Exosomes (Exo) have emerged as potent amplifiers of pro-tumorigenic signals to distant cells. The knowledge of their role in colorectal cancer (CRC) is continuously up-growing, although their contribution to metastasis remains largely unclear. Liu et al. (Clinical Science (2020) 134, https://doi.org/10.1042/CS20191087) in their work have described a novel mechanism by which CRC-derived Exo promote metastasis through the down-regulation of the deleted in liver cancer-1 (DLC-1), a gene involved in the epithelial-to-mesenchymal transition (EMT) event in cancer cells. The Authors also demonstrated an increase in serum exosomal miR-106b-3p in patients with metastatic CRC, suggesting its potential implication as a prognostic biomarker. These findings may be of great effort in clarifying the underlying mechanisms of CRC metastasis and provide new targets for future researches.
Metastasis is the major cause of death in patients suffering from colorectal cancer (CRC). Many efforts have been devoted in the past to unraveling the machinery driving the dissemination of malignant cells from the primary tumor toward distant sites. Particular interest has been dedicated to the role of angiogenesis in favoring the CRC metastasis process, and the inhibition of the vascular endothelial growth factor (VEGF) by angiogenesis inhibitors primarily improved the survival of patients with metastatic CRC . Other molecular targeting drugs, as anti-EGFR MoAbs, have further extended the overall survival at least in eligible patients. Additional steps to gain both invasive and pro-metastatic behavior by CRC cells include the escape from immune surveillance, as well as increased cell motility and ability to intravasate into lymphatic and blood vessels. Particularly, the loss of epithelial characteristics in favor of a mesenchymal-like phenotype, namely epithelial-to-mesenchymal transition (EMT), is essential for CRC cells to overcome the cell-to-cell adhesion, lose the apical-basolateral cell polarity as well as increase both motility and invasiveness .
In the last decade, many data have been produced in favor of the role of extracellular vesicles (EVs) in cancer. These particles are considered key intercellular communicators delivering both tumorigenic and pro-metastatic signals within the tumor microenvironment and potential metastatic sites . In this regard, exosomes (Exo) are nano-sized (40–130 nm) EVs constitutively released by all types of cells, which carry a variable amount of active molecules including proteins, DNA fragments as well as coding and non-coding RNAs. A great heterogeneity of these EVs, however, has been described. Distinct subpopulations include large (L)-Exo (>80 nm), small (S)-Exo (60–80 nm) and non-membranous nanoparticles (∼35 nm), namely ‘exomeres’, suggesting different biological roles [4,5]. Their role in both physiological and pathological events like cancer has overwhelmingly come to light and their contribution to CRC progression and metastasis has been intensively explored although partly elucidated. Exo take part, indeed, in almost all steps of the CRC cancerogenesis, since they affect cell proliferation and invasiveness, angiogenesis, evasion from immune surveillance and pre-metastatic niche formation. Moreover, up-growing knowledge has been progressively accumulated on the role of CRC-derived Exo within the tumor microenvironment to exert pro-tumorigenic effects on the stromal accessory cells. To this, several tumorigenic signals are conveyed from tumor cells to cancer-associated fibroblasts (CAFs) to display a highly proliferative and angiogenic phenotype as well as reprogram their metabolic asset or secretion of extracellular matrix (ECM)-remodeling proteins [6,7]. For a detailed description of the role of Exo in CRC we recommend a recent review of our group .
The contribution of tumor-derived Exo in the metastatic evolution of CRC also depends on the activation of the EMT. Previous works highlighted the delivery of exosomal miR-210 via CRC-derived Exo as a possible mechanism promoting the EMT in cancer . However, the present study mostly focused on the variation of both E-cadherin and N-cadherin expression following the stimulation of CRC cells by exosomal miR-210 with lesser attention to other associated events. A step forward in this field has been inspired by the data of Liu et al., describing a novel mechanism engaged by CRC cells to activate the EMT, which includes the Exo-mediated microRNA (miRNA) exchange .
Using a systematic approach, the authors profiled a large number of miRNAs from serum Exo of patients with both metastatic and non-metastatic CRC. Although the investigation included a small number of patients (n=10), they showed a higher expression of miR-106b-3p, miR-6842-3p and miR-151a-3p in patients with metastatic CRC as compared with the non-metastatic ones. The same miRNAs were overexpressed in formalin-fixed paraffin-embedded (FFPE) samples of CRC in comparison with normal mucosae, while only one of these, namely the miR-106b-3p, was found significantly increased in primary CRC samples of metastatic patients as compared with those with non-metastatic disease. In a wide validation cohort (n=100), the miR-106b-3p also resulted significantly enriched in serum Exo of patients with metastatic CRC while lower levels were detected in non-metastatic CRC or healthy controls. Moreover, increased Exo miR-106b-3p was correlated with worsened TNM stage and prognosis, while a decrement in circulating levels was observed after surgical tumor removal.
The role of miRNAs in cancer and particularly in CRC, has been widely described. They are small non-coding RNAs (19–22 nucleotides) that epigenetically regulate the protein expression by interfering with mRNA transcription. MiRNAs are variably expressed by cells and their deregulation in cancer development definitely affects a number of pathways controlling cell metabolism, proliferation, differentiation and apoptosis [11,12]. The increased expression of miR-106b-3p has been observed in various types of human cancer. Previous studies suggested that miR-106b-3p is involved in the down-regulation of p21 , namely a cell cycle inhibitor, and zinc and ring finger 3 (ZNRF3), the latter considered a negative regulator of the Wnt/β-catenin signaling pathway . Collectively, these data suggest that the miR-106b-3p supports the cell proliferation by dampening both apoptosis and DNA repair system while imprinting invasiveness in cancer cells via Wnt/β-catenin-mediated activation of the EMT machinery.
