinical effects. Most common side-effects associated with these therapies include diarrhea, nausea, pain, and weight gain. Further research is still needed to improve the efficiency of these potential therapeutic strategies. PD-L1 expression is also predicted to be dynamic over time as it displays intra-tumoral heterogeneity, thus complicating ongoing efforts to establish a predictive biomarker of response to the blockade (Smola et al. 2017). A detailed investigation into the complex regulatory interactions amongst effectors of the anticancer immune response is needed to comprehend problems of resistance and relapse. Research is ongoing to investigate the potential of new immune-modulatory molecules to expand upon current clinical applications. There are numerous inhibitory and stimulatory checkpoints that play a significant role, paving the way for new strategies (Table 1). Pre-clinical studies have recognized novel ICPs that play a role in the regulation of T cell response. These checkpoints, however, also affect other important modulators of the innate immune response. Their potential therapeutic use is providing significant preclinical outcomes and is currently being investigated in clinical trials. Drugs blocking LAG-3, TIM-3, TIGIT, or VISTA are being evaluated as well, as these are novel inhibitory pathways (ElTanbouly et al. 2019; Qin et al. 2019; Acharya et al. 2020; Chauvin and Zarour 2020; Shan et al. 2020).

There are a few notable biological components in the tumor microenvironment, including mesenchymal cells, fibroblasts, endothelial cells, hematopoietic cells, and an extracellular matrix, which provide both essential physical and biomechanical support (Lu et al. 2012; Pattabiraman and Weinberg 2014; Ruivo et al. 2017). Exosomes have been found to play roles in both triggering a cancer immune response as well as escaping immune surveillance, (Table 2). Initial evidence for exosomes in immune responses was shown with exosomes obtained from human EBV-transformed B-lymphocytes, which contain MHC class I and II on their surface (Raposo et al. 1996; Ruivo et al. 2017) displaying antigens to CD4+ and CD8+ T-cells and triggering the initial antigen-specific cascade for an adequate immune response (Holling et al. 2004; Ruivo et al. 2017). In the context of antitumor immune responses and anticancer vaccination, human DC-derived exosomes have been investigated due to their association and expression of MHC-I and -II molecules (Zitvogel et al. 1998; Ruivo et al. 2017). Research has shown that isolated exosomes expressing tumor antigens and DC-derived exosomes significantly reduce tumor burden in astrocytoma and breast carcinoma, specifically in mouse models (Zitvogel et al. 1998; Ruivo et al. 2017). Both DC-derived and tumor derived exosomes have shown considerable anti-tumor effects in experimental studies and much of their potential remains to be explored (Li et al. 2006). Another method that MHC positive tumor-derived exosomes use to activate a T-cell antigen-specific immune response is through cross-presentation of tumor antigens. However, understanding the impact of exosomes without MHC in modulating immune responses are on going (Ruivo et al. 2017). It is important to note that exosomes can elicit unwanted immune responses such as immune tolerance or immune evasion, but further investigations about the physiological functions of exosomes from various sources are necessary. Tumor-derived exosomes (TEX) also demonstrate an ability to escape immune surveillance and promote tumor development, as well as propagate dissemination of viral pathogens. TEX can affect proliferation, apoptosis, cytokine production, and activity of both innate and adaptive immune cells. For instance, TEX have been shown to promote the differentiation of monocytes into myeloid derived suppressor cells, which inhibit DC maturation (Valenti et al. 2006, 2007). Exosomes that are derived from prostate cancer samples were seen to decrease T-cell proliferation in a dose dependent manner based on expression of the Fas-ligand (Abusamra et al. 2005; Ruivo et al. 2017). Furthermore, there is a demonstrated link in tumor-derived EVs, specifically exosomes, in regulating ICPs, tumor formation and progression (Raimondo et al. 2020). Extracellular vesicles can have a plethora of effects on the formation and progression of tumors, including angiogenesis (Lang et al. 2017; Maji et al. 2017; Chiba et al. 2018), invasion (Li et al. 2018a; Emmanouilidi et al. 2019), metastasis (Costa-Silva et al. 2015; Chen et al. 2017a; Wortzel et al. 2019), and resistance to therapies (Boelens et al. 2014; Au Yeung et al. 2016; Raimondo et al. 2020). Understanding EV mediated immune regulation and cross talk is important in determining immunosuppression and pro-tumorigenic effects.

Table 2.

