OncomiRs: the discovery and progress of microRNAs in cancers

Calin et al. first made the connection between microRNAs and cancer by showing that miR-15 and miR-16 are located at chromosome 13q14, a region deleted in more than half of B-cell chronic lymphocytic leukemia (CLL). Detailed deletion and expression analysis showed that miR-15 and miR-16 are located within a 30 kb region of loss in CLL, and that both genes were deleted or downregulated in approximately 68% of CLL cases [8]. They further showed that miR genes were frequently located in cancer-associated genomic regions or in fragile sites. The full complement of miRNAs in a genome might be extensively involved in cancers [9]. Bottoni et al. found that miR-15a and miR-16-1 were expressed at lower levels in pituitary adenomas as compared to normal pituitary tissue. Their expression was inversely correlated with tumor diameter and with arginyl-tRNA synthetase expression, but was directly correlated with p43 secretion, suggesting that these miRNAs might influence tumor growth [10]. Cimmino et al. then demonstrated that miR-15a and miR-16-1 expressions were inversely correlated to Bcl2 expression in CLL and that both miRNAs negatively regulated Bcl2 at a posttranscriptional level. Bcl2 repression by these miRNAs induced apoptosis in a leukemic cell line model. Therefore miR-15 and miR-16 were natural antisense Bcl2 interactors that could be used for therapy of Bcl2-overexpressing tumors [11]. Recently, Garzon et al. showed that all-trans retinoic acid (ATRA) downregulation of Bcl2 and Ras was correlated with the activation of miR-15a/miR-16-1 [12].

Amplification and overexpression of the miR-17-92, which comprised 7 miRNAs and resided in intron 3 of the C13orf25 gene at 13q31.3, has been reported with pointers to functional involvement in the development of lymphoma and lung cancer. He et al. compared B-cell lymphoma samples and cell lines to normal tissues, and found that the levels of the primary or mature miRNAs derived from the miR-17-92 locus were often substantially increased in cancer cells. Their studies indicated that miRNAs could modulate tumor formation and function as oncogenes, implicating the miR-17-92 cluster as a potential human oncomicroRNAs (oncomiRs) [13]. O'Donnell et al. demonstrated that c-Myc activated expression of a set of 6 miRNAs on human chromosome 13 that was tied to the development of human lymphoma. It was found that the expression of E2F1 was negatively regulated by miR-17-5p and miR-20a in HeLa cells. Their findings revealed a mechanism through which the c-Myc protein simultaneously activated E2F1 transcription and limited its translation, allowing a tightly controlled proliferative signal [14]. Woods et al. proposed a model whereby miR-17-92 promoted cell proliferation by shifting the E2F transcriptional balance away from the pro-apoptotic E2F1 and toward the proliferative E2F3 transcriptional network [15].

On the other hand, Hayashita et al. found that miR-17-92 was markedly overexpressed in lung cancer, especially with small-cell lung cancer. Their findings suggested that marked overexpression of the miR-17-92 cluster with occasional gene amplification might play a role in the development of lung cancer and that the C13orf25 gene might well be serving as a vehicle in this regard [16]. Matsubara et al. showed that inhibition of miR-17-5p and miR-20a with antisense oligonucleotides could induce apoptosis selectively in lung cancer cells overexpressing miR-17-92, suggesting the possibility of oncomiR addiction to expression of these miRNAs in a subset of lung cancers. Their discoveries contributed towards better understanding of the oncogenic roles of miR-17-92, which might ultimately lead to the future translation into clinical applications [17].

