Abstract
Gastric cancer is a major cause of cancer mortality worldwide, with a low survival rate for patients with advanced forms of the disease. Over the recent decades, the investigation of the pathophysiological mechanisms of tumourigenesis has opened promising avenues to understand some of the complexities of cancer treatment. However, tumour regeneration and metastasis impose great difficulty for gastric cancer cure. In recent years, cancer stem cells – a small subset of tumour cells in many cancers – have become a major focus of cancer research. Cancer stem cells are capable of self-renewal and are known to be responsible for tumour initiation, metastasis, therapy resistance and cancer recurrence. Recent studies have revealed the key role of microRNAs – small noncoding RNAs regulating gene expression – in these processes. MicroRNAs play crucial roles in the regulation of a wide range of biological processes in a post-transcriptional manner, though their expression is dysregulated in most malignancies, including gastric cancer. In this article, we review the consequences of aberrant expression of microRNA-34 in cancer and cancer stem cells, with a specific focus on the miR-34 dysregulation in gastric cancer and gastric cancer stem cells. We address the critical effects of the aberrant expression of miR-34 and its target genes in maintaining cancer stem cell properties. Information collection and discussion about the advancements in gastric cancer stem cells and microRNAs can be useful for providing novel insights into patient treatment.
Introduction
Gastric cancer (GC) is a life-threatening malignant tumour in humans. It has the fourth highest incidence among all malignant tumours. GC is also the second leading cause of cancer-related mortality worldwide,1,2 with low survival and high recurrence rates for patients with advanced forms of the disease. 3 Tumour regeneration and metastasis impose great difficulty in the prevention and treatment of GC. 4 Currently, the only hope for cure rests on the removal of the malignant tissue – either endoscopically or by surgical resection. For advanced forms of the disease, the treatment consists of a combination of surgery, chemotherapy and radiation. Overall, the results of current therapy for advanced disease are poor, 2 and the 5-year survival rate is still less than 40%.4,5 Thus, new and effective strategies for cancer therapy should be established through further research, by providing precise knowledge of key molecular mechanisms by which a cancer cell is deregulated.
Cancer stem cells (CSCs) – a small subset of cells within the tumour bulk – show high capacities for sphere-forming, self-renewal, migration, invasion and resistance to chemo- and radiotherapy. CSCs contribute to the initiation, progression and recurrence of different types of tumours.6,7 On one hand, GC is prone to invasion, metastasis and multidrug resistance.8–11 On the other hand, metastasis, invasion and therapy resistance are found to be the consequences of existence of CSC in different types of tumours.6,12,13 Previous studies have also revealed that microRNAs (miRNAs or miRs) can regulate a variety of critical genes involved in invasion, metastasis and therapy resistance.8,14–16
miRNAs are small, noncoding RNAs that are ~22 nucleotides in length. They can regulate gene expression in a post-transcriptional manner by translational inhibition or cleavage of messenger RNAs (mRNAs), depending on the degree of complementarity to their target sequence.17–19 The discovery of miRNAs in 1993 20 is a landmark milestone in the field of molecular biology. miRNAs can regulate the expression of hundreds of their target genes, thereby playing a crucial role in a wide range of biological processes, such as apoptosis, 21 differentiation, 22 proliferation 23 and immune response.24,25 Most importantly, recent evidence indicates that miRNAs may function as tumour suppressors or oncogenes, and dysregulation in miRNA expression is associated with tumourigenesis and cancer progression through impact on a variety of tumour cell functions, including proliferation, migration, invasion, differentiation and tumour recurrence.26–29 Studies suggest that miRNAs play important roles in CSC proliferation and differentiation and tumour formation. 30 Therefore, exploration of dysregulated miRNAs and their functions in various CSCs would be beneficial for developing drugs or novel therapeutic methods to target and eliminate CSCs.
Several studies have investigated the aberrant expression of the microRNA-34 (miR-34) family and its role in cancer and CSCs, in order to provide potential new strategies for cancer treatment. Here, we review these studies, with a specific focus on the miR-34a dysregulation in GC and gastric cancer stem cells (GCSCs). We address the critical role of aberrant expression of miR-34a in maintaining CSC properties. Furthermore, the biological and molecular mechanism by which miRNAs regulate CSCs is also addressed. Discussion of the advancements in miRNA functions and abnormalities caused by dysregulated expression of miRNA in GCSCs will be useful for providing a novel insight into the development of effective miRNA-based therapies in the clinic and in GC treatment.
