Keywords

1 Introduction

The central dogma of molecular biology, which describes the flow of genetic information from DNA to proteins through the synthesis of RNA, can be summarized in a short sentence: “DNA transcribes RNA, which is then translated into proteins” [1]. However, protein-coding genes represent only 3% of the human genome. Our remaining genes code for RNAs that are not translated into proteins (ncRNAs). As their function was unknown for many years, these ncRNAs have been considered the dark matter of the human genome. Recently, there has been increasing evidence that ncRNAs actually play a critical role in the regulation of gene expression in both healthy and diseased cells [2, 3]. The family of ncRNAs can be divided into three major classes according to size: short, mid-length, and long ncRNAs. MiRNAs are short ncRNAs with an approximate length of 21 nucleotides, and due to their extraordinary abilities to regulate gene expression, they represent the most studied and characterized of all ncRNAs. Multiple cellular processes are controlled by miRNAs: proliferation, migration, invasion, apoptosis, differentiation, and drug resistance. Therefore, aberrant levels of miRNAs can alter these processes and lead to the development of cancer [4]. Genetic mutations (deletions, amplifications, mutation, translocations) and epigenetic alterations (methylation and histone modifications) are frequently observed in cancer cells and can be responsible for the dysregulated expression of miRNAs [5]. The tumor microenvironment can also regulate the expression levels of miRNAs of cancer cells. Soluble factors secreted in the tumor microenvironment, such as growth factors [e.g., epidermal growth factor (EGF)] and cytokines [e.g., transforming growth factor beta (TGF-β), IL-6], can affect miRNA expression [6]. Therefore, an altered tumor microenvironment, for example, during chronic inflammation, can have a great impact on tumor cell functions through the regulation of miRNA expression.

In the past few years, it has been discovered that miRNAs can also perform their regulatory functions outside the cells where they are expressed. Indeed, bioactive miRNA molecules can be secreted from their cells of origin into the extracellular space, be delivered to recipient cells (tumor and non-tumor cells), and regulate the recipient cells’ gene expression in a hormone-like fashion [7]. In this way, tumor cells can regulate their surrounding microenvironments and create favorable conditions allowing them to survive, proliferate, escape from attacks by immune cells, and eventually disseminate to distant organs and generate metastases.

2 MiRNA Biogenesis

The past decade has seen an increasing interest in the roles of miRNAs in tumor cells, due to miRNAs’ abilities to regulate the expression of genes controlling multiple cell processes frequently altered in cancer (e.g., cell cycle, proliferation, migration/invasion, differentiation, apoptosis) [4]. MiRNAs are some of the most abundant genes: there are 2588 miRNAs in humans, according to the latest miRNA database (miRBase). MiRNAs are short ncRNAs ~21 nucleotides (nts) in length, and they are encoded by sequences located within introns or exons of genes or in intergenic regions [8,9,10,11]. MiRNA genes are transcribed in the nucleus by RNA polymerase II (Pol II) as primary transcripts (pri-miRNAs), which are long several kilobases and contain characteristic hairpin structures. Following transcription, the pri-miRNA hairpin structure is recognized by a complex called Microprocessor—composed of the nuclear ribonuclease DROSHA (RNase III) and the essential cofactor DGCR8—which processes the stem-loop and generates a small hairpin-shaped RNA (pre-miRNA) of ~65 nucleotides in length. Then, the pre-miRNA is exported to the cytoplasm, where it matures. The export step is mediated by a transport complex composed of the pre-miRNA, Exportin 5, and the GTP-binding nuclear protein Ran-GTP. The complex drives the pre-miRNA through the nuclear pores and into the cytoplasm where it is released following the disassembly of the transport complex. Next, the pre-miRNA is cleaved by a complex composed of DICER (RNase III-type endonuclease) and the transactivation-responsive RNA-binding protein (TRBP), forming a double-stranded miRNA molecule. During the final maturation step, this RNA duplex is incorporated into the RNA-induced silencing complex (RISC), whose primary component is the Argonaut protein 2 (AGO2). In the RISC, the double-stranded miRNA is unwound, generating two single miRNA strands: the mature (guide) and passenger strands [12]. In general, the passenger strand is quickly degraded, whereas the mature miRNA strand, which remains in the RISC, is guided to the target-site sequence of a messenger RNA (mRNA) to either inhibit mRNA’s translation into protein or initiate its degradation. Particularly, miRNAs bind to specific binding sites in the 3′-untranslated regions (3′ UTRs) of mRNAs with different levels of complementarity, affecting the mechanism of translation inhibition. Perfect or near-perfect annealing between a miRNA and its binding site sequence induces the degradation of the target mRNA by RISC, whereas imperfect or partial annealing inhibits ribosomes’ access to the target mRNA, blocking translation. The power of miRNAs in gene expression regulation is vast, as a single miRNA can target hundreds of different mRNAs. Therefore, altered expression of miRNAs can have significant impacts on many biological functions and pathways regulated by target mRNAs, leading to transformation of normal cells into tumor cells and the progression of cancer.