Iorio et al. showed that miRNAs were aberrantly expressed in human breast cancer compared with normal breast tissue, with the most significantly downregulated miRNAs being miR-10b, miR-125b and miR-145, whereas the most significantly upregulated miRNAs being miR-21 and miR-155. The miRNA expression was also found to be correlated with specific breast cancer biopathologic features, such as estrogen and progesterone receptor expression, tumor stage, vascular invasion or proliferation index [18]. Si et al. found that miR-21 was highly overexpressed in breast tumors compared to the matched normal breast tissues. They found that the anti-miR-21-mediated cell growth inhibition was associated with increased apoptosis and decreased cell proliferation. Their results suggested that miR-21 functioned as an oncogene and modulated tumorigenesis through regulation of genes such as Bcl2 and it might serve as a novel therapeutic target [19]. Zhu et al. performed two-dimensional differentiation in-gel electrophoresis of tumors treated with anti-miR-21 and identified the tumor suppressor tropomyosin 1 as a potential miR-21 target. Downregulation of tropomyosin 1 in breast cancer by miR-21 might explain why suppression of miR-21 could inhibit tumor growth, further supporting the notion that miR-21 functions as an oncogene [20]. Meng et al. identified miR-21 was also highly overexpressed in malignant cholangiocytes. Inhibition of miR-21 increased sensitivity to gemcitabine, it modulated gemcitabine-induced apoptosis by phosphatase and tensin homolog deleted on chromosome 10-dependent activation of PI 3-kinase signaling [21]. Tran et al. found that miR-21 was highly expressed in the head and neck cancer cell lines. Several tumor suppressor genes were identified to be potential targets of miRNAs, including kinesin family member 1B isoform alpha, hypermethylated in cancer 2 and pleomorphic adenoma gene 1 [22].

Costinean et al. showed that E(mu)-mmu-miR-155 transgenic mice exhibited preleukemic pre-B-cell proliferation evident in spleen and bone marrow, followed by frank B-cell malignancy. Their findings indicated that miR-155 could induce polyclonal expansion, favoring the capture of secondary genetic changes for full transformation [23]. Using miRNA cloning and qRT-PCR of mature miRNAs, Fulci et al. demonstrated that miR-21 and miR-155 were dramatically overexpressed in CLL patients [24]. Besides, Roldo et al. showed that the expression of miR-103 and miR-107, associating with a lack of miR-155 expression, could discriminate pancreatic tumors from normal pancreas. Their results suggested that the alteration in miRNA expression was related to endocrine and acinar neoplastic transformation [25].

Felli et al. demonstrated that treatment of CD34+ progenitors with miR-221 and miR-222 caused impaired proliferation and accelerated differentiation of erythropoietic cells, coupled with the downmodulation of Kit protein. The decline of miR-221 and miR-222 during exponential erythropoietic growth unblocked Kit protein production at mRNA level, thus leading to expansion of early erythroblasts [26]. He et al. also showed that miR-146, miR-221 and miR-222 distinguished unequivocally between papillary thyroid carcinoma (PTC) and normal thyroid. The upregulation of these miRNAs was strongest associated with a dramatic loss of Kit transcript and its protein. Sequence changes in genes targeted by these miRNAs could contribute to the regulation of Kit involved in PTC pathogenesis [27]. Analyzing the genome-wide miRNA expression profile in human PTCs using microarray, Pallante et al. detected a significant increase of miR-181b, miR-221 and miR-222 in the comparison of PTCs with normal thyroid tissue. Further confirmation by Northern blot analysis and qRT-PCR, their results suggested miRNA deregulation as an important event in thyroid cell transformation [28].

The analysis of both glioblastoma tissues and glioblastoma cell lines allowed Ciafre et al. to identify a group of miRNAs whose expression was significantly altered in the tumor. miR-221 was strongly upregulated, whereas miR-128, miR-181a, miR-181b and miR-181c were downregulated in glioblastoma [29]. Pekarsky et al. discovered that the expression levels of miR-29 and miR-181 were inversely correlated with Tcl1 expression in CLL. Their results showed that miR-29 and miR-181 might be candidates for therapeutic agents in CLL overexpressing Tcl1 [30]. By in silico analysis, Mott et al. identified a putative target site in the Mcl1 mRNA and found that miR-29b was downregulated in malignant cells, consistent with Mcl1 protein upregulation. Enforced miR-29b expression reduced Mcl1 protein expression in the KMCH cholangiocarcinoma cells, thus miR-29 was an endogenous regulator of Mcl1 protein expression [31].