General properties of CSCs
Emerging evidence indicates that malignant tumours consist of a small subset of distinct cancer cells, defined as CSCs (typically less than 5% of total cancer cells, based on cell surface marker expression). 31 CSCs – also known as cancer-initiating cells – were initially isolated from patients with acute myeloid leukaemia. 32 At present, the existence of CSCs has been proved in many solid tumour types, such as breast cancer,33,34 glioblastoma,35,36 colon cancer, 37 pancreatic cancer, 38 prostate cancer, 39 ovarian cancer, 40 liver cancer, 41 cholangiocarcinoma 42 and GC.6,43 The spheroid body formation assay, in which the cells are cultured in serum-free low attachment conditions supplemented with basic fibroblast growth factor (FGF) and epidermal growth factor (EGF), is a practical method for the enrichment of CSC from tumours or cancer cells.44–46
CSCs are capable of self-renewal, proliferation and multi-lineage differentiation. They produce different phenotypes of non-tumourigenic tumour cells – constantly enlarging the tumour mass.6,47 CSCs can mediate tumourigenicity, metastasis and resistance to chemo- and radiotherapy. 48 CSCs are also known as a slow-cycling or quiescent subpopulation. These properties are thought to be a major reason that current anti-cancer therapies are more successful in targeting proliferating non-CSCs than CSCs 49 ; thereby, the surviving CSCs will lead to recurrence of tumour, even many years after treatment has been ended.7,50 To be most effective, cancer therapy must be directed against both the resting CSCs and the proliferating cancer cells. 47 However, it seems that killing differentiated non-CSCs in the tumour bulk is not as important as killing CSCs. This is because after removal of CSCs, we can assume that a tumour bulk is a benign tumour without metastasis or tumour initiation ability. Also, in comparison with CSCs, differentiated non-CSCs may have a limited lifetime.
Previous studies have confirmed that GCSCs play an important role in the tumourigenicity of GC, and are the major cause of invasion, metastasis and drug resistance observed in this type of cancer.6,12 This suggests that GCSCs should be the central therapeutic target for GC treatment. Thus, isolated and well-characterized GCSCs are the ideal model of therapeutic studies for cancer research. In the following section, we describe the most commonly used cell surface markers for the enrichment of GCSCs from GC tissues or cell lines.
It is important to note that the expression of CSC markers is highly associated with tumourigenesis, degree of malignancy and tumour grading. 51 Therefore, CSC markers can be ideal targets for cancer therapy.
GCSC surface markers
GCSCs were first isolated and identified in 2009. Takaishi et al. 43 isolated CD44(+) cells from GC cell lines. It was found that these CD44(+) cells were able to form spherical colonies in serum-free culture medium and develop tumours in severe combined immunodeficient (SCID) mice, while the CD44(−) populations had significantly reduced the tumourigenic ability in vitro and in vivo. Han et al. isolated GCSCs from human GC tissues by using two cell surface markers – the epithelial cell adhesion molecule (EpCAM) and CD44. In vitro experiments showed that these cells grew into cancer spheres in serum-free medium. They were found to have the ability to self-regenerate, potential for multiple differentiation and greater anti-cancer drugs resistance compared with other subpopulation cells. Furthermore, it was confirmed that these cells formed solid tumours in nude mice. 6 Thereafter, use of EpCAM and CD44 surface markers in GCSCs isolation was also reported in other studies.3,45 In the study by Chen et al., CSCs were isolated from tumour tissues and the peripheral blood of human gastric adenocarcinoma. These cells, which carried CD44 and CD54 surface markers, generated tumour when injected into immunodeficient mice, differentiated into gastric epithelial cells in vitro and self-renewed in vitro and in vivo. 52
CD133 (prominin-1) – originally classified as a marker of primitive haematopoietic and neural stem cells 53 – has been identified as a CSC marker in various types of cancers, including brain, 54 prostate 55 and colon cancer.37,56 Several studies have confirmed the expression of CD133 marker in GC cells42,57,58 and have investigated the clinical role of its expression in patients with GC. Data show that CD133(+) GC cells had high potential for malignancy grades. Patients with CD133(+) had stronger chemotherapy resistance and recurrence. In addition, CD133 expression was associated with poor prognosis in GC patients.57,59,60 Although CD133 expression has been widely investigated as a candidate for human GCSC marker in cancerous tissues or cell lines3,43,52,61, to our knowledge, there is no well-documented finding confirming CD133 as a valuable GCSC marker. Several studies that increase doubt about the consideration of CD133 as a GCSC marker are referred to in.3,62,63
CD44 has been recognized as a CSC marker in many types of cancer, including colon,64,65 breast, 33 pancreas, 38 prostate, 39 cholangiocarcinoma, 42 ovary, 66 liver, 41 small intestine, 67 head and neck68,69 and stomach cancer. 6 CD44 is the earliest marker used for the isolation of GCSCs. 43 Our review of the existing literature shows that almost all published studies related to GCSC research6,12,43,45,52,70,71 used CD44 – either individually or in combination with other marker(s) – for the isolation, identification and characterization of these cells. Thus, we further discuss the function of CD44 in GC and the association between miR-34 and CD44 expression.