Role of exosome in various stages of cancers

Role	Mechanisms and specifics
Cancer detection	Cancer exosomes play a role in diagnosis of malignancies such as prostate, pancreas, breast, ovarian cancers, glioblastoma, and melanoma via elevation of concentration in systemic circulation (Skog et al. 2008; Li et al. 2009; Logozzi et al. 2009; Nilsson et al. 2009; Corcoran et al. 2011; Melo et al. 2015; Kalluri 2016)
Biomarkers—mRNA	Nucleic acids in exosomes including mutant mRNA such as EGFRvIII variant can be biomarkers for glioblastoma (Skog et al. 2008). Enriched and specific miRs for diagnosis and monitor progression of cancer (Taylor and Gercel-Taylor 2008; Rabinowits et al. 2009; Tanaka et al. 2013; Long et al. 2015; Pfeffer et al. 2015; Kalluri 2016; Li et al. 2016)
Biomarkers—DNA	Information about cancer specific mutations as exosomes in serum of pancreatic cancer patients contained the entire genomic double-stranded DNA after whole genome sequencing (Kalluri 2016)
Tumorigenesis	Intercellular transfer of oncoproteins by exosomes facilitate tumorigenesis (Al-Nedawi et al. 2008; Demory Beckler et al. 2013). Macrophage migration inhibitory factory (MIF) is elevated in circulating exosomes of PDAC patients that have disease progress after diagnosis compared to controls or PDAC patients with no disease after 5 years (Kalluri 2016)
Tumor immunity	Tumor nano environment has EVs of different sizes and apoptotic bodies, with exosomes from cancer cells having protumorigenic or anti-tumorigenic environments by changing the response of supporting tissues. Exosomes from DCs can activate T and B cells, while exosomes from cancer cells can be a source of tumor antigens presented to activated T cells (Zitvogel et al. 1998; Wolfers et al. 2001; Andre et al. 2002a, b; Schartz et al. 2002; Hwang et al. 2003). Exosomes may aid in immune evasion by impairing DC maturation (Yu et al. 2007; Kalluri 2016)
Tumor angiogenesis	When cancer-derived exosomes are taken up by ECs, in response to enhanced exosome production from intra-tumoral hypoxia, angiogenesis is stimulated partly through stimulation of proangiogenic secretion of ECs (Skog et al. 2008; Hong et al. 2009; Park et al. 2010; Umezu et al. 2013) Tumor-derived exosomes can impact tumor growth, metastasis, and structural integrity of ECs (Kalluri 2016). Tumor-derived exosomes can impact tumor growth, metastasis, and structural integrity of ECs (Kalluri 2016)
Fibroblast activation	Stimulation by cancer cell-derived exosomes can help cancer-associated fibroblasts acquire pro-tumorigenic traits. Cancer-cell derived exosomes that are originated from prostate cancer cells induce activation of myofibroblasts (Webber et al. 2015; Kalluri 2016)
Tumor growth regulation	Cancer exosomes are involved in conversion of benign epithelial cells into malignant cells, the remodeling of the extracellular matrix (ECM) in causing the promotion of cancer cell invasion, modulation metastatic sites resulting in enhancing metastases in diseases, and inducing harmful phenotypes into recipient cells (Kalluri 2016)

Though further research is necessary, exosomes have also been explored in anticancer immunization or vaccinations for cancers including melanoma, lung, and colorectal by isolating them from ex vivo expanded cells due to their potential to present antigens (Ruivo et al. 2017). This potential is based on the capacity of free EVs to stimulate T cells, which can be enhanced if these vesicles are derived from DCs, as well as their ability to penetrate privileged tissue space like the central nervous system (CNS) (Robbins and Morelli 2014).

Neuroinflammation

Neuroinflammation has a pivotal role in the development and progression of many different neurodegenerative diseases like Multiple Sclerosis (MS), HIV associated neurocognitive disorder (HAND), HTLV-1 associated myelopathy/tropical spastic paraparesis (HAM/TSP), Alzheimer’s disease (AD), and Parkinson’s disease (PD) (Yang et al. 2008; Wu et al. 2009; Williams et al. 2014; Anderson et al. 2016; Bottcher and Priller 2016; Fromentin 2016; Vera et al. 2016; Li and Barres 2018; Sperk et al. 2018; Wykes and Lewin 2018; Katuri et al. 2019). It is now well established that immune cells’ presence and activity within the CNS contribute to neuroinflammation and degeneration. Consequently, ICP association with neuroinflammation is seen in a variety of pathological conditions.