Examined by RT-PCR, Bandres et al. identified 13 significantly deregulated mature miRNAs in colorectal cancer, including miR-31, miR-96, miR-133b, miR-135b, miR-145 and miR-183. In addition, the expression level of miR-31 was correlated with the stage of colorectal cancer [32]. Akao et al. found that miR-143 and miR-145 expression levels were extremely reduced in the colon cancer cells. The transfection of each precursor miRNA into the cells demonstrated a significant growth inhibition in human colon cancer DLD-1 and SW480 cells, whereas Erk5 was determined to be the target gene of miR-143 [33].

Laneve et al. showed that miR-9, miR-125a and miR-125b acted in an additive manner by repressing the truncated isoform of the neurotrophin receptor tropomyosin-related kinase C. They found that the downregulation of this isoform was critical for regulating neuroblastoma cell growth [34].In vitro functional studies of neuroblastoma cell lines indicated that miR-184 played a significant role in apoptosis. Chen and Stallings suggested that neuroblastoma derived Myc myelocytomatosis viral related oncogene might mediate a tumorigenic effect through directly or indirectly regulating the expression of miRNAs that were involved with neural cell differentiation and/or apoptosis [35].

Welch et al. showed that miR-34a on chromosome 1p36.23 was generally expressed at lower levels in unfavorable primary neuroblastoma tumors relative to normal tissue. miR-34a directly targeted the mRNA encoding E2F3 and significantly reduced E2F3 protein levels. Their results suggested that miR-34a acted as a suppressor of neuroblastoma tumorigenesis [36]. Chang et al. showed that miR-34a was frequently absent in pancreatic cancer cells. They demonstrated that this miRNA was directly transactivated by p53. miR-34a-responsive genes were highly enriched for those that regulated cell-cycle progression, apoptosis, DNA repair and angiogenesis. It was likely that an important function of miR-34a was the modulation and fine-tuning of the gene expression program initiated by p53 [37].

Nervi et al. found that the expression level of miR-223 was correlated with the differentiation fate of myeloid precursors. The activation of both pathways of transcriptional regulation by the myeloid lineage-specific transcription factor CCAAT/enhancer-binding protein-α (C/EBPa) and posttranscriptional regulation by miR-223 appeared essential for granulocytic differentiation and clinical response of acute promyelocytic leukemia blasted to ATRA [38]. Using miRNA microarray platform and qRT-PCR, Garzon et al. reported the expression of miRNAs in acute promyelocytic leukemia patients and cell lines during ATRA treatment. They found upregulation of miR-107 targeted nuclear factor 1-A, a gene involving miR-223 and C/EBPa in a regulatory loop during granulocytic differentiation. Besides, ATRA downregulation of Ras and Bcl2 was shown to correlate with the activation of let-7a miRNA [12].

Johnson et al. showed that the let-7 miRNA family negatively regulated let-60/Ras. Loss of let-60/Ras suppressed let-7 miRNA complementary sites, restricting reporter gene expression in a let-7-dependent manner. let-7 miRNA expression was lower in lung tumors than in normal lung tissue, while Ras protein was significantly higher in lung tumors, providing a possible mechanism for let-7 miRNA in cancer [39]. Akao et al. found that the levels of Ras and c-Myc proteins in let-7 miRNA low-expressing DLD-1 human colon cancer cells were lowered after the transfection with let-7a-1 precursor miRNA, whereas the levels of both of their mRNAs remained almost unchanged. Their findings suggested the involvement of let-7 miRNA in the growth of colon cancer cells [40]. Meng et al. demonstrated that let-7a miRNA was upregulated and contributed to the survival effects of enforced Interleukin-6 activity in malignant human cholangiocytes. It contributed to the constitutively increased phosphorylation of the signal transducers and activators of transcription-3 factors by a mechanism involving neurofibromatosis 2 [41]. Brueckner et al. showed that the human let-7a-3 precursor miRNA on chromosome 22q13.31 was associated with a CpG island. They identified let-7a-3 precursor miRNA as an epigenetically regulated miRNA gene in lung adenocarcinomas with oncogenic function and suggested that aberrant miRNA gene methylation might contribute to the human cancer epigenome [42].