3 Regulation of miRNA Expression in Human Cancers

3.1 Genetic Alteration

Human miRNAs are frequently located at chromosomal fragile sites and in genomic regions involved in cancer [13]. These regions are associated with increased probabilities of various genetic alterations (such as deletions, insertions, amplifications, single point mutations, transitions, and transversions) that occur in different cancer types with different frequencies [5]. These genetic alterations have significant impacts on the cellular levels of miRNAs, leading to their aberrant expression and, accordingly, altered regulatory functions in several pathways involved in tumorigenesis and tumor progression. The first evidence of the correlation between genetic alterations and aberrant miRNA expression was found for the tumor suppressors miR-15a and miR-16-1, which are located at chromosome 13q14. Deletion of this region, which occurs in more than 65% of chronic lymphocytic leukemia (CLL) cases, as well as in 50% of mantle cell lymphomas, 16–40% of multiple myelomas, and 60% of prostate cancers, causes a reduction in miR-15a and miR-16-1 levels [14, 15]. Other examples include the tumor suppressor miR-34a mapped on 11q23-q24, which is lost in breast and lung cancer; miR-123 located at 9q33, which is frequently deleted in non-small cell lung cancer (NSCLC); and miR-145 and miR-143 located at 5q32, which is frequently deleted in myelodysplastic syndrome [13]. On the other hand, genomic amplifications induce elevated levels of some miRNAs. The oncogenic miR-21 is located at 17q23 and is amplified in many tumors, including breast, colon, lung, pancreas, stomach, and prostate tumors [16]. The potent oncogenic miR-17-92 cluster is encoded by the 13q31-32 locus, which undergoes amplification in large B-cell lymphoma, follicular lymphoma, mantle cell lymphoma, and primary cutaneous B-cell lymphoma [13, 17] and is overexpressed in many cancer types, including lymphoma, colon, lung, breast, pancreas, and prostate cancer [16, 18, 19]. Elevated levels of miR-569 are associated with amplification of the 3q26.2 locus in ovarian and breast cancer [20]. Mutations at the DNA level can also affect the transcription of miRNAs or the maturation of pri- and pre-miRNAs, resulting in altered expression of mature miRNAs. Point mutations located in regions containing pri-miRNA recognition and processing motifs, which enhance pri-miRNA processing, determine reductions in mature miRNA levels. For instance, mutations in the basal UG and/or CNNC motifs affect the processing of pri-miR-16 and pri-miR-30a into their mature forms [21].

Mutations can also occur in the key regulators of miRNA biogenesis, DICER and DROSHA, greatly contributing to aberrant expression of miRNAs and cancer development. Indeed, mutations in DICER and DROSHA have been identified in diverse types of cancers and enhance cellular transformation and tumorigenesis [22,23,24,25].