Based on these data and their observations in metastatic CRC patients, Liu et al.  speculated that the miR-106b-3p could be directly involved in reprogramming the metastatic behavior of cancer cells. They developed an in vitro model to assess the effect of the up-regulation of this miRNA in CRC cells and demonstrated its role in favoring both migration and invasiveness. Noteworthily, the overexpression of miR-106b-3p was associated with the activation of EMT markers, such as N-cadherin and vimentin, while in silico analysis and functional experiments revealed the deleted in liver cancer-1 (DLC-1) as a key target gene of this miRNA. Further experiments demonstrated that Exo deliver in vitro the miR-106b-3p to poorly invasive CRC cells with consequent activation of their EMT machinery and increased invasiveness by direct targeting of DLC-1. Moreover, following their stimulation with Exo enriched with miR-106b-3p, CRC cells acquired in vivo an increased metastatic potential to the lungs at least in mice models as described. The authors, therefore, concluded that miR-106b-3p exchange via Exo is implicated in the EMT of CRC cells and is thus pivotal for acquiring their metastatic capacity. Additionally, it is not excluded that exosomal transfer of miR-106b-3p to other cells may also have a role in preparing the pre-metastatic niche and regulating the organotropism of CRC cells. This topic is of particular interest since recent works have highlighted the capacity of tumor-derived Exo to selectively prepare the future metastatic bed by driving circulating cancer cells toward predicted sites. In this context, Exo are captured by specific organs depending on their integrin repertoire which drives their binding with resident target cells, thus starting to prepare the future metastatic niches only in those sites that are permissive for their anchorage and fusion [15,16].
The present study is another excellent example supporting the role of Exo in cancer development and sounds of particular interest, since it provides new data about the role of EMT in CRC metastasis. In summary, the authors have discovered a novel mechanism exploited by CRC cells to strengthen their metastatic potential through the exosomal exchange of miR-106b-3p, whose role is to down-regulate the DLC-1. This gene was originally identified in rats and was often found to be deleted in human hepatocellular carcinoma . It is generally expressed in normal tissues, while it also recurs as inactive or even lost in many human cancers, thus suggesting DLC-1 as a potential tumor suppressor gene. The DLC-1 protein is a 122 kDa multidomain protein responsible for catalyzing GTP hydrolysis and inactivation of downstream GTP Rho proteins, which was found involved in organelle development, cytoskeletal dynamics and cell movement . However, the exact molecular mechanisms responsible for DLC-1 suppression in cancer metastasis program has long remained unknown. Interestingly, the work of Zhang et al., published few weeks before Liu et al.’s  study, revealed that the DLC-1-RhoA signaling in non-small cell lung cancer (NSCLC) cells was able to inhibit the TGF-β1-induced expression of CD105, namely a strong regulator of the EMT . Collectively, these data suggest a key role of both miR-106b-3p overexpression and DLC-1 down-regulation for CRC cells in acquiring their mesenchymal phenotype (Figure 1A). On the other hand, this activity is apparently amplified by the release from CRC cells of miR-106b-3p containing Exo, which in turn reinforce and spread these signals to distant sites and maximize the lung metastatic propensity of the own cells (Figure 1B). A major limitation of the study, however, is the lack of data relative to the possible role of exosomal miR-106b-3p in driving CRC metastasis to any other organ, particularly the liver, since they used the tail vein injection for cancer cell inoculation into mice. This model classically generates only lung metastasis, due to the efficient entrapment of tumor cells into the lungs, while the optimal way to generate a systemic disease is the intracardiac injection of cancer cells, thus necessarily deserving further confirmations.
Exosomal miR-106b-3p promotes lung metastasis from CRC cells by activating the EMT machinery
Despite much evidence produced in this field, however, studies on EVs did not lead to a clear translation to clinical oncology utilization. Nevertheless, Exo still retrieve much attention by researchers and up-growing scenarios are continuously discovered concerning their contribution to tumor progression, metastasis and drug resistance [8,20]. Several studies with cancer patients including CRC investigated the possible use of circulating EVs as biological sources for ‘liquid biopsy’ potentially confirming both prognostic and predictive utility [21,22]. For instance, Liu et al. suggest the measurement of miR-106b-3p in serum Exo from metastatic CRC patients with prognostic purpose, although it is questionable how this information could be applied to drive a therapeutic decision. Moreover, a strong limitation for these purposes is the lack of standardized and universally accepted methods for the EV isolation and downstream analyses. In this context, several commercial kits are available for a rapid and easy purification but a suitable yield still requires the ultracentrifugation of biological fluids with significant effects on costs and time spent as a major restriction for the applicability of Exo as a high-throughput diagnostic tool. Further use of Exo as innovative systems for RNA- or drug-delivering is now under investigation. To this purpose, the composition of the phospholipidic bilayer of Exo protects their cargos from the phagocytosis by reticuloendothelial system contributing to diminished blood clearance, thus suggesting Exo to be suitable vectors for efficient delivery of drugs to cancer cells [23,24].
In conclusion, the paper from Liu et al.  emphasizes the role of Exo in CRC metastatic spread and strengthens the relevance of EMT in its tumor progression. The clinical applications of such interesting results, however, need further pre-clinical models for the potential use of miR-106b-3p or DLC-1 as putative therapeutic targets.
The authors declare that there are no competing interests associated with the manuscript.