The cell surface molecule CD44 is a transmembrane glycoprotein that plays a role in the facilitation of cell–cell and cell–matrix interactions through its affinity to hyaluronic acid. It is involved in cell adhesion and the assembly of growth factors on the cell surface. CD44 is encoded by a single gene containing 20 exons. The standard form (CD44s) consists of exons 1–5 and 15–20 (10 exons). The variable exons are identified as v1–v10. The different combinations of the 10 variant exons, through alternative splicing, generate multiple CD44 variants (CD44v).72,73
The standard isoform CD44s is expressed in a variety of normal tissues. It has been reported that CD44 isoforms containing variant exon 6 (CD44v6) is correlated with tumour invasion, progression and metastasis in various kinds of malignancies, including GC.73–75 Lau et al. 3 reported that CD44v8–10 is the major CD44 variant in GC and verified its role as a GCSC marker. The invasive and migratory capacities of the CD44(+) GC stem-like cells were much higher compared with those of the CD44(−) cells. 12 The CD44(+) GC cells showed increased resistance for chemotherapy- or radiation-induced cell death. 43 Furthermore, GC cells with removed CD44 had significantly reduced tumourigenic ability in vitro and in immunocompromised mice.3,43 In the following sections, an attempt is made to review the relationship of GCSCs and miR-34 expression on the basis of existing literature.
miRNA-34 family members
The miRNA-34 family (miRNA-34(s)) is highly conserved in the evolutionary context. In vertebrates, there are three miR-34 members – miR-34a, miR-34b and miR-34c – which are generated from two distinct genomic loci. miRNA-34b and miRNA-34c are generated by processing a bicistronic transcript from chromosome 11q23, while miRNA-34a is encoded by its own transcript, located within the chromosome 1p36. Assessment of the miRNA-34 family expression in normal mouse tissues showed that miRNA-34a is ubiquitously expressed with highest levels in brain, whereas miRNA-34b/c are predominantly expressed in the lung, with low expression in the brain and scarce or undetectable expression in other tissues. 76 The nucleotide sequences of the three miRNA-34 family members are shown in Table 1.
Sequences of the mature miRNA-34 family members retrieved from miRBase (www.mirbase.org).
miRNA-34: microRNA-34; miRNA: microRNA.
Over 50% of human cancers have mutant p53 – a sequence-specific transcription factor. 47 Although most miR-34 family members are known to be directly regulated by p53,76–79 some studies have reported that miR-34a is regulated independent of the p53 pathway.80,81 More than 60% of human protein-coding genes are putative targets of miRNA. 82 This point emphasizes their crucial roles in diverse biological processes. As a family of highly conserved miRNAs, miR-34(s) play important roles in the cell. miR-34 is involved in cell cycle arrest, either transient or permanent (senescence), and accumulates the cells in the G1 phase; thus, it regulates cell cycle progression.47,76,78,83,84 In addition, miR-34 induces apoptosis.10,77,79,85
The functions mentioned above, namely, cell cycle arrest, senescence and apoptosis, are tumour-suppressive mechanisms. Therefore, the miRNA-34(s) inactivation may contribute to tumourigenesis. In support of this statement, studies have revealed that miR-34(s) function as a potential tumour suppressor and their expression is down-regulated or lost in various types of cancers.86–91 The down-regulation of miR-34(s) gene expression may occur due to several reasons, including mutations that inactivate p53 in tumour cells,47,92 mutation of the p53-binding site in genes encoding miR-34(s)77,78 and inactivation of miR-34(s) by aberrant CpG methylation of promoter.93–96 Furthermore, the deletion of genes that include miR-34(s) (in chromosomes 1p36 and 11q) is often observed in various types of cancers and is another mechanism that leads to reduced miR-34 expression in cancer.97,98
Re-expression of miR-34(s) in cancer cells was found to induce apoptosis, cell cycle arrest or senescence.31,94,99 In addition, previous data demonstrated that overexpression of miR-34 (via transfection into tumour cells) inhibited the growth of cultured cancer cells 100 and metastasis91,101 and increased radiosensitivity at low doses of radiation.102,103 Indeed, we can conclude that the restoration of miR-34 to cells – through which it is down-regulated – inhibits tumour malignancy. Wiggins et al. showed that the uptake of miRNA-34a mimics has no effect in normal cells. This may suggest that the targeted delivery of miRNA to tumour tissues would not be necessary, 100 obviating the obstacles associated with the use of some delivery vehicles, including virus-based delivery carriers. 104