3.2 Epigenetic Regulation

Besides the genetic alteration reported above, epigenetic regulation can also contribute to the aberrant expression of miRNAs in cancer. Indeed, DNA methylation and histone modification play important roles in the regulation of miRNA expression. Many miRNAs are embedded in CpG islands, and their promoter regions can undergo heavy methylation. DNA methylation and histone acetylation induce chromatin remodeling that regulates the transcription machinery’s access to promoter regions, controlling miRNA expression [26]. The first evidence of alteration of miRNA expression by epigenetic changes was in breast and bladder cancer. Particularly, a rapid alteration of miRNA levels was measured in response to inhibitors of histone deacetylase (HDACi) [27] and DNA methylation [28]. Another important example of epigenetic regulation is represented by the miR-34 family, whose expression is repressed by hypermethylation in a variety of cancer types including breast, ovarian, esophageal, gastric, colon, renal, pancreatic, NSCLC, acute lymphocytic leukemia (ALL), and CLL. The downregulation of the tumor suppressor miR-34 is particularly relevant, as miR-34 cooperates with TP53 to suppress prostate cancer by regulating the stem cell compartment [29]. On the other hand, hypomethylation can induce the overexpression of some miRNAs, such as the oncogenic miRNAs (oncomiRs) miR-21 and miR-29b. Hypomethylation determines high levels of these miRNAs in breast cancer and is associated with aggressive characteristics of tumor cells [30].

3.3 MiRNA Regulation by Oncogenes and Tumor Microenvironment

Other mechanisms can regulate the expression levels of miRNAs. Oncogenes and tumor suppressor genes can activate or repress the expression of miRNAs working as transcription factors or repressors. Therefore, either overexpression of oncogenes or downregulation of tumor suppressor genes can impact miRNA levels in cancer cells. For instance, TP53 directly transactivates miR-34 transcription by binding to the miR-34 promoter [31]. The oncogene MYC positively regulates the transcription of the miR-17-92 cluster and can also repress the expression of miR-34 [32].

The tumor microenvironment can also play an important role in regulating miRNA expression in cancer cells. A variety of cytokines produced in the tumor microenvironment are released directly by tumor cells or by tumor associated cells (e.g., immune and stromal cells) [33, 34] and can regulate the expression of miRNAs involved in tumor pathogenesis and progression. For example, the pro-inflammatory cytokine IL-6 induces the expression of TWIST, SNAIL, and ZEB1, key transcription factors that regulate the epithelial-mesenchymal transition (EMT). During EMT, tumor cells lose epithelial characteristics and acquire motility and invasive abilities [35, 36]. TWIST can bind directly to the putative promoter of miR-10b and induce its transcription [37], whereas SNAIL and ZEB1 bind to E-boxes in the miR-34a promoter, repressing miR-34a expression [38]. ZEB1 can also repress the expression of members of the miR-200 family by binding directly to their promoter sequences [39,40,41]. IL-6 can also suppress the expression of miR-34 through the activation of its signaling pathway mediator STAT3, which can bind to a conserved STAT3-binding site in the first intron of the miR-34 gene [42]. The TGF-β/Smad pathway regulates the expression of miR-155, and increased levels of miR-155 induce the acquisition of migration and invasion ability in breast cancer by targeting RHOA [43]. Additionally, growth factors can regulate the expression of miRNAs. For instance, EGF can modulate the expression of miR-30b, miR-30c, miR221, and miR-222, which play important roles in gefitinib-induced apoptosis and EMT in NSCLC [44].

Therefore, in addition to genetic alterations in tumor cells, soluble factors released in the tumor microenvironment play significant roles in the regulation of the expression of miRNAs in tumor cells. This may be particularly important under conditions of chronic inflammation [45, 46], both in the early stages of cancer formation and later during the progression of the disease, as cytokines released in the microenvironment can provide signals that regulate the expression of miRNAs controlling tumor cell growth, differentiation, motility, angiogenesis, and resistance to treatment, as well as anti-tumor immune response [47,48,49,50,51].

4 miRNAs Regulate All Cancer Hallmarks

The hallmarks of cancer are the six fundamental biological capabilities acquired by tumor cells during the multistep process regulating the development of human tumors: self-sufficiency in growth signals, insensitivity to growth-inhibitory (antigrowth) signals, evasion of programmed cell death (apoptosis), limitless replicative potential, sustained angiogenesis, and tissue invasion and metastasis [52]. MiRNAs are involved in the regulation of all of these hallmarks. Accordingly, the aberrant expression of miRNAs significantly impacts the dysregulation of the stepwise cancer development. Some examples are reported below.

To move into an active proliferative state, normal cells require growth signaling provided by growth factors, extracellular matrix components, and cell-to-cell adhesion/interaction molecules. Whereas this dependence on growth signaling is a characteristic of normal cells, tumor cells show a greatly reduced dependence on exogenous growth stimulation. An example is the activation of the oncogene RAS that allows tumor cells to grow independently of external signals. It has been reported that all three RAS genes (K-RAS, N-RAS, and H-RAS) are directly regulated by the let-7 miRNA family [53].

