3.3.1. Cancer-Derived Exosomes as a Key Regulator of the Epithelial–Mesenchymal Transition (EMT)
Undoubtedly, metastasis is the most dramatic consequence of cancer, responsible for about 90% of cancer deaths globally
[51].
Metastasis is a multistep process, which involves local invasion, intravasation, transport, extravasation, and colonization
[52]. These steps require a series of genetic, biochemical, and morphological deregulations that are present in an evolutionarily conserved developmental program known as the epithelial–mesenchymal transition (EMT)
[40][53][54][55].
The EMT is a natural process of transdifferentiation of epithelial cells to mesenchymal cells that is crucial for embryogenesis
[56][57][58] and re-epithelization in tissue repair
[59]. During embryogenesis, the EMT (EMT type I) gives rise to mesoderm (responsible for the formation of muscle, bone, and connective tissues) during gastrulation and neural crest delamination (which results in glial cell, adrenal gland, and epithelial pigmented cell formation)
[54][60]. In adult life, the EMT plays a key role in tissue re-epithelization during wound healing (EMT type II)
[59][61][62] but, when inappropriately active, such as occurs in carcinogenesis (EMT type III), the EMT causes important disturbances in epithelial tissue homeostasis and integrity, leading to cancer cell spread and metastasis
[53][63].
The EMT (type III) is a consequence of cancer progression away from the cancer cells from the stroma, which is responsible for providing nutrients and oxygen support to the cells, creating a hypoxic environment. In addition, the partial reduction in the oxygen pressure leads to the activation of hypoxia-inducible factor 1 alpha (HIF-1α) in both cancer cells and cancer-associated fibroblasts (CAFs)
[64][65][66].
HIF-1α nuclear translocation promotes the upregulation and stabilization of Snail and Twist, resulting in cadherin switching, which is characterized by the downregulation of E-cadherin (leading to a loss of intercellular adhesion and consequent activation of the Wnt/β-catenin pathway) and N-cadherin upregulation in cancer cells
[67][68][69]. Combined with the F-actin reorganization of invadopodia sites, these actions create sites of transient adhesion that confer cell motility, facilitating the dissemination of cancer cells
[53][70].
HIF-1α also acts as a key regulator of metabolic plasticity, promoting genetic and metabolic deregulations
[54][71][72]. These deregulations drive the oxidative metabolism to glycolytic metabolism. This process is crucial to guaranteeing the energy supply (ATP) in hypoxic conditions
[54]. In addition, glycolytic metabolism increases lactate production, which is generated as a byproduct of glycolysis.
L-Lactate is an important oncometabolite produced by the glycolytic cells within the TME, promoting a metabolic symbiosis between cancer cells and cancer-associated fibroblasts (CAFs)
[73]. However, due to its high toxicity, L-lactate is transported out of the cytoplasm of CAFs to the extracellular compartment by a monocarboxylate transporter (MCT4), whose expression is upregulated by HIF-1α
[74]. Thus, when released into the TME, the L-lactated CAFs can be uptaken by the MCT1 present in the plasma membrane of glycolytic cancer cells, which acts as a fuel source
[75]. This is because cancer cells can oxidize the L-lactate to pyruvate in the mitochondria by lactate dehydrogenase, providing intermediate metabolites to the tricarboxylic acid cycle (TCA)
[75][76].
However, the L-lactate exported to the extracellular space promotes the acidification of the TME
[75]. The TME’s acidification inhibits the activation and proliferation of CD4+ and CD8+ lymphocytes, natural killer (NK) cells, and dendritic cells (DC)
[75] as well as causes the polarization of the macrophages toward the M2 phenotype
[75], contributing to immune evasion, which is recognized as a hallmark of cancer
[77]. The TME’s acidification also induces the synthesis of metalloproteinases (MMPs) in both cancer and stromal cells, facilitating extracellular matrix (ECM) degradation and, therefore, cancer cell migration and spread
[54][78].
Interestingly, studies have demonstrated that activation of HIF-1α by hypoxia increases the secretion of exosomes in both cancer
[79][80][81][82] and non-cancer cells within the TME
[83][84]. For this reason, hypoxia has been explored to increase the production of mesenchymal stem cell-derived exosomes for novel therapeutic strategies based on cell-free therapy
[10][84][85]. This occurs because the hypoxia increases the L-lactate production and, therefore, reduces the pH, increasing the exosome release and uptake, contributing to the crosstalk between cancer and non-cancer cells within the TME
[86][87][88].
In this sense, numerous studies have provided evidence that hypoxic cancer-derived exosomes regulate different EMT-related pathways in a miRNA-dependent manner
[82][89][90]. In this context, it was reported that the miR-665 identified in hepatocellular carcinoma-derived exosomes can downregulate Hippo signaling through directly targeting tyrosine phosphatase receptor type B (PTPRB)
[91], serving as a novel invasive biomarker for this malignancy
[92]. This is because the Hippo tumor suppressor signaling pathway is crucial to controlling cell proliferation and apoptosis by inhibiting the oncogenic coactivators Yes-associated protein (YAP)/transcriptional coactivator with the PDZ-binding motif (TAZ)
[93][94].