The down-regulated expression of miR-34(s) was not reported in all human cancer samples evaluated in each study. For example, Bommer et al. 76 demonstrated that the expression of two miRNA-34(s) dramatically reduced in six of 14 (43%) non–small cell lung cancers (NSCLCs). In the report by Tazawa et al., 97 nine of 25 human colon cancers (36%) showed a decrease in expression of miR-34a compared with the counterpart normal tissues. In the study by Wiggins et al., 20 of 32 NSCLC tumour samples (63%) showed reduced miR-34a expression in comparison with the normal adjacent lung samples. Significantly, the results of this study showed that exogenous miR-34a not only inhibits cell lines with reduced endogenous miR-34a expression but also inhibits cell lines with normal miR-34a expression levels. 100 In the study by Mudduluru et al., 24 of 44 NSCLC patients (54.5%) showed a significant down-regulated expression of miR-34a, while 12 patients (27.3%) had a significant up-regulated expression of miR-34a. Interestingly, NSCLC patients with miR-34a up-regulation showed a positive association towards longer survival. 105
miRNA-34 in GC
Dysregulated expression of miR-34(s) was also reported in GC. Ji et al. demonstrated that in p53-deficient human GC cells (Kato III), the restoration of functional miR-34 inhibited cell growth and induced chemosensitization and apoptosis. This indicates that miR-34 may restore p53 function. 47 Cao et al. found that miR-34a expression levels in the GC cell lines (SGC-7901, MGC80-3, NCI-N87 and HGC-27) were significantly lower than the levels in the GES-1 normal gastric epithelial cell line. The results of this study demonstrate that after overexpression of miR-34a in HGC-27 cells, cellular viability, the proliferation index and invasion were significantly decreased, while cellular apoptosis was significantly increased. 10 Zhang and colleagues investigated the expression level of miR-34a in 137 GC patients through quantitative real-time polymerase chain reaction (qRT-PCR). Results of the study revealed that 64 patients had high miR-34a levels and 73 patients had low miR-34a levels (the median miR-34a expression level in tumour tissue was 2.44 (range: 0.12–29.83)). They also found that low level of miR-34a expression in GC patients was associated with lymph node involvement, advanced tumour/node/metastasis (TNM) stage, poor tumour differentiation, high tumour recurrence rate and poor 5-year survival rate. 106
Taken together, the published literature established that miR-34 is a new tumour suppressor whose expression is decreased in a large variety of cancers (such as gastric, colon, lung, breast, pancreas, glioblastoma, head and neck squamous cell and osteosarcoma), providing impetus to the functional restoration of miR-34 as a novel therapy for cancers. However, our literature review on the altered expression of miR-34 in GC showed that some studies have reported the up-regulation of miR-34 in GC (listed in Table 2). Also, miR-34 was not determined as a deregulated miRNA in several studies that aimed to identify aberrantly expressed miRNAs in GC.114–118 The lack of agreement among the different studies profiling miRNA expression may be due to the use of different analytical approaches, different platforms of (miRNA) microarray, different processing methods, small number of samples and cancer samples with different degrees of differentiation or with different clinicopathological stages of cancer, and may probably be related to the differences among study populations. In support of the last mentioned probability, few studies have reported that differential expression patterns of miRNAs in GC can be influenced by the ethnic background and socio-cultural status in which the cancers arise.109,115 The inconsistencies in these reports require further studies to draw a definite conclusion that is applicable in subsequent decision-making for cancer treatment.
Reports of hsa-miR-34a expression in GC detected by miRNA microarray and/or qRT-PCR analysis.
