Background
Ranked as the third most frequently detected cancer worldwide, colorectal cancer (CRC) also holds the position of the second most significant contributor to cancer-related mortalities. The prevalence has risen worldwide recently, particularly in low- and middle-income countries, underscoring the imperative need for enhanced approaches to early detection, prevention, and treatment of this disorder [
1,
2]. While certain strategies have shown promise in addressing CRC, the overall prognosis for this condition remains unfavorable [
3,
4]. Roughly one-third of individuals diagnosed with CRC ultimately experience the development of metastatic disease and face a notable risk of recurrence following surgical intervention [
5]. This highlights the requirement for innovative biomarkers for diagnosis and prognosis that offer sufficient sensitivity and specificity.
Investigation into microRNAs (also known as miRNAs or miRs) as promising novel biomarkers for tracking tumor evolution and advancement is gaining ground [
6]. miRNAs—short, endogenous, non-coding RNA molecules—function to down-regulate the expression of protein-coding genes by binding to the 3’-untraslanted region (3’-UTR) of messenger RNAs (mRNAs). miRNAs also exert control over the expression levels of a vast majority of protein-coding genes within human cells, thereby impacting nearly all developmental processes and diseases observed in humans. On the other side, the dysregulation of miRNAs contributes to the initiation and progression of cancer, as evidenced by the presence of aberrant miRNA expression in the majority, if not all, forms of malignancies [
7,
8]. Interestingly, profiling the expression pattern of miRNAs will be useful to effectively categorize tumors with precision and forecast the prognosis. In fact, the remarkable stability of miRNAs in biological fluids, their straightforward extraction and quantification, along with established sensitivity and specificity, render them exceptionally well-suited for biomarker research, although further studies are needed to discover the roles of miRNAs in human cancers. Within this framework, numerous investigations have emphasized the utility of serum-derived circulating miRNAs as reliable biomarkers for the early detection of CRC [
9,
10].
Recently, exosomes have gained recognition as a pivotal element in the study of cancer progression. These are distinctive extracellular vesicles, ranging in size from 30 to 100 nm, that are released from most types of cells [
11,
12]. Exosomes encapsulate proteins, lipids, DNA, mRNA, and miRNAs, constituting a significant focus of interest. Notably, a substantial portion of the RNA content within exosomes consists of miRNAs, which have significant impacts on targeted cells [
13]. The tumor microenvironment (TME) is a complex entity composed of different stromal cells [
14]. As an active stromal component of the TME, cancer-associated fibroblasts (CAFs) release exosomes that can contribute to the formation of the pre-metastatic microenvironment and support the survival and growth of tumor cells [
15]. Exosomal miRNAs released by CAFs can be transferred to recipient tumor cells and affect gene expression [
16]. However, our understanding of the precise role of miRNAs secreted by CAFs in the process of tumor progression remains elusive.
In this study, our primary aim was to investigate the expression profiles of a specific set of five serum-derived circulating miRNAs (miR-122-5p, miR-139-3p, miR-143-5p, miR-193a-5p, and miR-20a-5p) in individuals diagnosed with CRC. Additionally, we sought to explore potential associations between the expression levels of these miRNAs and clinical features in CRC patients. Furthermore, our study aimed to elucidate whether exosomal miR-20a-5p secreted from CAFs promotes CRC progression by regulating the phosphatase and tensin homolog (PTEN) gene, thereby shedding light on the functional roles of miR-20a-5p in CRC tumorigenesis. We also investigated whether CAF-derived exosomes may result in the activation of the transcription factor NF-κB in CRC cells. Our findings proposed that the increased nuclear translocation of NF-κB p65 in CAF exosome-treated CRC cells may be associated with elevated production of interlukin-6 (IL-6).
Materials and methods
Functional enrichment analyses
In order to elucidate the regulatory roles of miRNAs and identify significant molecular pathways from the Kyoto Encyclopedia of Genes and Genomes (KEGG) [
17], we utilized the DNA Intelligent Analysis (DIANA)-miRPath v3.0 software [
18]. Additionally, for the identification of validated target pathways associated with candidate miRNAs, we employed the miRTarBase v7.0 algorithm [
19]. As a standard practice, we established a threshold of
P-value < 0.05, and the false discovery rate (FDR) was determined using Fisher's exact test. To retrieve robust experimentally supported candidate miRNAs, we utilized the miRTargetLink Human algorithm [
20], which also facilitated the extraction of the interaction network.