MS is a neurological disease that has considerable neuroinflammation mediated by the infiltration of autoreactive immune cells into the CNS. Although the etiology of MS remains elusive, MS-associated neuroinflammation often leads to the destruction of myelin in the CNS, causing deficits seen in patient populations (Bottcher and Priller 2016). Interestingly CTLA-4, which is important in downregulating T-cell activation, was shown in one study to be dysregulated and linked to overall susceptibility and severity of the clinical course of MS. This study was also the first to show that both membrane-bound and soluble CTLA-4 levels were affected, which may have led to inadequate down regulation of T-cell response (Karabon et al. 2009; Liu and Zhang 2014; Gerdes et al. 2016). TIM-3, another co-inhibitory immune molecule, has decreased expression on T cells in peripheral blood and cerebrospinal fluid in MS patients. Interestingly, TIM-3 expression is seen on T cells of MS patients with stable disease and patients with reduced MS relapses (Anderson et al. 2016). Intriguingly, it has been investigated whether glatiramer acetate and IFN-beta could be successful treatments of MS because they modulate TIM-3. Unfortunately, in clinical trials MS patients that glatiramer and IFN-beta worked for, TIM-3 levels were comparable to healthy controls and were ineffective (Yang et al. 2008). As such, understanding how to manipulate TIM-3 for MS treatment strategies remains to be explored. Like TIM-3, reduced expression of TIGIT also correlates with the diagnosis with of autoimmune disorders (Dixon et al. 2018). Similarly, sPD-Ll was found to interact with membrane-bound PD-1 and act as an autoimmune antibody, blocking the function of membrane-bound form (Wu et al. 2009). Interestingly, this study also found a new pathogenic mechanism in which overexpressed PD-1 attempts to correct itself using the over-production of sPD-1 (Wu et al. 2009).

As mentioned previously, two immune checkpoint receptors, CTLA-4 and PD1, and its associated ligands PD-L1 and PD-L2 are important in regulating anti-tumor immune responses, but have impacts in neuroinflammatory autoimmune diseases (Lu et al. 2019). CTLA-4 has a promoter that is polymorphic and related to susceptibility to systemic lupus erythematous (SLE). (Taha Khalaf 2011). In addition, other aspects of CTLA-4 linked to SLE include modulation of humoral responses through affecting follicular T helper cells, follicular regulatory T cells, and regulatory T cells (Finck et al. 1994; Daikh and Wofsy 2001; Lu et al. 2019). Similarly, PD1 has a notable presence in SLE and recent attention has been garnered to these two ICP receptors due to the fact that studies have demonstrated that blocking them is related to more immune-related adverse events (irAEs) which resemble SLE (van der Vlist et al. 2016; Lu et al. 2019; Ramos-Casals et al. 2020). The reasoning behind blocking CTLA-4 and PD-1 with anti-CTLA-4 antibodies and anti-PD-1 antibodies for different cancers is due to demonstrated clinical success in treatment, despite side effects resembling rheumatic diseases such as SLE (Lu et al. 2019).

With respect to virus associated CNS complications, PD-1 has been consistently reported to be upregulated in chronic HIV infections (Sperk et al. 2018; Wykes and Lewin 2018). HIV particles in the CNS, along with monocytes, lead to the neuroinflammation seen in the CNS. Interestingly these proinflammatory cells and proteins remain in the CNS even when peripheral HIV viral load is low (Williams et al. 2021). When patients with HIV are treated effectively, the viral load in the CNS remains high and correlates with cognitive impairments, although this could be due to the poor penetration of anti-retroviral drugs into the CNS (Vera et al. 2016). In HAM/TSP patients, inflammation of the spinal cord drives symptomology. In these patients levels of PD-1 and other inhibitory immune checkpoints like TIGIT, TIM-3, and LAG-3 are elevated and correlate with the degree of inflammation (Clements et al. 2021).

Exosomes also play significant roles in the context of neuroinflammation. All neural cells, whether it be neurons, astrocytes, oligodendrocytes, or microglia, release EVs that include exosomes and ectosomes in both pathological and normal conditions (Gupta and Pulliam 2014; Brites and Fernandes 2015). Origins of exosomes being used to aid delivery of therapeutic agents in treating inflammatory conditions have been discovered and advancing over the past decade (Sun et al. 2010; Yang et al. 2017). Exosomes can mediate the delivery of genes without resulting in adverse immune reactions, while other gene therapy methods such as viral vectors and lipid nanoparticles often result in activation of the host immune system. (Marcus and Leonard 2013). Studies found that exosomes were able to carry and transmit curcumin, the primary bioactive substance that is in turmeric, which has anti-inflammatory properties (Kalani et al. 2016; Wang et al. 2019a), lead to increased stability and bioavailability of curcumin and a higher survival rate in patients with lipopolysaccharide (LPS)-induced septicemia. Evidence supports curcumin and other anti-inflammatory agents such as JSI124 can reach the brain via exosomes, alleviating LPS induced brain inflammation in mice (Zhuang et al. 2011; Yang et al. 2017).