Normal tissue is characterized by the presence of multiple antiproliferative signals that aim to maintain cellular quiescence and tissue homeostasis. The transcription factor FOXO1 is a tumor suppressor that regulates cell cycle progression, proliferation, and apoptosis. FOXO1 is the target of three miRNAs, miR-96, miR-182, and miR-183, and its repression by increased levels of these miRNAs leads to increased proliferation and reduced apoptosis [54].

Tumor cells’ ability to expand in number is determined by the balance between the rates of cell proliferation and cell attrition. Apoptosis, also known as programmed cell death, represents a major source of attrition. Tumor cells are characterized by resistance to apoptosis. The tumor suppressor miR-34 induces cell cycle arrest and subsequent caspase-dependent apoptosis by targeting BCL-2, an important anti-apoptotic regulator [55].

Cells have a finite replicative potential, and after a certain number of cell divisions, they stop growing and enter into senescence. It was reported that during senescence, miR-29 and miR-30 are upregulated and target the oncogene MYBL2, inhibiting DNA synthesis [56].

Angiogenesis is an important process that supplies oxygen and nutrients for cell function and survival. Significant amounts of proangiogenic factors are produced and secreted in tumor tissue to promote neovascularization, which supports the growth of tumor cells. The vascular endothelial growth factor (VEGF) is the most important proangiogenic factor regulating neoangiogenesis and is highly expressed in most cancers. It is induced by low oxygen concentration (hypoxia) in the tumor microenvironment. Hypoxia can modulate expression of several hypoxia-regulated microRNAs (HRMs), some of which, such as miR-210, miR-26, and miR-181, are directly controlled by hypoxia-inducible factor (HIF) [57]. It was found that miR-210 is involved in the regulation of endothelial cell chemotaxis and tubulogenesis [58]. Another HRM, miR-27a, targets the zinc finger gene ZBTB10, a negative regulator of the specific-protein (SP) transcription factors (such as Sp1, Sp3, and Sp4), resulting in the induction of Sp-dependent survival and angiogenic genes, including survivin (BIRC5), VEGF, and VEGF receptor 1 (VEGFR1) [59].

Metastasis is a complex process involving multiple steps that endow tumor cells with invasive properties (migration and invasion ability), intravasation (blood or lymphatic vasculature), blood circulation survival, extravasation, and growth at a new site [60]. It was reported that miR-10b can promote invasion and metastasis by targeting the transcription factor homeobox D10 (HOXD10) [37], which represses the expression of genes involved in cell migration and extracellular matrix remodeling: RHOC, α3 integrin (ITGA3), matrix metalloproteinase-14 (MMP-14), and urokinase-type plasminogen activator receptor (UPAR) [61].

Increasing evidence suggests that two additional hallmarks may be involved in the regulation of cancer development: reprogramming of energy metabolism and evasion of immune destruction. Tumor cells are able to adjust their metabolic pathways according to their energy requirements. The metabolism of glucose and glutamine represents the major source of energy for cells, and pathways using these two nutrients are often altered in cancer cells [62]. Oncogenes and miRNAs are involved in the regulation of these metabolic pathways. For instance, MYC modulates the metabolism of glutamine by repressing miR-23a/b, resulting in increased expression of their target protein, mitochondrial glutaminase [63]. Furthermore, glycolytic pathways are regulated by several miRNAs, including miR-378 [64] and miR-143 [65].