Subject recruitment and sample processing
The study included a cohort of forty-five patients diagnosed with CRC and having primary tumors, recruited from Shariati and Rasoul Akram Hospitals in Tehran, Iran. Additionally, a group of healthy controls, consisting of 40 individuals matched in terms of age and sex, with no prior history of malignancies or chronic diseases, were also enrolled. The study received approval from the ethics committee of Islamic Azad University, and all participants provided written informed consent. Confirmation of CRC diagnoses was established through histopathological analyses conducted on surgically resected tumor samples. Tumor stages in CRC patients were determined using the staging system based on Tumor Node Metastasis (TNM) criteria outlined by the American Joint Committee on Cancer and the Union for International Cancer Control [
21]. The clinicopathological characteristics of the patients are comprehensively documented in Supplementary Table
1. To isolate serum samples, approximately 5 mL of peripheral blood was collected from each participant and subsequently subjected to centrifugation at 1,000 × g for 10 min at 4 °C. The obtained sera were then stored at -80 °C for further analysis. To isolate stromal fibroblasts, primary tissues were obtained from three CRC patients with locally advanced stage III tumors who had undergone surgery. These patients had received no prior chemo- or radio-therapy. Primary normal colon fibroblasts (NFs) were isolated from the paired adjacent non-tumor tissues.
Isolation and culture of primary fibroblasts
Fibroblasts were enzymatically isolated from both tumor and non-tumor CRC tissues using collagenase A, following the established protocol [
22]. These cells were cultured in Dulbecco’s Modified Eagle’s medium nutrient mixture F12 (DMEM/F12) supplemented with 10% fetal bovine serum (FBS). Fibroblasts within the first six passages were utilized for this study. Culture supernatants were collected when the fibroblasts reached over 70% confluency. Subsequently, the collected supernatants were centrifuged and stored at -80 °C until further use.
Isolation and characterization of exosomes
Exosomes were isolated from the conditioned media of isolated fibroblasts by differential centrifugation as we previously described [
23]. Briefly, cells were cultured in a medium containing exosome-depleted FBS at 37 °C with 5% CO2. The cell supernatants were collected and subjected to centrifugation at 300 × g for 10 min to pellet cells, followed by centrifugation at 10,000 × g for 30 min at 4 °C to further remove dead cells and debris. The resulting supernatant was then filtered through a 0.22-μm filter to exclude microvesicles larger than 200 nm. The filtered supernatant underwent an additional ultracentrifugation step at 100,000 × g for 70 min at 4 °C. To ensure the removal of any remaining protein contamination, the initial exosome pellet was resuspended in PBS and subjected to another round of ultracentrifugation. The resulting exosome-enriched pellets were reconstituted in a small volume of PBS or lysis buffer as required. The intact exosomes suspended in PBS were then stored at -80 °C for future experiments.
The morphology of purified exosomes was observed using a scanning electron microscopy (SEM). To accomplish this, a portion of the exosome suspension was fixed with 2.5% glutaraldehyde and applied onto 200-mesh Formvar-coated grids. Subsequently, the grids were stained using uranyl acetate and lead citrate before being examined using a Digital SEM, KYKYEM3200, China. The size range of the exosomes was determined using a Malvern Zetasizer Nano ZS90 (Malvern, UK), following the manufacturer's instructions. The protein concentration in exosome preparations was quantified using a BCA protein assay kit (Pierce Chemical Co., Rockford IL). The presence of the exosome-specific protein CD9 was detected in exosome preparations using a western blotting assay.
Cellular internalization of purified exosomes
Fibroblast-derived exosomes were labeled with PKH26 (Sigma Aldrich, USA) following established protocols. Subsequently, the labeled exosomes were introduced to a subconfluent layer of CRC cells and incubated at 37 °C for 12 h. After incubation, the cells were rinsed thrice with PBS and fixed with 4% paraformaldehyde. For nuclear staining, 4′-6-diamidino-2-phenylindole (DAPI) was used. The cellular internalization of labeled exosomes was visualized using a spectral confocal microscope (Nikon Eclipse TiE, Japan).