For the treatment of neurological diseases, predominant research strategies utilize exosomes derived from mesenchymal stem cells (MSCs). MSCs have self-renewing capacity and are important in neural tissue repair (Pittenger et al. 1999; Donega et al. 2014; Takeda and Xu 2015; Bagher et al. 2016; Yang et al. 2017). The MSC environment significantly impacts the content of the exosomes that will be secreted, leading to different results in the tissues they act upon. MSCs-derived exosomes have been shown to reduce neuroinflammation, increase neurogenesis and angiogenesis and improve functional recovery after a traumatic brain injury (TBI) in animal models (Zhang et al. 2015, 2017; Kim et al. 2016; Yang et al. 2017). In the context of TBI, neuroinflammation plays a critical role in its pathogenesis during the primary and secondary stages (Lozano et al. 2015; Yang et al. 2017). Primary injury results in irreversible brain damage, and secondary injury has major impacts on prognosis and the recovery process of the injury. Both positive and negative effects are seen with neuroinflammation, as an initial activation of microglia followed by astrocyte activation, recruitment of leukocytes, and formation and release of inflammatory mediators promote healing. However, as this persists or becomes hyperactive, indiscriminate tissue damage can propel CNS pathology (Morganti-Kossmann et al. 2001; Chiu et al. 2016; Yang et al. 2017). Benefits of MSCs-derived exosomes include low immunogenicity, and very small size permitting diffusion and crossing the blood–brain barrier (BBB) to target a variety of cells, as well as capabilities of transporting biological molecules (Yang et al. 2017). The activation of astrocytes and microglia are important in initiating and progressing neuroinflammation after TBI, and MSCs-derived exosomes have demonstrated anti-inflammatory responses, which can suppress inflammation associated damage (Zhang et al. 2013; Yang et al. 2017). However, the precise mechanism in treating neuroinflammation with MSCs-derived exosomes remains unclear. MSCs-derived exosomes are being considered, as they may increase induction of anti-inflammatory cytokines, decrease in pro-inflammatory cytokines (Zhang et al. 2015; Yang et al. 2017), and inhibit the activation of macrophages and hypoxic inflammation (Lee et al. 2012; Phinney et al. 2015; Yang et al. 2017). Neuroprotective effects have also been shown through the growth of neurites in rat primary neurons following occlusion of the middle cerebral artery (Xin et al. 2012; Marcus and Leonard 2013).

In Alzheimer’s, exosomes were implicated in amyloid-beta (Aβ) spreading between cells, neurotoxicity, overexpression of tau, and neurofibrillary lesions (Saman et al. 2014; Brites 2015; Brites and Fernandes 2015). Studies have demonstrated exosomes as biomarkers for Parkinsons through an increase in brain-derived exosomes and their target proteins, such as alpha-synuclein or DJ-1 in neuronal-derived exosomes in plasma (Zhao et al. 2018; Ohmichi et al. 2019; Vacchi et al. 2020). This has also improved diagnostic accuracy for PD (Jiang et al. 2020). Extracellular vesicle surface markers CD25 and CD146 were also identified to have a moderate correlation with cognitive impairment in PD (Vacchi et al. 2020). This supports the notion that there is a link between inflammation and cognitive decline, as CD25 supports immune cell activation and CD146 is important in serving as a therapeutic target for cerebrovascular disorders (Elgueta et al. 2009; Chen et al. 2017b; Vacchi et al. 2020). Neuroinflammation often also occurs with systemic inflammatory conditions, as seen in rheumatoid arthritis, sepsis, type 2 diabetes mellitus, and obesity (Semmler et al. 2008; Fuggle et al. 2014; Miller and Spencer 2014; Srodulski et al. 2014; Li and Barres 2018). Because of this, there is likely a connection between the peripheral immune system and the neuroimmune system. As exosomes are significant mediators in cell-to-cell communication with neighboring or distant cells, studies have hypothesized that blood-borne exosomes can be involved between peripheral and CNS immune systems. For instance, when serum-derived exosomes are transfused from LPS-challenged mice into control mice, there is an upregulation in microglial and astrocytic activation and additional increases in the expression of inflammatory cytokines in the brain (Miller and Spencer 2014; Li et al. 2018b). Considering this, exosomes can be a factor in mediating neuroinflammation in central and peripheral nervous system.

Exosomes also have a prominent role in SLE. In this autoimmune disease, damage to multiple organs occur as a result of abnormal activation of autoreactive T cells, and hyperactive B cells producing autoantibodies that form immune complex deposits, tissue inflammation, and complement activation (Rahman and Isenberg 2008; Lee et al. 2016; Xu et al. 2019). SLE is a systemic autoimmune disease, where the immune response is attacking various organs and tissues at the same time (Xu et al. 2019). Microparticles, such as miRNAs that can induce inflammatory responses have shown a higher concentration of immunoglobulins and complement compared to that of healthy controls (Fabbri et al. 2012; Nielsen et al. 2012; Østergaard et al. 2013; Lee et al. 2016). Therefore, there is evidence that suggests circulating exosomes are immunologically active in SLE patients, with a direct correlation on exosome levels and activity of the disease. A possible mechanism behind this correlation is that apoptotic cells in inflamed tissues could be releasing more exosomes into the blood, as in SLE there is an issue with clearing out cellular debris (Cohen and Caricchio 2004; Kuhn et al. 2014; Lee et al. 2016). Although evidence remains limited on investigating exosomes from MSCs in SLE animal models or in human patients (Xu et al. 2019), a study that isolated exosomes from healthy controls and patients with SLE showed supportive results (Lee et al. 2016). Serum levels of exosomes were shown to be significantly higher in SLE patients, as well as CD81 and CD63 levels—proteins which often correspond to exosomes. In theory, removing circulating exosomes in SLE can be a novel therapeutic approach.