According to the immune surveillance hypothesis, the immune system plays an important role in recognizing and eliminating incipient and advanced stage tumors (metastasis). However, many tumors are able to evade immune cell attack using tumor immune escape mechanisms that include alterations in: tumor antigen processing and presentation by human leukocyte antigen (HLA) class I and II; signal transduction pathways; expression of co-stimulatory and co-inhibitory molecules; and secretion of immune-suppressive mediators [66]. Recently, the roles of miRNAs in regulating tumor immunogenicity and antitumor immune responses have been unveiled [67, 68]. For instance, miR-9 and miR-346 regulate the expression of MHC class I antigen processing machinery (APM) components and interferon (IFN)-induced genes [69, 70], and miR-148a and miR-181a target the expression of HLA-C and HLA-A, respectively [71, 72]. The major histocompatibility complex class I-related molecules (MICs) A and B are the ligands of the activating NK cell receptor NKG2D, which mediates NK cell-mediated cytotoxicity. The expression of MICA and MICB is controlled by miR-20a in breast cancer stem cells, resulting in reduced sensitivity to NK cell-mediated lysis and enhanced metastatic potential [73]. The B7 family includes both co-stimulatory and co-inhibitory molecules (CD80, CD86, CD28, CTLA-4, PD-1, PD-L1, PDL2, ICOSL) that play important roles in immune responses [74]. It has been found that miRNAs can regulate the expression of B7 family members. For instance, the expression of PD-L1 is controlled by miR-570 in gastric cancer [75], miR-34a in acute myeloid leukemia (AML) [76], miR-200 in NSCLC [77], and miR-138-5p in colorectal cancer (CRC) [78]. Recently, it was found that the expression of high levels of miR-124 could reverse the immunosuppressive phenotype of glioma cancer stem cells by targeting STAT3 signaling and reducing the generation of FoxP3+ regulatory T cells (Treg) [79].

5 Circulating miRNAs as Tumor Biomarkers

Recently, it was observed that miRNAs can be released into the extracellular space [80] and can be detected in many biological fluids, such as serum, plasma, urine, saliva, and breast milk [81]. These circulating miRNAs can be actively secreted outside the cell either encapsulated within exosomes [82] or in an extracellular vesicle-free manner associated with the Ago2 protein [83, 84]. They can also be passively secreted into the blood circulation as a result of apoptotic [85] or necrotic cell death [86]. Importantly, aberrant levels of miRNAs can be detected not only in tumor cells but also in the biological fluids of cancer patients, possibly reflecting the expression patterns of the tumor tissues from which circulating miRNAs originate [87]. Due to their extraordinary stability in body fluids, resistance to storage handling, and the ease of assessment by quantitative PCR and miRNA microarrays [80, 88], circulating miRNAs are considered suitable biomarker molecules to differentiate normal from diseased states and monitor both progression of cancer and response to therapy (Table 6.1). Indeed, tumor-specific miRNAs were identified for the first time in the serum of patients with diffuse large B-cell lymphoma; in these patients, high levels of miR-21 were correlated with improved relapse-free survival [89]. Since then, many other studies have been published reporting the potential use of circulating miRNAs as tumor biomarkers in different types of cancer [7], such as miR-141 in prostate cancer [80], miR-486, miR-30d, miR-1, and miR-499 in NSCLC [95], miR-17-3p and miR-92 in CRC [96], miR-195 and let7-a in breast cancer [87], and miR-500 in liver cancer [97].

Table 6.1 Circulating miRNAs in cancer
Full size table

The potential use of tumor-specific miRNAs as diagnostic markers for cancer has been confirmed not only in serum and plasma but also in other body fluids, such as saliva and urine. For instance, levels of miR-125a and miR-200a were significantly lower in the saliva of cancer patients with oral squamous-cell carcinoma compared to healthy controls [100]. In another study, it was reported that miR-31 levels were higher in the saliva of patients with oral squamous-cell carcinoma compared to healthy controls, and a decrease in miR-31 levels was measured after tumor resection. The latter result was also demonstrated in plasma [101]. Finally, increased levels of miR-126, miR-152, and miR-182 were found in the urine of patients with bladder cancer, and the ratios of miR-126 to miR-152 and miR-182 to miR-152 could indicate the presence of bladder cancer with a specificity of 82% and a sensitivity of 72% [102].

Circulating miRNAs were also studied for their ability to predict prognosis and response to therapy. For instance, miR-125b and miR-532-3p predict the efficacy of rituximab-mediated lymphodepletion in chronic lymphocytic leukemia patients [90]. Six serum microRNAs can predict lymph node metastasis in cervical squamous cell carcinoma patients: miR-1246, miR-20a, miR-2392, miR-3147, miR-3162-5p, and miR-4484 [98]. In hormone-refractory prostate cancer, high serum miR-21 levels could identify patients who were resistant to docetaxel-based chemotherapy [94]. Circulating let-7a and miR-16 levels can predict progression-free survival and overall survival in patients with myelodysplastic syndrome [92]. Circulating miR-200c levels significantly predict prognosis and response to therapy in patients undergoing neoadjuvant chemotherapy for esophageal cancer [99]. Plasma levels of miR-155 can predict response to therapy in patients with chronic lymphocytic leukemia [91]. Finally, high levels of miR-141 in the serum of breast cancer patients were associated with shorter brain metastasis-free survival and were an independent predictor of both progression-free survival and overall survival [93].