GC: gastric cancer; miRNA: microRNA; qRT-PCR: quantitative real-time polymerase chain reaction; ND: not determined; NA: not available.
As listed in Table 2, studies that reported the up-regulated expression of miR-34a in GC commonly used the microarray technique for analysis. Although microarray appeared as one of the most popular technologies for miRNA expression profiling, the results should be verified by qRT-PCR due to the high rate of false-positive up-regulated results. qRT-PCR showed higher sensitivity, specificity and reproducibility compared with the microarray. 119 However, some studies used different reagents and analytical approaches to improve the sensitivity, specificity and reliability of miRNA microarray experiments.120–122
miRNA-34 in GCSC
In stem cells, miRNAs can promote either self-renewal or differentiation; therefore, they are able to determine the fate of stem cells. 45 However, the exact roles of different miRNAs in the regulation of CSCs have not been fully elucidated.
Sorted CD44(+) GC stem-like cells and CD44(−) cells from the same cell line (MKN-45 or KATO III) showed different patterns of miRNA expression. 91 Liu et al. 123 demonstrated that non-adherent spheroid body–forming cells from the GC cell line MKN-45 in defined serum-free medium possessed GCSC properties and overexpressed CSC-related genes (Oct4, Sox2, Nanog and CD44), compared with the parental cells. In a subsequent work, they used a miRNA microarray to evaluate the miRNA expression profiles in the spheroid body–forming cells and the parental cells (MKN-45 cell line), and found that 182 miRNAs, each with more than twofold change, were differentially expressed between the spheroid body–forming cells and parental cells. Most of them (173), containing hsa-34a-5p, showed very low-level expression in the spheroid body–forming cells, which were increased upon differentiation. Inconsistent with the results from the miRNA microarrays, the qRT-PCR results showed that miR-34a-5p was up-regulated in the spheroid body–forming cells, compared with the parental GC cells (the relative ratio of 2.57 ± 0.07). 46 In this study, no comparison was performed between miR-34 expression level in spheroid body–forming cells and normal gastric cells.
miR-34 is an important regulator of both normal stem cells and CSCs. 99 In the study by Ji et al., the restoration of miR-34 in GC cells (Kato III cell line) inhibited the growth and formation of the tumoursphere, which is reported to be correlated to the self-renewal of CSCs. They found that miR-34-mediated suppression of self-renewal was related to the direct modulation of downstream targets Bcl-2, Notch and HMGA2, indicating that miR-34 may be involved in GCSC self-renewal/differentiation decision-making. 47 Prostate CSCs are enriched in the CD44(+) cell population.39,55 Liu et al. showed that miR-34a expression was down-regulated in CD44(+) prostate cancer cells. They validated CD44 as a direct and functional target of miR-34a, which is negatively regulated by this miRNA. Systemically delivered miR-34a inhibited prostate cancer metastasis and prolonged the survival of tumour-bearing mice. 99 In the next section, we address several miR-34a target genes implicated in carcinogenesis and cancer development. These also include genes involved in the self-renewal and survival of CSCs.
Target genes of miR-34a
The identification of target mRNAs is important for understanding the miRNA functions as well as the molecular mechanisms by which the regulation of different processes is mediated by miRNAs. Recognition of target mRNAs by miRNAs depends mainly on the 2–8-base-seed sequence of miRNA 5′ ends. miRNAs bind through this sequence to the complementary sites in 3′-untranslated region (3′-UTR) of the mRNA, leading to its translational repression or degradation, thereby negatively regulating the protein expression. The interaction with 3′-UTR is often imperfectly complementary, meaning that each miRNA could regulate the expression of many mRNAs.10,30,115,124 Although the 3′-UTRs of mRNAs are the most frequent regions of miRNA targeting, the open reading frames (ORFs) and 5′-UTR can also be targeted. 125
miRNA binding to the 3′-UTR of target genes is the basis for most target-site-prediction algorithms. miRanda, TargetScan and PicTar are examples of algorithms commonly used to predict human miRNA target genes.26,46,126 These algorithms provide an ability to search for a potential binding site for miRNAs within the 3′-UTR of target mRNA. 127 Studies show that there are more than hundreds of predicted target genes for each miRNA. Also, several genes were identified as potential targets of two or more miRNAs. 46 However, bioinformatic analysis and prediction programmes are the primary methods and are known to have high false-positive rates; their predictions are not in agreement.26,46 Thus, the results of the prediction algorithms need to be further investigated in order to determine whether the predicted target genes are the authentic targets of the miRNA.