Transfection assays
CRC cells were transfected with either miR-20a mimic or a negative control (25 nM), as well as miR-20a-5p inhibitor or inhibitor negative control (50 nM), using Lipofectamine® 2000 (Thermo Fisher Scientific, USA). The miR-20a-5p mimic, miR-20a-5p inhibitor and their corresponding negative controls were obtained from GenePharma Co., Ltd. and the sequences were as follows: miR-20a-5p mimic, sense: 5′-UAAAGUGCUUAUAGUGCAGGUAG-3′; miR-20a-5p mimic, antisense: 5′-ACCUGCACUAUAAGCACUUUAUU-3′; miRNA negative control: 5′-UUGUACUACACAAAAGUACUG-3′; miR-20a-5p inhibitor: 5′-CUACCUGCACUAUAAGCACUUUA-3′; inhibitor negative control: 5′-UGACUGUACUGAACUCGACUG-3′. Additionally, two independent siRNAs targeting PTEN or a negative control siRNA (si-NC; obtained from GenePharma) were introduced into the cells using Lipofectamine®2000 (Thermo Fisher Scientific, USA). For the luciferase reporter assay, CRC cells were co-transfected with psi-CHECK2 luciferase reporter vector containing the 3′-UTR of PTEN and miR-20a-5p mimic or the negative control. After 48 h, the relative dual-luciferase activity was measured using the dual luciferase reporter assay system (Promega, USA).
RNA extraction and real-time quantitative PCR
Total RNA extraction from cells, patient serum, and isolated exosomes was performed using the TRIzol reagent (Invitrogen). Subsequently, 1 μg of total RNA was reverse transcribed into cDNA using the PrimeScript 1st Strand cDNA Synthesis kit (TAKARA, Japan). For miRNA quantification, the extracted total RNA underwent polyadenylation with polyA polymerase enzyme (NEB). Then, cDNA synthesis was carried out using an anchored oligo (dT) primer, following a previously described method [
24]. Reverse transcription quantitative PCR (RT-qPCR) was conducted on an ABI Step One Detection System (Applied Biosystems, USA). Relative expression levels were normalized to U6 small nuclear RNA (snRNA) and glyceraldehyde-3-phosphate dehydrogenase (GAPDH). Additionally, cel-miR-39 served as an internal reference for analyzing exosomal miRNA expression. Data analysis was performed using the 2
−ΔCt method, where ΔCt is calculated as Ct (Target)—Ct (Reference), and fold changes were determined using the 2
−ΔΔCt method [
25].
NF-κB activation assay
To evaluate NF-κB activity, nuclear and cytosolic fractions were separated utilizing a NF-κB Activation Assay Kit (FIVEphoton Biochemicals, San Diego, CA, USA) following the manufacturer’s guidelines. Subsequently, protein concentrations in the lysates were quantified using the Bradford assay, and the NF-κB p65 protein level in the nuclear and cytoplasmic preparations was detected by western blotting.
Western blot assay
Cellular or exosomal total proteins were extracted using radioimmunoprecipitation assay (RIPA) lysis buffer (Sigma-Aldrich, USA), separated on 10–12% SDS–polyacrylamide gels (SDS-PAGE) and transmitted to a polyvinylidene difluoride (PVDF) membrane. To prevent non-specific binding, a solution of 5% bovine serum albumin (Merck) in TBST (10 mM Tris-buffered saline with 0.05% Tween 20) was employed. The blots were then probed with specific primary antibodies, appropriately diluted in TBST (1:1000). Following rinsing, the blots were treated with horseradish peroxidase (HRP)-conjugated secondary antibodies and then subjected to chemiluminescence (ECL, Amersham, Buckinghamshire, UK) for visualization. β-actin was utilized as a loading control.
Enzyme-linked immunosorbent assay
The IL-6 levels in the supernatant of CRC cells were quantitatively measured after incubating them with 100 µg/mL exosomes derived from CAFs or NFs for 48 h, using enzyme-linked immunosorbent assay (ELISA). The cell culture supernatants were used undiluted for the assay.
Proliferation assay
SW480 CRC cells were seeded at a density of 2 × 104 cells/mL and then incubated with different concentrations of CAF-derived exosomes (25, 50, 100 µg/mL), NF-derived exosomes (100 µg/mL) or transfected with a miR-20a inhibitor. After 48 h of incubation or transfection, the viability of the cells was evaluated in triplicate using the trypan blue exclusion assay, and the results were compared to corresponding control cells.
Transwell migration assay
SW480 CRC cells were suspended in serum-starved media and subsequently seeded into transwell chambers with 8 μm pore size (BD Biosciences). In the lower chamber, a medium containing 10% FBS served as a chemoattractant. Following a 24-h incubation period, the cells that migrated through the membrane and adhered to its lower surface were stained with 1% crystal violet (ThermoFisher Scientific) dissolved in methanol and then quantified using a light microscope.