6 Extracellular Vesicles Mediate Intercellular Communication in the Tumor Microenvironment

During the past few years, there has been increasing evidence to support the concept that miRNAs are able to mediate intercellular communication. This exchange of genetic information is mediated by extracellular vesicles (EVs), carrying miRNAs and other molecules, that are secreted by donor cells and taken up by recipient cells through several mechanisms [103] (Fig. 6.1). The first evidence of miRNA transfer was provided by Valadi et al. [82], who showed that functional RNA molecules (mRNAs and miRNAs) are transferred between mast cells through exosomes. Exosomes are extracellular vesicles of endosomal origin with diameters ranging from 30 to 100 nm [104]. The generation of exosomes is a highly controlled multi-step process [105] (Fig. 6.1). In the first step, the cell membrane buds inward, forming early endosomes in the endocytic pathway. Then, the early endosome membrane invaginates to generate multivesicular bodies (MVBs), each consisting of a large endosome containing exosomes of different sizes (called intraluminal vesicles or ILVs). During this second inward budding, the exosomes are loaded with different cellular components, including coding RNAs (mRNAs), short and long non-coding RNAs (miRNAs and lncRNAs), proteins, and DNA [82, 106,107,108,109]. In the final step, the fusion of MVBs with the plasma membrane allows the release of ILVs into the extracellular space as exosomes. The regulation of exosome formation, cargo loading, and secretion involves several mechanisms [110]. Ceramide, synthesized by the neutral sphingomyelinase 2 (nSmase2), is involved in the budding of ILVs from MVBs and in exosome secretion [111, 112]. RAB proteins, such as RAB11, RAB27 and RAB35, participate in vesicle trafficking and exosome secretion [113]. The endosomal sorting complex required for transport (ESCRT) mainly regulates protein sorting into MVBs in a ubiquitin-dependent manner [114]; whereas miRNA loading into exosomes is regulated by ceramide and heterogeneous nuclear ribonucleoprotein A2B1 (hnRNPA2B1) [112, 115].

Fig. 6.1
figure 1

Biogenesis of exosomes and cell-to-cell communication. Exosome biogenesis starts with the inward budding of the plasma membrane to form a clathrin-coated vesicle (CCV) and then an early endosome. Next, a second inward budding of the endosome membrane will generate a multivesicular body (MVB) containing exosomes. During the second inward budding, exosomes are loaded with their cargo (mRNAs, ncRNAs, proteins, and DNA fragments). The MVB can be directed either to the lysosome for degradation and recycling of MVB components or to the plasma membrane for secretion. Finally, the MVB fuses with the plasma membrane, and exosomes are released into the extracellular space. Secreted exosomes can be taken up by recipient cells through several mechanisms: (1) receptor-mediated endocytosis; (2) phago- and micropino-cytosis; (3) direct fusion with the recipient plasma membrane; (4) clathrin-, caveolin-, and lipid raft-mediated endocytosis

Full size image

Exosomes are secreted from all types of cells and represent potent vehicles for cell-to-cell communication [116], as they can naturally deliver genetic and protein cargo to recipient cells and regulate these cells’ biological functions. This intercellular communication mechanism is particularly important in cancer, as tumor cells produce significant amounts of EVs. Accordingly, the altered composition of cancer cell-derived exosomes’ cargo can mediate dysregulated signaling. Furthermore, other components of the tumor microenvironment, such as mesenchymal stromal cells (MCS), cancer associated fibroblasts (CAFs), and immune cells (macrophages, dendritic cells, T cells, and NK cells), can participate in EV-mediated crosstalk with tumor cells and regulate their biological functions (Fig. 6.2). This generates a niche that facilitates tumor progression by regulating proliferation, differentiation, angiogenesis, metastasis, anti-tumor immune responses, and drug resistance.