To test whether a miRNA directly recognizes its predicted target gene transcripts through the presumed binding site, many studies performed luciferase reporter assay.47,76,128 In this analysis, a reporter vector is constructed, consisting of the luciferase coding sequence followed by the 3′-UTR of the target gene. The ectopic expression of a miRNA mimic significantly reduces the luciferase activity of a construct containing the 3′-UTR of target gene in the transfected cells. Indeed, reduction of the luciferase activity shows that the transfected miRNAs are functional and confirms that the gene is a direct target of studied miRNA. qRT-PCR and western blot analysis are often used to detect the variations of target gene expression and the protein level after ectopic expression of miRNA.
mRNAs targeted by miR-34a are highly enriched by transcripts that control the cell cycle, apoptosis, DNA repair and angiogenesis.77,84,94,126 Most of them are oncogenes that are frequently up-regulated in human tumours. 31 miR-34a has the ability to down-regulate these oncogenes.
Table 3 shows several main target genes of miR-34a that play a pivotal role in oncogenesis. In published studies, these targets were first predicted by various miRNA target prediction algorithms. In silico analysis identified miR-34a binding site(s) in their 3′-UTRs. After this, experimental validation confirmed these genes as miR-34a direct targets. Among these target genes, several genes are associated with pivotal signalling pathways of the stem cell, and high levels of their expression can be involved in the self-renewal and survival of CSCs, such as Notch, Wnt/β-catenin, Bcl-2, Myc, JAG1, SIRT1 and LEF1.47,87,141,146,148–150
Validated direct target genes of miR-34a implicated in carcinogenesis based on published experimental findings.
BCL2: B-cell CLL/lymphoma 2; SIRT1: silent information regulator 1; CDK: cyclin-dependent kinase; MET: hepatocyte growth factor receptor; PDGFR: platelet-derived growth factor receptor; mTORC2: mammalian target of rapamycin complex 2; LRP: low-density lipoprotein (LDL) receptor–related protein; TCF/LEF: T-cell factor/lymphocyte enhancer factor; DLL1: Delta-like 1; Tgif2: transforming growth factor-beta–induced factor 2; 3′-UTR: 3′-untranslated region; qRT-PCR: quantitative real-time polymerase chain reaction; TGF-β: transforming growth factor β; NAD: nicotinamide adenine dinucleotide.
As mentioned above, in order to study the mechanism of miR-34a action on its putative target gene expression, the mRNA expression of target gene upon miR-34a ectopic overexpression was determined by qRT-PCR. It is important to note that among the target genes listed in Table 3, DLL1 and Notch1 mRNA levels in choriocarcinoma cell lines (BeWo and JEG-3 cells) 145 and Smad4 mRNA levels in cholangiocarcinoma cell lines (QBC939 and HuCCT1) 128 were not significantly influenced by the overexpression or inhibition of miR-34a, while the protein expression levels of these genes were found decreased by transfection with the miR-34a mimics, suggesting that the expression of the genes was primarily inhibited by miR-34a at the translational level.128,145
Here, an attempt was made to identify several direct target genes of miR-34a whose aberrant expressions play important roles in carcinogenesis. However, many genes are indirectly regulated by miR-34a and play roles in cancer development and progression. For example, survivin, encoded by the gene BIRC5, acts as an anti-apoptotic protein and a cell cycle regulator implicated in the G1-to-S transition. Survivin has oncogenic potential and overexpresses in many human cancers.10,89,151 On the basis of current literature, miR-34a can negatively regulate survivin protein expression.10,89,152 Although prediction algorithms such as TargetScan predict a seed sequence for miR-34a at 3′-UTR of survivin, to the best of our knowledge, direct binding of miR-34a to the 3′-UTR of survivin was not confirmed by luciferase reporter assays in previously published studies. miR-34a was significantly down-regulated in head and neck squamous cell carcinoma (HNSCC) tissues and cell lines. Studies with tumour samples from HNSCC patients revealed an inverse relationship between miR-34a and survivin as well as miR-34a and E2F3 levels. Ectopic overexpression of miR-34a in HNSCC cell line (UM-SCC-74A) markedly decreased E2F3 and survivin levels. In the rescue experiments, survivin expression was completely restored by a miR-34a-resistant isoform of E2F3a, thereby suggesting that miR-34a may indirectly regulate survivin expression via E2F3a. 89 In another study, Chen et al. demonstrated that cell death process induced by CDK1 inhibitors was dependent, although not completely, on the induction of the miR-34a expression, which subsequently down-regulated the expression of MYCN and then decreased the transcriptional activation of MYCN on the survivin promoter. In other words, MYCN regulated the expression of survivin as a transcription factor and acted upstream of survivin. In this study, a critical role for the miRNA-34a–MYCN–survivin pathway was proposed in cell death induced by CDK1 inhibition in neuroblastoma cells. 151