Statistical analysis
The data are presented as means ± standard deviation (SD) and were subjected to analysis using two-tailed Student's t-tests or one-way ANOVAs, as deemed appropriate for the respective comparisons. A significance threshold of p-value < 0.05 was utilized for this study. The diagnostic accuracy of the candidate serum-derived miRNAs was evaluated using a receiver operating characteristic (ROC) curve test, which involved calculating the area under the curve (AUC) along with 95% confidence intervals. All statistical analyses as well as visualization were carried out using GraphPad Prism version 9.0 (GraphPad Software, Inc., La Jolla, CA, USA).
Discussion
CRC stands out as one of the most perilous malignancies, with its worldwide occurrence and fatality rates witnessing a substantial rise in recent decades. Despite extensive documentation of genetic and epigenetic alterations in CRC, the intricate molecular mechanisms underpinning the development of this disease still require further investigation. Additionally, the discovery of novel therapeutic targets is of utmost importance for enhancing the treatment of CRC. There is also a pressing need for non-invasive diagnostic tools to reduce the mortality rates associated with CRC [
33,
34].
In recent times, the field of exosome biology and its clinical implications in oncology have garnered growing interest. Among the various types of cargo transported by exosomes, miRNAs have come under the spotlight [
31,
35,
36]. Here we aimed to identify robust biomarkers for CRC and subsequently elucidate the mechanisms by which the most significant biomarker influences tumorigenesis. Our findings established a signature of five-circulating miRNAs (miR-20a-5p, miR-122-5p, miR-139-3p, miR-143-5p, and miR-193a-5p) as a novel non-invasive diagnostic biomarker for patients with CRC (Fig.
1). Consistent with our results, Selven et al. also demonstrated that high expression levels of miR-20a-5p are positively correlated with a favorable prognosis in advanced stages of CRC [
37]. Additionally, miR-122-5p has been identified to promote CRC progression [
38]. Therefore, these studies provide evidence of the oncogenic nature of these miRNAs in CRC. Our findings indicated that miR-20a-5p and miR-122-5p displayed associations with advanced stages of tumorigenesis, while miR-139-3p, miR-143-5p, and miR-193a-5p showed positive correlations with the earlier stages of tumorigenesis, specifically TNM stages I and II (Fig.
1). These findings collectively highlight the potential of these miRNAs as valuable biomarkers for CRC diagnosis and prognosis. However, there remains an unanswered question regarding the specific roles that these candidate miRNAs play during CRC tumorigenesis.
CAF, as a prominent component of the TME, plays a crucial role in promoting tumor progression by releasing an array of soluble factors or structurally remodeling the extracellular matrix [
39]. Additionally, these stromal cells can induce a more malignant activity of tumor cells through the release of exosomes that eventually promote tumor progression [
40]. The impact of exosomes on CRC progression has been extensively documented. Notably, CAF-derived exosomes can enhance metastasis and confer resistance to chemotherapy in CRC [
41,
42]. CAFs and NFs exhibit distinct variations in terms of their morphology and gene expression [
42]. CAFs stand out as the predominant stromal components and hold crucial roles in shaping the TME and impacting the behavior of tumor cells, primarily through the release of proteolytic enzymes, growth factors, and cytokines [
43]. While CAFs have historically been categorized as α-SMA + myofibroblasts [
44], various other markers, e.g., FAP, fibroblast-specific protein 1 (FSP1), and platelet-derived growth factor receptor α (PDGFRα) have been documented as well [
45]. Our data consistently revealed an up-regulation in the expression levels of
α-SMA and
FAP within the isolated CAFs. Conversely, NFs exhibited elevated expression of Vimentin, a marker highly expressed in fibroblasts (Fig.
2A). At the next step, we showed that the expression analysis of candidate miRNAs in both CAFs and NFs exhibited a comparable pattern in patients’ samples. Specifically, miR-20a-5p and miR-122-5p were found to be elevated in CAFs, while miR-139-3p, miR-143-5p, and miR-193a-5p displayed increased expression levels in NFs (Fig.
2G). Subsequent analysis revealed a consistent expression pattern in exosomes derived from both CAFs and NFs. However, concerning miR-20a-5p, we identified notably higher expression levels in exosomes derived from CAFs compared to the other candidates (Fig.