Viral Infections

Many infectious diseases negatively regulate immune cell functions via immune checkpoints, and it has been increasingly evident that chronic infections from pathogens exploit immune checkpoints to escape immune surveillance. In chronic viral infections, the phenomenon of T-cell exhaustion has been widely recognized and is correlated to elevated ICP expression. HIV-specific CD8 T cells upregulate PD-1 expression as their effector functions decline and in-vitro blockades of PD-1 have been shown to restore some function (D’Souza et al. 2007). In HTLV-1 infections, patients with HTLV-1 associated myelopathy have been seen to have elevated co-expression of multiple inhibitory ICPs such as TIM-3, TIGIT, PD-1 and LAG3 on both CD4+ and CD8+ T cells, and blockade strategies with monoclonal antibodies for one or more of these ICPs restored some function when evaluating TNFa and IL-2 expression after treatment (Clements et al. 2021). Since the immune system is responsible for responding to invasive pathogens, it is not surprising that PD-1 has been shown to be highly upregulated in chronic viral infections (Sperk et al. 2018; Wykes and Lewin 2018). As a major regulatory factor in T-cell exhaustion seen in chronic HIV infections (Sperk et al. 2018), PD-1 is also shown to be a significant contributor to the latency of HIV infections. This is further corroborated as administration of anti-PD-1 significantly decreased detectable plasma HIV RNA (Evans et al. 2018). While PD-1 is a significant regulator, it is not the only ICP that is elevated in chronic infections associated with T-cell exhaustion as CTLA-4, LAG-3, TIGIT, and TIM-3 have also been liked to viral infections with distinct expression patterns, particularly in HIV and HTLV-1 (Kaufmann and Walker 2009; Jin et al. 2010; Johnston et al. 2014; Chew 2016; Liu et al. 2016; Cao et al. 2018; Gu et al. 2018; Dong et al. 2019; Graydon et al. 2019; Scharf et al. 2020; Clements et al. 2021). T-cell exhaustion is described as T cells lacking effector functions or proliferation accompanied by accumulated ICP expression on its surface. TIM-3 being a notable ICP found on exhausted T cells (de Armas et al. 2019) worth mentioning that sTIM-3 and sCD40 are elevated in patients exposed to though uninfected with HIV compared to those unexposed and uninfected with HIV (Li et al. 2020). Notably, dysregulation of sTIM-3 and sCD40 is linked to the impaired cell-mediated immunity seen in this patient population (Li et al. 2020). PD-1, TIGIT, and LAG-3 are also largely expressed on cells with HIV proviral DNA and are highly predictive for cells containing viral RNA (Fromentin et al. 2016).

The discovery of ICP inhibitors has played an important role in the treatment of virus-associated cancers, including human papillomavirus (HPV), Hepatitis B virus (HBV), Hepatitis C virus (HCV), and Epstein-Barr virus (EBV). These carcinogenic viruses have evolved strategies to evade the host immune system by increasing the expression of ICP proteins. HPV infection is an important factor in the development of head and neck squamous cell carcinoma (HNSCC). Two of the most used ICIs used in the treatment of HNSCC are pembrolizumab and nivolumab, which are PD-1 inhibitors. A clinical trial, which investigated the anti-tumor activity in patients with metastatic HNSCC, showed that the treatment of ICIs was more effective in HPV-positive HNSCC. Additionally, a systematic review that evaluated the efficacy of ICIs on HNSCC revealed that there was higher overall survival or progression-free survival in HPV-positive patients in comparison to HPV-negative patients (McCusker et al. 2020). HBV and HCV both induce carcinomas as well. One review investigating outcomes of HBV/HCV patients with advanced-stage cancer treated with PD-1 inhibitor monotherapy and anti-CTLA-4 monotherapy concluded that ICIs are considered safe and effective in infected patients with cancer. However, they also found that patients can possibly develop reactive hepatitis through uncertain mechanisms. One possible theory is that blocking CTLA-4 may impair the function of T cells to suppress hepatic virus through the activation of T regulatory cells (Treg). This is based on the fact that Treg, which suppresses effector CD4+ and CD8+T cell activation, express upregulated CTLA-4 (Pu et al. 2020). EBV is a virus that is associated with human neoplasms, one of them being non-Hodgkin lymphomas (NHL). PD-L1 expression was seen to be significantly higher in EBV-positive patients compared to EBV-negative NHL, and better therapeutic effect of ICIs in EBV-positive lymphoma is thought to be correlated with the higher expression of PD-L1 (Gao et al. 2019). EBV-induced nasopharyngeal carcinoma also expresses higher levels of PD-L1 compared to EBV-negative NPC (Gao et al. 2019). Initial observations of cancer patients with HIV being treated with ICIs are emerging as well. Phase I and II trials are exploring the effects of ICIs in patients infected with HIV that have advanced solid tumors and lymphomas (Abbar et al. 2020; Gambichler et al. 2020; Sahin et al. 2020). The use of ICIs in 73 HIV-positive patients that presented with primary tumors (non-small-cell lung cancer, melanoma, Kaposi sarcoma, or anal cancer) were given either PD-1 inhibitors, CTLA-4 inhibitors, or both, showed a response rate of 67% for Kaposi sarcoma, 30% for non-small cell lung cancer, and 27% for melanoma (Rajdev et al. 2018; González-Cao et al. 2019; Gambichler et al. 2020). However, additional trials are needed since there was a lack of toxicity studies in these patients. Nonetheless, this was unique as HIV-positive patients are normally excluded from many studies due to concerns of increased viral replication in the setting of T-cell activation (Adashek et al. 2019). Recently in pre-clinal studies, BTLA/HVEM inhibitors have been assessed in restoring some of the T-cell dysfunction associated with chronic viral infections. Similar to PD-1 and CTLA-4 inhibitors, partial recovery of exhausted cells was noted (Zhang et al. 2011; Chibueze et al. 2013; Kinosada et al. 2017).