Signalling pathways affected by miR-34a
miR-34a has multiple targets and thus works on multiple cell signalling pathways. Published data show that direct target genes of miR-34a are important mediators that regulate the various cell signalling pathways, such as Smad4 in transforming growth factor β (TGF-β)/Smad signalling pathway 128 ; Rictor, a core component of mammalian target of rapamycin (mTOR) complex 2, in phosphatidylinositide 3-kinase (PI3K)/AKT/mTOR and Wnt/β-catenin signalling pathways 136 ; and Bcl-2 in mitochondrial apoptosis signalling pathway 153 and AKT/nuclear factor-kappa B (NF-κB) signalling pathway. 154 Furthermore, several key components of the signalling pathways Notch and Wnt/β-catenin, which have shown to have major roles in cancer and CSC biology, are negatively regulated by miR-34a. In the following sections, we provide a brief summary of these two signalling pathways and the consequences of the aberrant expression of their key components in cancer cells.
Wnt/β-catenin signalling pathway
The canonical Wnt signalling pathway controls various cellular functions, including cell fate determination, embryonic development, survival, migration, metabolism, cell cycle regulation, proliferation, the maintenance of stem cell niches in adult tissues and the induction of epithelial–mesenchymal transition (EMT) programmes.138,139,155
The canonical Wnt/β-catenin signalling pathway is activated by the binding of secreted glycoproteins of the Wnt family as ligands to Frizzled receptors and LRP5/6 co-receptors in the plasma membrane. The LRP5/6 is phosphorylated. Dishevelled (Dsh) is recruited to the plasma membrane to interact with Frizzled and inactivate the destruction complex of β-catenin. Thus, cytoplasmic β-catenin is stabilized and accumulated and can translocate to the nucleus, where it interacts with T-cell factor/lymphocyte enhancer factor (TCF/LEF) transcription factors. This complex activates the transcription of a large number of downstream target genes.139,155 As noted in Table 3, Wnt1, LRP6, β-catenin and TCF/LEF involved in canonical Wnt signalling pathway have been identified as direct targets of miR-34a.
In addition to the importance of the canonical Wnt pathway in normal cell function, the aberrant activation of Wnt signalling is frequently involved in CSC generation and cancer progression. Wnt1-positive NSCLCs show higher proliferation capacities, and patients with Wnt1-positive NSCLCs have lower overall survival rates. 156 High levels of expression of Wnt1 and β-catenin in patients with prostate carcinoma were associated with advanced metastasis. 157 It was shown that the expression of LRP6 was up-regulated in a subset of human breast cancers. LRP6 silencing in breast cancer cells attenuated Wnt/β-catenin signalling and reduced cell proliferation as well as in vivo tumour growth. 140 Gao et al. found that the knockdown of LEF1 – a major transcription factor of Wnt pathway – inhibited glioblastoma multiforme (GBM) U251 cell migration, invasion and proliferation. Furthermore, down-regulation of LEF1 expression inhibited the self-renewal capacity of U251 GBM stem-like cells and decreased the expression level of the GBM stem-like cell markers such as CD133 and nestin. 141 In addition, Ishimoto et al. 49 found that Wnt signalling strongly enhanced the expansion of CD44(+) stem-like cells in the squamo-columnar junction in the presence of prostaglandin E2 signalling, leading to gastric tumourigenesis in K19-Wnt1/C2mE mice. Notably, CD44 – a well-known CSC marker – was found as a target of Wnt/β-catenin signalling pathway. 158
Notch signalling pathway
The highly conserved Notch signalling pathway is involved in various cell processes, such as proliferation, differentiation, apoptosis, adhesion, angiogenesis and EMT, depending on the cell type.130,159,160
Notch receptors are type I single-pass transmembrane proteins; they function as extracellular signal recipients and also act as transcription factors regulating gene expression inside the nucleus. In mammals, there are four Notch receptors (Notch 1–4) and five Notch ligands – Jagged (JAG) 1, 2 and Delta-like ligand (DLL) -1, -3, -4. Upon ligand binding, the transmembrane domain of Notch receptor is cleaved by the gamma-secretase complex. This cleavage allows the release and translocation of the intracellular domain of Notch into the nucleus, where it acts as a transcription factor to control the expression of downstream genes.130,160,161 Gamma secretase inhibitors have been used to inhibit Notch signalling. 161 As we have previously mentioned (Table 3), Notch1, Notch2, JAG1 and DLL1 have been identified as direct targets of miR-34a.