However, considering the plethora of biomolecules, especially miRNAs, delivered by cancer-derived exosomes, the mechanism of action of these vesicles on EMT could not be limited only to the Hippo signaling pathways.
In this sense, Yue et al.
[95] showed that exosomal miR-301a, secreted by hypoxic glioblastoma cells, targets transcription elongation factor A like 7 (TCEAL7), leading to the activation of the Wnt/β-catenin signaling pathway, resulting in the expression of the EMT-related transcription factors Snail, Slug, and Twist. Similar results were verified by Nam et al.
[96], who demonstrated that miR-301a functions as an oncogene in prostate cancer by directly targeting the p63 tumor suppressor, leading to loss of E-cadherin and EMT.
Thus, it is not surprising that cancer-derived exosomes can regulate different steps of the EMT, including cancer progression
[97], dissemination
[98][99], ECM remodeling
[100][101], stemness
[102], and metastasis
[103], though different miRNAs.
Interestingly, studies have demonstrated that exosomes derived from cancer-associated macrophages can also regulate stem cells’ dormancy
[104] and cell migration and invasion
[105], providing evidence that exosomes are also implicated in metastasis.
In this sense, lung cancer cell-derived exosomes (from the A59 and H358 cell lines) alter the transcriptional and bioenergetic signature of M0 macrophages, leading them to an M2 phenotype
[106]. However, the M2 macrophage-derived exosomes can transfer miR-21-5p and miR-155-5p to cancer cells, promoting the downregulation of transcription factor Brahma-related gene-1 (BRG1), leading to cell migration and invasion in colon cancer cells
[105][107]. Gastric cancer showed similar results; M2 macrophage-derived exosome-mediated apolipoprotein E (ApoE) transfer was found to increase the cancer cell migration in a PI3K/Akt signaling pathway activation-dependent manner
[108].
3.3.2. Exosomes in Angiogenesis
Tumor vascularization is crucial to guaranteeing the support of nutrients and meeting oxygen needs to sustain cancer growth. For this reason, the activation of HIF-1α also serves as a signal to induce sustained angiogenesis
[64][109]. Once phosphorylated, HIF-1α induces the expression of vascular endothelial growth factor (VEGF)
[109][110][111][112]. VEGF binds to VEGF receptors (VEGFRs)-1, -2, and -3, which are expressed on vascular endothelial cells, regulating vessel formation through endothelial cell migration
[113][114].
In this context, studies have demonstrated that cancer-derived exosomes act as a key regulator of angiogenesis
[115][116]. This is because exosomes derived from cancer cells can stimulate endothelial cell migration and tube formation independently of uptake
[117]. This response is mediated by the 189-amino-acid heparin-bound isoform of VEGF, which, unlike other common isoforms of VEGF, is preferentially enriched on the exosome surface
[117].
However, cancer-derived exosomes can also promote angiogenesis in an uptake-dependent manner. In this sense, Li et al.
[118] showed that hepatocellular carcinoma-derived exosomes transporting lysyl oxidase-like 4 (LOXL4) induce angiogenesis. In another study, Zhang et al.
[119] demonstrated that ovarian cancer-derived exosomes expressing prokineticin receptor 1 (PKR1) promote angiogenesis by promoting the migration and tube formation of HUVEC cells. Similar results were also described by Umezu et al.
[120], who demonstrated that hypoxia increases the production of multiple myeloma cell-derived exosomes transporting miR-135b, which can bind to factor-inhibiting hypoxia-inducible factor 1 (FIH-1) in endothelial cells, enhancing the formation of endothelial tubes. In another study, Zeng et al.
[121] showed that colorectal cancer-derived exosomes drive miR-25-3p to endothelial cells, targeting Kruppel-like factors 1 and 4 (KLF2 and KF4, respectively) and promoting vascular permeability and angiogenesis.
3.3.3. Cancer-Derived Exosomes Contribute to Pre-Metastatic Niche (PMN) Formation
Angiogenesis contributes to both cancer cell and cancer-derived exosome dissemination. However, the outcome of cancer metastasis depends on the interactions between metastatic cells and the host microenvironment
[122]. These interactions between the cancer cells (“seeds”) and the host microenvironment (“soils”) were first discovered by the English surgeon Stephen Paget in 1889
[122]. About 40 years later (in 1928), James Ewing postulated that metastasis is determined by a mechanism associated with hemodynamic factors of the vascular system
[123]. In a complementary hypothesis postulated in the 1970s, Isaiah Fidler demonstrated that, although the mechanical properties of blood flow are important, metastatic colonization only occurs at certain organ sites (organotropism)
[123]. Fidler’s theory was supported by additional discoveries, which revealed that tumors induce the formation of microenvironments in distant organs, facilitating the survival and outgrowth of cancer cells before they arrived at these sites
[123][124][125][126]. These predetermined microenvironments are termed ‘pre-metastatic niches’ (PMNs)
[127].