Interestingly, virally infected cells have also been seen to secret exosomes that carry viral RNA and proteins capable of eliciting responses in uninfected recipient cells. HTLV-1 EVs were seen to induce the expression of cytokines involved in migration and promote cell to cell contact. Furthermore, different fractions of exosomes were seen to carry different viral proteins and increased overall HTLV-1 spread in a humanized mouse model (Pinto et al. 2021). Similarly, different HIV proteins have been found packaged in exosome in infected and transformed cells as well as in a closely related retrovirus, simian immunodeficiency virus (SIV), and has been reviewed elsewhere (Patters and Kumar 2018). The potential exosomes have in promoting viral pathology and dissemination has led to many exciting hypotheses concerning the pathogenesis of many viruses and their associated diseases. Considering that virally infected cells secret exosomes that can enhance viral dissemination, and dysregulated ICP expression has been associated with chronic viral infections, the potential these factors have synergistically in overall viral pathology needs to be evaluated.

Mechanisms of Immune Modulation

Immune Checkpoints

The primary signal for T-cell activation occurs when tumor antigens are presented to T cells by antigen-presenting cells through the interaction between MHC and T-cell receptor. Activation is completed when there is another costimulatory signal, this often involves B7 on APCs and CD28 on T-cells interacting with one another (Prete and Salvatori 2000). TCR/CD28 interaction leads to intracellular signaling pathway activation, which results in an increase in the production of cytokines, including interleukin-2, which further promotes T cell expansion (Granier et al. 2017). There are several co-receptors that function as negative modulators of the immune response at various molecular checkpoints. The activation of ICPs is induced by chronic exposure to the antigen and is marked by the progressive loss of proinflammatory cytokine production, the loss of cytotoxic activity, the decrease in proliferative potential, and an increase in apoptosis. Examples of proinflammatory cytokines include tumor necrosis factor-alpha (TNFα) and interferon gamma (IFNγ), IL-6, IL-2, etc. (Marhelava et al. 2019).

CTLA-4 is expressed by T cells during the time of their initial response to the antigen (at the priming phase) and is then transported to the cell surface in a proportional manner related to the antigen stimulation through exocytosis. CTLA-4 is induced on CD4+and CD8+T cells after early activation but is constitutively expressed on regulatory T-cells. CTLA-4 is a member of the immunoglobulin superfamily and is homologous to CD28. Its locus is very close to the CD28 locus, so they both have very similar protein sequences. However, it binds to the ligand B7 with a greater affinity, acting as a competitive inhibitor that prevents CD28-mediated signaling. This consequently leads to the inactivation of specific T-cells and inhibition of IL-2 production (Granier et al. 2017). PD-1 is also a member of the immunoglobulin superfamily and is more widely expressed than CTLA-4. The PD-1/PD-L1 pathway is not associated with initial T-cell activation but plays an important role in peripheral tolerance to protect from exacerbated immune responses and autoimmunity. It controls inflammatory responses in tissues that are maintained by effector T cells. Activated T cells upregulate both PD-1 and inflammatory signals in the tissue to induce PD-L1 expression. It is also expressed over a longer time frame than CTLA-4 and can be seen on activated T cells, B cells, and natural killer cells (Granier et al. 2017). The cytoplasmic tail of PD-1 has two structural motifs. After the interaction of PD-1 with its ligands, there is phosphorylation of the tyrosine residues, allowing for cytoplasmic tyrosine phosphatase recruitment. These phosphatases, including SHP2, antagonize the signal of TCR and CD28, subsequently altering the signaling of pathways of PI3K, ERK, RAS, and VAV. This results in a decrease in the activation of transcription factors, including AP-1, NFAT, and NF-kB, leading to downregulation of T-cell activity and proliferation, which protects normal tissues from destruction (Prete and Salvatori 2000). The mechanisms associated with other inhibitory receptors such as TIM-3, LAG-3, and VISTA are under investigation, although many have postulated that they follow a mechanism similar to that of the TNF superfamily receptors. LAG-3 is found on activated T cells, B cells, Natural killer cells, and plasmacytoid DCs. The ligand associated with LAG-3 is MHC II and its mechanism of action consists of competitive inhibition involving antigen and CD4+ T-cell TCR interaction. Two other ligands of LAG-3 have been recognized, LSECtin and Galactin-3, which inhibit antitumor T-cell response through cis and trans interactions with LAG-3. More recently, Fibrinogen-like protein 1 has been recognized as a major inhibitory ligand for LAG-3 (Wang et al. 2019b). VISTA is another member of the CD28 protein family and functions as a coinhibitory ligand on APCs, which suppress T-cell responses, including cytokine production and T cell proliferation (Granier et al. 2017). It has an extracellular domain that is similar to that of PD-L1. Studies have shown that VISTA acts as a receptor on naïve T cells along with functioning as a ligand on APCs with roles in regulating naïve T cell quiescence and peripheral tolerance; however, the binding partners of VISTA and mechanisms are still unknown (Lines et al. 2014; ElTanbouly et al. 2020). ICPs have a complex regulatory network, and their interactions from classical, soluble, or exosomal sources can have significant impacts on immune regulation.