Fig. 6.2
figure 2

Exosome-mediated cell-to-cell communication in the tumor microenvironment. Exosomes mediate intercellular communication between tumor cells and other cellular components of the tumor microenvironment. EC endothelial cells, MM BM-MSCs multiple myeloma bone marrow mesenchymal stromal cells, DC dendritic cells, Th cells T helper cells, CD8+ CTL CD8+ cytotoxic T lymphocytes, Treg T regulatory lymphocytes, NK natural killer cells, TAM tumor-associated macrophages, CAFs cancer-associated fibroblasts, NF normal fibroblasts

Full size image

Bone marrow mesenchymal stromal cells from the tumor microenvironment of multiple myeloma patients (MM BM-MSCs) support the growth of multiple myeloma (MM) cells, whereas the bone marrow mesenchymal stromal cells from healthy donors (BM-MSCs) inhibit the growth of MM cells. Exosomes from MM BM-MSCs contain lower levels of the tumor suppressors miR-15 and miR-16 and higher levels of IL-6 and CCL2 compared with exosomes from BM-MSCs. The exosomes released from MM BM-MSCs can deliver their cargo to MM cells, playing a crucial role in MM pathogenesis, tumor growth, and disease progression [117]. The transfer of exosomal miRNAs to endothelial cells can promote angiogenesis and metastasis. It was found that exosomal miR-9 secreted by tumor cells induced endothelial cell migration and an in vivo increase in endothelial density, which promoted tumor growth [118]. Exosomes secreted by the leukemia cell line K562 carry miR-210, which increases tube formation by human umbilical vein endothelial cells [119]. Exosomal miR-210 can also be secreted by breast cancer cells and taken up by endothelial cells, promoting angiogenesis [112]. Exosomal miR-105 secreted from breast cancer cells can target cellular tight junctions and disrupt vascular endothelial barriers during early premetastatic niche formation [120].

Exosomes released by tumor cells can also contribute to the dissemination of malignant cells by remotely regulating a metastatic site. Indeed, exosomes from melanoma cells conditioned lymph node tissue and induced microanatomic niches that allowed metastasis of melanoma cells to lymph nodes [121]. Another study showed that exosomes released by renal cancer stem cells stimulated angiogenesis and the formation of a premetastatic niche in lung tissue [122]. Exosomes can also mediate communication between tumor and immune cells. Exosomes secreted from tumor associated macrophages (TAMs) can deliver miR-223 to breast cancer tumor cells, increasing their invasive abilities [123]. Cancer cell-derived exosomes containing miRNAs can also regulate the functions of immune cells. For instance, exosomal miR-21 and miR-29 released by mouse lung cancer cells can bind to Toll-like receptors (TLRs) 8 and 7 of mouse macrophages and activate the NF-κB pathway. This induces an inflammatory response mediated by TNF-α and IL-6, which promote tumor growth and metastasis [124]. An interesting miRNA-mediated bidirectional crosstalk between neuroblastoma (NBL) cells and monocytes was recently described. Particularly, exosomal miR-21 released by NBL cells can induce the expression of miR-155 in human monocytes. In turn, miR-155 is delivered from monocytes to NBL cells through exosomes and regulates resistance to cisplatin treatment [125].

Tumor cell-derived exosomes can also regulate the functions of immune cells present within the tumor microenvironment [126, 127]. Dendritic cells (DCs) play an important role in activation of anti-tumor immune responses, and their functions can be altered by tumor-derived-exosomes. For instance, exosomal miR-212-3p released from pancreatic tumor cells (PANC-1) can be transferred to DCs and affect their immune functions by inducing immune tolerance [128]. MiR-203 is expressed in PANC-1 cells, and its exosome-mediated delivery induced the downregulation of TLR4, TNF-α, and IL-12, resulting in impairment of immune response activation [129]. Exosomes from nasopharyngeal cancer cells containing miR-24-3p impair T cell proliferation and differentiation into Th1 and Th17 cells and promote the induction of T regulatory CD4+ CD25high Foxp3+ lymphocytes (Tregs) [130]. Tumor-derived microvesicles can deliver miR-214 to CD4+ T cells and promote the expansion of Tregs by targeting phosphatase and tensin homolog (PTEN), resulting in enhanced immune suppression [131]. High levels of miR-210 and miR-23a are present in hypoxic tumor-derived microvesicles and can be transferred to natural killer (NK) cells, leading to impairment of cytotoxicity against different tumor cells in vitro and in vivo [132]. Tumor-derived microvesicles inhibit proliferation and induce apoptosis of activated CD8+ T cells [133].