2H). Out of all the candidate miRNAs, we concentrated our attention on miR-20a-5p for several key reasons: firstly, it displayed characteristics consistent with oncomiRs; secondly, it exhibited the most robust correlation with cancer-related features identified through functional enrichment analyses; and thirdly, its expression level was significantly higher than that of any other candidate miRNAs in both CAFs and the exosomes derived from CAFs.
As cancer progresses, tumor cells undergo processes such as detachment from the basement membrane, invasion into nearby tissues, and the establishment of colonies in distant locations. EMT plays a crucial role in regulating these processes, particularly in metastasis [
46]. In the context of CRC, miR-20a-5p has been identified to induce EMT, indicating its oncogenic role. Interestingly, miR-20a-5p's function seems to vary based on tissue context; for example, in endometrial cancer, miR-20a-5p inhibits EMT by targeting STAT3 [
47]. However, in CRC, miR-20a-5p promotes EMT, suggesting its tissue-specific function. Here, we showed a correlation between elevated levels of miR-20a-5p and increased aggressiveness of CRC cells. PTEN, a well-known tumor suppressor gene frequently disrupted in cancers, including CRC, is often silenced at the transcriptional level [
48]. Various studies have highlighted the regulation of PTEN by miRNAs in different cancer types [
49]. Given that miR-20a directly targets PTEN (Fig.
3B, C), we speculated whether the exosomal transfer of miR-20a secreted from CAFs might facilitate CRC cell progression partly through targeting PTEN. We identified a reduction in
PTEN expression in cells transfected with miR-20a-5p mimic (Fig.
3D). Consistently, Fang et al. [
50] identified that miR-20a-5p contributes to hepatic glycogen synthesis through targeting p63 to modulate p53 and
PTEN expression. Notably, the miR-20a-5p/PTEN axis was also identified by Zhang et al., who proposed that the up-regulation of miR-20a-5p enhances hepatocyte proliferation by targeting the PTEN/AKT signaling pathway [
51], aligning with our findings. Our study revealed a close association between PTEN inhibition and increased CRC cell proliferation and migration, a phenomenon partially attributable to the actions of miR-20a-5p (Fig.
3F-H). Notably, Coronel-Hernández et al. [
52] made a similar observation regarding miR-26a's ability to reduce PTEN levels in CRC cells, a process that correlates positively with heightened rates of cell proliferation and migration. Furthermore, our study revealed that inhibiting PTEN led to increased expression levels of EMT markers in CRC cells (Fig.
3I-K). Consistently, our data revealed that miR-20a-5p can induce EMT markers in CRC cells, partly through PTEN inhibition, and enhance the expression of E-Cadherin—an epithelial marker for EMT. These findings are particularly intriguing and add depth to our understanding of the role of miR-20a-5p in CRC, suggesting a potential mechanism for its oncogenic function, ultimately driving elevated cell proliferation and migration in CRC.
Exosomal transfer of miRNAs is instrumental in modulating the TME. Our research focused on understanding the intercellular communication mediated by exosomes within the context of the CRC microenvironment. Our data demonstrated that miR-20a was enriched in CAF-derived exosomes and was capable of being transferred as cargo to CRC cells (Fig.
4A). Importantly, the exosomal transfer of miR-20a-5p secreted from CAFs was associated with enhanced proliferation and migration rates of CRC cells (Fig.
4C-E). Thus, it can be inferred that exosomal miR-20a released by CAFs potentially enhances the aggressive behavior of CRC cells. Our findings further indicated that exosomes originating from CAFs led to heightened levels of the NF-kB p65 transcription factor. Consequently, NF-κB p65 exhibited increased activity within the nuclei of CRC cells treated with exosomes derived from CAFs (Fig.
5A, B). IL-6 has been identified as an important tumor-promoting cytokine that enforces proliferation and anti-apoptotic effects in tumor cells [
53]. The expression of IL-6 is markedly elevated in CRC and is closely associated with CRC development [
54]. Given that the hallmark of vascular NF-κB activation involves IL-6 production, and it has been demonstrated that the promoter region of the IL-6 gene contains a putative NF-κ B-binding site [
55], we investigated IL-6 production in CAF Exo-treated CRC cells. As illustrated in Fig.
5, CRC cells stimulated with CAF-derived exosomes for 48 h exhibited significantly higher IL-6 production compared to CRC cells treated with control PBS or normal fibroblast-derived exosomes (NF Exo). Collectively, our study may provide new insights into the mechanisms through which CAF-derived exosomes exert their oncogenic functions, specifically though the NF-kB p65/IL-6 axis.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.