Exosomes on Immune Modulation

There is extensive literature that delves into EV associated immune regulation (Li et al. 2006; Greening et al. 2015; Wang et al. 2019c; Wu et al. 2019; Zhou et al. 2020; Kugeratski and Kalluri 2021). EVs have many roles in modulating immune response depending on their cargo, particularly when containing pathogen associated molecules such as dsDNA, viral RNA, or proteins as well as tumor associated antigens. Signaling mediated by exosomes is an important factor in the inflammatory response. The secretion of exosomes by antigen-presenting cells offers therapeutic benefits through the stimulation or attenuation of the immune response (Shenoda and Ajit 2016). By delivering a variety of bioactive macromolecules, exosomes exert their biological functions in cells through a variety of pathways, in which target cells can be influenced (Zhang et al. 2019). For instance, intracellular signaling cascades can be triggered when there is direct contact between exosome membrane proteins and the recipient cell plasma membrane. Protein fragments can interact with cell membrane receptors after protein cleavage on the membrane of the exosomes. There can be fusion of the exosomes with the cell membrane leading to cargo release. Lastly, phagocytosis can cause the whole vesicle to internalize (Shenoda and Ajit 2016). Antigen presentation is a very important step that occurs in the immune response. APCs present antigens to T cells through the formation of a contact point between the cells called an immunological synapse. The peptide antigens that are presented by APC are bound to MHC class I and class II molecules. As exosomes carry functional MHC-peptide complexes to regulate antigen-specific CD8+ and CD4+ T-cell responses, they play a very important role in the presentation of these complexes, the formation of the immunological synapse, and subsequent response (Barros et al. 2018). These presentations can either be direct or indirect. During direct presentation, T cells are activated because antigen-specific T cells directly engage the MHC-peptide complexes on the exosomes. This occurs when peptide MHC class I complexes on cells directly interact with CD8+ T cells and activate them instead of through an APC (den Haan et al. 2000; Wang et al. 2001; ten Broeke et al. 2013). In indirect presentation, antigens on MHC complexes carried by exosomes are acquired by APCs. These APCs present the peptide-MHC complex to T cells after the antigens go through further processing, thus activating CD4+ T cells (Dolan et al. 2006). This phenomenon, also known as cross–dressing, is more efficient when the exosomes come from mature DCs treated with LPS (Robbins and Morelli 2014). Mature exosomes derived from DCs can activate immune response via the TNF-α pathway. Epithelial cells induced by TNF-α pathway have been shown to secrete pro-inflammatory cytokines, such as MCP-1, IL-8, TNF-α, and RANTES. Exosomes derived from DC’s also promote anti-tumor responses through the activation of immune effector cells (Obregon et al. 2009).

The assessment of pathways stimulated by tumor exosomes, which lead to immune suppression, gives insight into potential cancer therapies. Tumor-derived exosomes are part of the immunosuppressive mechanisms that cancer cells use to progress through the inhibition of immune surveillance processes. Furthermore, exosomes have apoptosis-inducing ligands that can contribute to T-cell apoptosis. These ligands include Fas ligand, TNF-related apoptosis-inducing ligand, or galectin 9 (Barros et al. 2018). Exosomes that have TGF-β1 can inhibit the activation of NK cells by disrupting IL-2 signaling. NK cells can also have a reduced ability to distinguish malignant cells because exosomes carrying NKG2D ligands diminish NKG2D expression (Lundholm et al. 2014). Through the transfer of miRNA, exosomes can also inhibit toll-like receptor 4 and inhibit MHC class II genes in DCs (Naseri et al. 2020).