In the context of the “seed and soil” theory (Paget’s theory), the exosomes are similar to fertilizers, which can make barren land fertile and facilitate the colonization of cancer cells
[127][128][129][130]. This occurs because exosomes exhibit adhesion molecules on their surface, particularly integrins (ITGs), which bind to the ECM and organ-specific PMN receptors
[128]. Supporting this theory, in a study evaluating the biodistribution of exosomes from different cancer cell lines, Hoshino et al.
[131] provided evidence that cancer-derived exosomes are preferentially uptaken by tissues commonly recognized as metastatic sites. The authors also demonstrated that this site-specific biodistribution is associated with high expression levels of integrins (ITGα6, ITGβ4, and ITGβ1 for lung tropism; ITGβ5 and ITGαv for liver tropism; and ITGβ3 for brain tropism)
[131], reinforcing the view that the integrins involved in PMN formation.
Cumulative studies have provided evidence that the local inflammatory microenvironment drives the formation of PMNs as revisited by Guo et al.
[127]. In this sense, the exosomes play a key role in the metastatic process, inducing immune suppression in the PMN. This is because cancer cells release exosomes carrying programmed death-ligand 1 (PD-L1)
[127]. When PD-L1 binds to programmed death receptor 1 (PD-1), which is mainly expressed on macrophages and activated T or B cells, it provides an inhibitory signal, inducing T cell apoptosis and/or inhibiting T cell activation and proliferation
[132]. Thus, PD-L1/PD-1 binding allows the exosomes to circulate through the bloodstream without being recognized by immune cells
[127][133][134].
In addition, cancer-derived exosomes contain many immunomodulatory molecules that can impair the immune cell function, resulting in an immunosuppressive pre-metastatic microenvironment
[127]. These molecules can induce natural killer (NK) cell dysfunction, inhibit antigen-presenting cells, block T cell activation, and enhance apoptosis
[135][136].
However, the effects of cancer-derived exosomes in PMN formation are not limited to immune suppression. Studies have demonstrated that exosomes released from hypoxic tumors increase angiogenesis and vascular permeability in the PMN by carrying different miRNAs, such as miR-105 and miR-25-3p, which can disrupt the vascular endothelial barrier by targeting specific gene products
[130][131][137].
3.3.4. Exosomes in Cancer Stem Cell (CSC) Formation
Cancer stem cells (CSCs), also known as tumor-initiating cells (TICs), are a subset of cancer cells that share various features with stem cells, including the ability to self-renew and differentiation into the heterogeneous lineages of cancer cells, producing a variety of tumor cell subpopulations
[138][139][140][141]. In addition, these cells can induce cell cycle arrest (quiescent state), conferring chemo- and radio-resistance. This is because many common chemotherapeutic agents target the proliferating cells to lead to their apoptosis
[139]. Furthermore, CSCs overexpress ATP-binding cassette (ABC) transporters, increasing chemotherapeutics’ efflux
[142][143][144]. In addition, by exhibiting a high capability to repair DNA damage, the CSCs are resistant to radiation therapy (RT)
[145][146]. Thus, although the origin of CSCs remains incompletely understood
[147], it is clear that these cells are currently involved in therapeutic resistance
[148].
Cumulative evidence has shown that genomic instability contributes to CSC formation and accelerates the development of many genetically variable cancer stem cells, increasing the intratumor heterogeneity
[53][149][150][151][152].
However, recent studies have provided evidence that cancer-derived exosomes mediate crosstalk between the EMT and cancer stem cell (CSC) formation, acting as a key regulator of cell plasticity
[138].
In this sense, numerous studies have shown that cancer-derived exosomes mediate the instability of cadherins (which was verified during the EMT) in recipient cells by transferring oncogenic microRNAs and long non-coding RNAs (lncRNAs) as revisited by Wang et al.
[153].
The loss of E-cadherin, mediated by these non-coding RNAs
[153], promotes β-catenin release into the cytoplasm
[154]. Once translocated to the nucleus, β-catenin downregulates not only cell-junction-related genes (E-cadherin and claudin-7)
[53][155] but also upregulates stemness-related genes, facilitating the formation of CSCs
[156][157][158].
In addition, studies have also demonstrated that cancer-derived exosomes mediate drug resistance in several malignancies, which is considered a major impediment in medical oncology
[159].
Basically, there are two main types of resistance in cancer: (i) inherent resistance, where insensitivity already exists before treatment; and (ii) acquired resistance, which subsequently appears following the initial positive response
[159]. Interestingly, studies have demonstrated that cancer-derived exosomes mediate the acquired resistance by transferring microRNAs as revised by Bach et al.
[159].