Mayr et al. reported that the chromosomal translocations associating with human tumors disrupted the repression of High mobility group A2 (Hmga2) by let-7 miRNA. They found that the loss of miRNA-directed repression of an oncogene provided a mechanism for tumorigenesis, and disrupting a single miRNA-target interaction could produce an observable phenotype in mammalian cells [43]. Lee and Dutta also demonstrated that ectopic expression of let-7 miRNA reduced Hmga2 and cell proliferation in lung cancer cell. Their results suggested that some tumors activated the oncogene through chromosomal translocations that eliminated the oncogene's 3' untranslated region with the let-7 miRNA target sites [44]. Hebert et al. reported that Hmga2 expression in head and neck squamous cell carcinoma cells was regulated in part by miR-98. They showed that the transfection of pre-miR-98 during normoxia diminished Hmga2 and potentiate resistance to doxorubicin and cisplatin. Their findings implicated the role of miR-98 as a key element in modulating tumors during hypoxia [45].

Stem cells have the ability to escape cell cycle stop signals, which are similar to cancer cells. On the basis of cell cycle markers and genetic interactions, Harfield et al. reported that dicer-1 mutant germline stem cells were delayed in the G1 to S transition, which was dependent on the cyclin-dependent kinase inhibitor dacapo, suggesting that miRNAs were required for stem cells to bypass the normal G1/S checkpoint. The miRNA pathway might be part of a mechanism that made stem cells insensitive to environmental signals that normally stop the cell cycle at the G1/S transition [46].

Lu et al. used a bead-based flow cytometric miRNA expression profiling method to present a systematic expression analysis of miRNAs from human cancer samples, including colon, liver, pancreas and stomach cancers. The miRNA expression profiles were able to successfully classify poorly differentiated tumors. Their findings highlighted the potential of miRNA profiling in cancer diagnosis [47].

Analyzing the entire miRNAome in pituitary adenomas and in normal pituitary samples by microarray and RT-PCR, Bottoni et al. showed that miRNA expression could differentiate micro- from macro-adenomas and the treated from non-treated patient samples. Several miRNAs were involved in cell proliferation and apoptosis. Predictive miRNAs could be potentially useful diagnostic markers, improving the classification of pituitary adenomas [48].

With the application of in situ RT-PCR, Lee et al. showed that the aberrantly expressed miR-221, miR-301 and miR-376a were localized to pancreatic cancer cells but not to stroma or normal acini or ducts. Aberrant miRNA expression offered new clues to pancreatic tumorigenesis and might provide diagnostic biomarkers for pancreatic cancer [49].

Takamizawa et al. reported that expression of let-7 miRNA was frequently reduced in human lung cancers, and that reduced let-7 miRNA expression was significantly associated with shorter postoperative survival. In addition, overexpression of let-7 miRNA in A549 lung adenocarcinoma cell line inhibited lung cancer cell growth in vitro. Their study reported the potential clinical and biological effects of let-7 miRNA alteration [50]. Yanaihara et al. identified miRNA expression profiles which could discriminate lung cancers from noncancerous lung tissues as well as molecular signatures that differed in tumor histology. High hsa-miR-155 and low hsa-let-7a-2 precursor miRNA expression were found to be correlated with poor survival of lung adenocarcinomas. Their results indicated that miRNA expression profiles were not just diagnostic markers, but also prognostic markers of lung cancer [51].

Roldo et al. showed that the overexpression of miR-21 in pancreatic tumors was strongly associated with both high Ki67 proliferation index and the presence of liver metastasis. Their results suggested that the alteration in miRNA expression was related to progression of malignancy [25]. On the other hand, Bloomston et al. compared the expression pattern of miRNAs in pancreatic cancer with those of normal pancreas and chronic pancreatitis using miRNA microarrays. Differentially expressed miRNAs were identified which could differentiate pancreatic cancer from normal pancreas, chronic pancreatitis, or both. High expression of miR-196a-2 was found to predict poor survival [52].