Conclusions and Future Perspectives

In this review, we summarized the many roles exosomes and ICPs have in cancer and neuroinflammation, including their biogenesis, activation mechanisms, and subsequent effector functions individually and in concert. Many studies have reported the immunomodulatory effects of exosomes. Cells stimulated with exosomes isolated from infected, or cancer derived cells can induce changes in gene expression, promote cell-to-cell interactions, promote cell adhesion, trigger induction of interferon and innate immune responses inflammation, and act as signaling molecules between cancer cells and surrounding cells within the tumor microenvironment. ICP expression is known to be hijacked by many infectious pathogens and cancers to evade immune surveillance, and differential expression levels of soluble ICP is often associated with poor prognosis. The overexpression of immunomodulatory ligands such as PD-1, CTLA-4, LAG-3, BTLA, VISTA, TIGIT, and TIM-3 have been shown to induce T-cell exhaustion, making necessary the development and continued improvement of ICP-blocking therapeutic strategies. Though anti-PD-1 and anti-CTLA-4 immunotherapies exist as FDA-approved solid tumor therapies, they are limited by the varying responsiveness of a diverse roster of cancers, as well as the development of resistance to treatment in a variety of mechanisms (Fares et al. 2019; Tsuyoshi Fujita et al. 2020). One such mechanism of resistance to ICP inhibition is mediated by the formation of anti-drug antibodies (Enrico et al. 2020). While the evolution of monoclonal antibodies from murine to chimeric, humanized, and finally fully human antibodies has reduced immunogenicity, there is still debate in comparing fully human antibodies with humanized antibodies in terms of immunogenicity in an oncologic, as opposed to inflammatory clinical setting (Enrico et al. 2020). Studying immunogenicity of ICP inhibitors has proven difficult due to heterogeneity between inhibitors, though, as an obstacle to fully effective treatment, further research is needed on this topic (Enrico et al. 2020). Furthermore, dermatological, and gastrointestinal toxicities may occur in individuals undergoing checkpoint inhibition therapy, along with liver and thyroid toxicity due to a loss of peripheral tolerance. While not life-threatening in most cases, abating the side-effect profile of checkpoint inhibition is a key target for further research, along with circumventing resistance (Tsuyoshi Fujita et al. 2020). Exosomes containing infection derived peptides and RNA as well as soluble ICPs have been seen clinically in cerebral spinal fluid from individuals with retroviral infections and in tumor microenvironments. Both ICPs and EV’s have significant impacts in cancer, infection, and neuroinflammation independently, however recent evidence suggests a synergistic role in driving immune responses as ICPs can be incorporated into exosomes. Given that the preservation of immunomodulatory effects exists for exosomal ICPs, vital questions remain on the effects of ICP blockade on exosomal ICPs. Another recent review also highlights the potential EV mediated communication shapes innate and adaptive immune response (Pelissier Vatter et al. 2021). While current treatments do not specifically target exosomes, it is necessary to elucidate the effects of presently available ICP-based therapies on exosomal ICPs, and whether specific blockade of exosomes may have any differential effects. Exosomal ICPs represent a major conceptual and mechanistic advancement in the understanding of immune responses as well as dysfunctions associated with chronic conditions. The complexity of the immune microenvironment prompts comprehensive evaluation to understand clinical implications and therapeutic approaches when utilizing ICP immunotherapies and exosomal delivery methods.

Abbreviations

ICP

Immune checkpoints

EV

Extracellular vesicles

MHC

Major histocompatibility complex

TCR

T cell receptor

CTLA-4

Cytotoxic T lymphocyte antigen-4

PD-1

Programmed death-1 receptor

PD-L1

Programmed death-1 ligand

LAG-3

Lymphocyte activation gene

TIM-3

T-cell immunoglobulin and mucin protein-3

TIGIT

Cell immunoreceptor with Ig and ITIM domains

ILV

Intra-luminal vesicles

MVB

Multivesicular bodies

ncRNA

Non-coding RNA

CAM

Cell adhesion molecules

APC

Antigen presenting cells

IFN-γ

Interferon gamma

ICI

Immune checkpoint Inhibitor

HIV

Human immunodeficiency virus

HPV

Human papillomavirus

HTLV-1

Human T cell leukemia Virus

HBV

Hepatitis B virus

EBV

Epstein-Barr virus

HNSCC

Head and neck squamous cell carcinoma

NHL

Non-Hodgkin lymphomas

DC

Dendritic cell

TEX

Tumor derived exosomes

MS

Multiple sclerosis

MSC

Mesenchymal stem cells

TBI

Traumatic brain injury

TNFα

Tumor necrosis factor-alpha