The role of Notch in cancers seems to be complex, depending on factors such as tissue type.159,160 Nevertheless, abnormal activation of Notch signalling (most commonly its overactivation) has been shown to be associated with carcinogenesis. In the following section, we refer to some studies. Purow et al. reported that Notch1, DLL1 and JAG1 were overexpressed in glioma cell lines and primary human gliomas. Knockdown of Notch1, DLL1, or JAG1 expression induced apoptosis and inhibited proliferation in multiple glioma cell lines. Furthermore, they injected human glioma cells with knocked-down Notch1 or DLL1 expression into mice and observed a significant prolongation of survival. 159 Another study showed that patients with breast tumours expressing high levels of JAG1 or Notch1 had significantly poorer overall survival compared with patients expressing low levels of these genes. 162 JAG1 expression was associated with a basal phenotype and recurrence in lymph node–negative breast cancer. 163 Pang et al. found that the down-regulation of Notch1, using gamma-secretase inhibitor and RNA interference, reduced the invasiveness of cervical cancer and trophoblast cell lines (HeLa and JAR cells, respectively). However, the forced expression of intracellular domain of Notch did not have significant effect on cell proliferation. They also demonstrated that miR-34a inhibited the expression of JAG1 and its receptor Notch1, and reduced the invasion capacity of HeLa and JAR cells through the regulation of the Notch pathway. 130 Published studies have provided evidence that Notch signalling also regulates self-renewal and survival of CSCs. For example, Simmons et al. showed that Notch1 inhibition reduced mammary tumoursphere-forming activity in vitro and tumour-initiating activity in vivo – a characteristic associated with CSCs. They also identified the embryonic stem cell pluripotency transcription factor Nanog as a Notch1-regulated gene in mammary tumour spheres and breast cancer cell lines. 161 Tatarek et al. 164 found that Notch1 inhibition in Tal1/Lmo2 mouse model of T-cell acute lymphoblastic leukaemia (T-ALL) significantly reduced or eliminated leukaemia CSCs and extended animal survival rate.
Conclusion
In conclusion, CSCs are responsible for tumour initiation, metastasis and therapeutic resistance. Because of the resistance of CSCs to traditional chemotherapy and radiotherapy, new treatments targeting CSCs are needed for improving patient survival rate. Thus, understanding the molecular mechanism that regulates the biological functions of CSCs is necessary. The tumour suppressor miR-34 plays crucial roles in the regulation of important biological processes, including cell cycle arrest, senescence and induction of apoptosis. However, miR-34 expression is down-regulated or lost in various types of cancers including GC, thereby playing a pivotal role in carcinogenesis, as now a huge volume of data strongly links deregulated miRNAs expression to the development of many cancers. Significantly, several key mediators of multiple signalling pathways, whose abnormal activation is important in cancerogenesis, have been identified as valid target genes of miR-34a. For example, JAG1 – a validated target of miR-34a – is a ligand for Notch signalling pathway; it can also be induced by other signalling pathways such as TGF-β and Wnt/β-catenin. The ability to affect multiple cancer signalling pathways seems to be an important benefit of the therapeutic use of miRNA-34a. This concept implies that miR-34a have the potential to regulate multiple malignant aspects of cancer biology, including metastasis, CSC generation, tumour angiogenesis, neoplastic cell growth and survival and therapy resistance. It has been shown that re-introduction (or restoration) of down-regulated miR-34a in cancer cells reactivates cellular pathways and thus provides a promising opportunity to treat cancer. However, most of our information about the function of the miRNA-34a as well as its role in CSC generation or differentiation comes only from studies in cell culture models. Clearly, further studies should be conducted to investigate the potential of tumour suppressor miR-34a in cancer treatment or the improvement of patient survival rates.
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Ethical approval
This review article does not contain any studies with human participants or animals performed by any of the authors.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by Immunogenetics Research Center, Faculty of Medicine, Mazandaran University of Medical Sciences (grant number 2119).
Informed consent
Not applicable to this type of document.