Cancer-associated fibroblasts (CAFs) are the major components of tumor stroma and can participate in exosome-mediated crosstalk with tumor cells. Indeed, it was found that exosomes released by breast cancer CAFs had increased levels of miRs-21, -378e, and -143 compared to normal fibroblasts. Transfer of these exosomes to breast cancer cells induced a significantly increased capacity to form mammospheres, induced stem cell and EMT markers, and promoted anchorage-independent cell growth [134]. High levels of the pro-metastatic miR-9 are found in various breast cancer cell lines. Exosome-mediated delivery of miR-9 can modify the properties of human breast fibroblasts and promote in vivo tumor growth by enhancing the switch from the normal fibroblast (NF) state to the CAF state [135].

Exosomes can also play important roles in regulating drug resistance [136,137,138]. It was reported that exosomes from tamoxifen-resistant breast cancer cells (MCF-7TamR) could transfer miR-221/222 and induce drug resistance in recipient breast cancer cells by targeting p27 (CDKN1B) and ERα (ESR1) [139]. In advanced renal cell carcinoma (RCC), the bioactive lncRNA named lncRNA Activated in RCC with Sunitinib Resistance (lncARSR) can be incorporated into exosomes. LncARSR can then transmit sunitinib resistance to sensitive cells by competitively binding miR-34/miR-449 to facilitate AXL and MET expression [140]. Exosomes can also perform their regulatory functions by interacting with drugs in the extracellular space. Indeed, it was found that exosomes released by the HER2-overexpressing tumor cell lines SKBR3 and BT474 express a full-length HER2 molecule, can bind to trastuzumab (anti-HER2 humanized monoclonal antibody), and accordingly reduce free molecules of trastuzumab. Exosomes with bound trastuzumab have been found in both HER2-positive tumor cell-conditioned supernatants and serum from breast cancer patients, which resulted in modulation of sensitivity to trastuzumab [141]. Drug-efflux pumps inserted in the exosome membrane can mediate drug sequestration from the cytoplasm. The presence of drug efflux transporters P-glycoprotein (P-gp) and ATP-Binding Cassette G2 (ABCG2) on EVs of breast cancer cells enabled the influx of drugs into the microvesicular compartment, resulting in an active sequestration of chemotherapeutic drugs from the cytoplasm [142, 143]. Then, chemotherapeutic drugs encapsulated inside EVs/exosomes can be expelled by active secretion [144, 145]. Interestingly, the transfer of EVs containing drug efflux pumps from drug-resistant to drug-sensitive tumor cells can contribute to the acquisition of multidrug resistance phenotypes by recipient cells [138].

7 Conclusions

MiRNAs play crucial roles in the regulation of physiological functions in normal cells; therefore, alterations in miRNA expression levels have significant impacts on cells’ biology. Indeed, aberrant miRNA levels are associated with carcinogenesis and cancer progression. Dysregulation of the expression of miRNAs results from regulatory events at both the intracellular level (genetic and epigenetic modifications) and the extracellular level (signaling from the tumor microenvironment). Because a single miRNA has the potential to regulate the expression of up to 100–200 target genes, it is easy to understand how one or a few genetic alterations inside a cell or a disruption in homeostasis in the microenvironment can significantly impact many biological functions, as shown by the roles of miRNAs in regulating the hallmarks of cancer. Altered levels of circulating miRNAs can reflect a pathological status; therefore, these miRNAs can serve as predictive and prognostic biomarkers of cancer. Via secretion into the extracellular space, miRNAs can also perform their regulatory functions outside their cells of origin. Indeed, circulating miRNAs are important mediators of cell-to-cell communication, regulating crosstalk both locally, among different cellular components of the tumor microenvironment, and remotely, by regulating premetastatic niches.

A comprehensive understanding of miRNA functions at multiple levels will allow for the development of more precise and less toxic targeted treatments for cancer.