MicroRNAs (miRNAs or miRs) are a family of small (19–25 nucleotides in length) non-coding RNAs that have a key role in the regulation of gene expression through the inhibition or the reduction of protein synthesis following mRNA complementary sequence base pairing (1–4). A single or multiple mRNAs can be targeted at the 3′ untranslated region (3′UTR), coding sequence (CDS) or 5′ untranslated region (5′UTR) sequence, and it is calculated that >60% of human mRNAs are recognized by miRNAs (1–4). The miRNA/mRNA interaction occurs at the level of the RNA-induced silencing complex (RISC) and causes translational repression or mRNA degradation, depending on the degree of complementarity with target mRNA sequences (5–8). Since their discovery and first characterization, the number of human miRNAs identified and deposited in the miRBase databases (miRBase v.21, www.mirbase.org) has increaed (it is >2,500) (9–16) and the research studies on miRNAs have confirmed the very high complexity of the networks constituted by miRNAs and RNA targets (17–22).

Alterations in miRNA expression have been demonstrated to be associated with different human pathologies, and guided alterations of specific miRNAs have been suggested as novel approaches for the development of innovative therapeutic protocols (23,24). Studies have conclusively demonstrated that miRNAs are deeply involved in tumor onset and progression, either behaving as tumor-promoting miRNAs (oncomiRNAs and metastamiRNAs) or as tumor suppressor miRNAs (25,26). In general, miRNAs able to promote cancer target mRNAs coding for tumor suppressor proteins, whereas miRNAs exhibiting tumor suppressor properties usually target mRNAs coding oncoproteins (see the scheme depicted in Fig. 1A). This has a very important implication in diagnosis and/or prognosis, including the recent discovery that the pattern of circulating cell-free miRNAs in serum allows us to perform molecular analyses on these non-invasive liquid biopsies with deep diagnostic and prognostic implications. This research field has confirmed that cancer-specific miRNAs are present in extracellular body fluids, and may play a very important role in the crosstalk between cancer cells and surrounding normal cells (27–32).

Interestingly, the evidence of the presence of miRNAs in serum, plasma and saliva supports their potential as an additional set of biomarkers for cancer. The extracellular miRNAs are protected by exosome-like structures, small intraluminal vesicles shed from a variety of cells (including cancer cells), with a biogenesis connected with endosomal sorting complex required for transport machinery in multivesicular bodies (29). For instance, miR-141 and miR-221/222 are predicted biomarkers in liquid biopsies from patients with colon cancer (33,34).

On the other hand, tumor-associated miRNAs are suitable targets for intervention therapeutics, as previously reported (35–44) and summarized in Fig. 1B. The inhibition of miRNA activity can be readily achieved by the use of miRNA inhibitors and oligomers, including RNA, DNA and DNA analogues (miRNA antisense therapy) (45–47), small molecule inhibitors, locked nucleic acids (LNAs) (48–53), peptide nucleic acids (PNAs) (54–57), morpholinos (58–60), miRNA sponges (61–67), mowers (68) or through miRNA masking that inhibits miRNA function by masking the miRNA binding site of a target mRNA using a modified single-stranded RNA complementary to the target sequence (69–75). On the contrary, the enhancement of miRNA function (miRNA replacement therapy) can be achieved by the use of modified miRNA mimetics, either synthetic, or produced by plasmid or lentiviral vectors carrying miRNA sequences (76–81).

Several miRNAs exhibit onco-suppressor properties by targeting mRNAs coding oncoproteins (82–105). Therefore, these onco-suppressor miRNAs have been found to be often downregulated in tumors. For instance, Fernandez et al (106) recently described the intriguing tumor suppressor activity of miR-340, showing the miR-340-mediated inhibition of multiple negative regulators of p27, a protein involved in apoptosis and cell cycle progression. These interactions with oncoprotein-coding mRNA targets determine the inhibition of cell cycle progression, the induction of apoptosis and growth inhibition. The miR-340-mediated downregulation of three post-transcriptional regulators [Pumilio RNA-binding family member (PUM)1, PUM2 and S-phase kinase-associated protein 2 (SKP2)] correlates with the upregulation of p27. PUM1 and PUM2 inhibit p27 at the translational level, by rendering the p27 transcript available to interact with two oncomiRs (miR-221 and miR-222), while the oncoprotein SKP2 inhibits the CDK inhibitor at the post-translational level by triggering the proteasomal degradation of p27, showing that miR-340 affected not only the synthesis but also the decay of p27. Moreover their data confirm the recent identification of transcripts encoding several pro-invasive proteins such as c-Met, implicated in breast cancer cell migration, RhoA and Rock1, implicated in the control of the migration and invasion of osteosarcoma cells, and E-cadherin mRNA, involved in the miR-340-induced loss of intercellular adhesion (106 and refs within).

Recently, miR-18a was demonstrated to play a protective role in colorectal carcinoma (CRC) by inhibiting the proliferation, invasion and migration of CRC cells by directly targeting the TBP-like 1 (TBPL1) gene. The onco-suppressor activity of miR-18a in CRC tissues and cell lines was supported by the finding that the content of this mRNA is markedly lower in tumor cells with respect to normal control tissues and cells (107). In addition Xishan et al (108) found that miR-320a acts as a novel tumor suppressor gene in chronic myelogenous leukemia (CML) and can decrease the migratory, invasive, proliferative and apoptotic behavior of CML cells, as well as epithelial-mesenchymal transition (EMT), by attenuating the expression of the BCR/ABL oncogene. Furthermore Zhao et al (109) demonstrated that miR-449a functions as a tumor suppressor in neuroblastoma by inducing cell differentiation and cell cycle arrest. Finally, Kalinowski et al (110) and Gu et al (111) demonstrated the significant role of miR-7 in cancer which functions by directly targeting and inhibiting key oncogenic signaling molecules involved in cell cycle progression, proliferation, invasion and metastasis. A partial list of onco-suppressor miRNAs is presented in Table I.

miRNAs can act as oncogenes and have been demonstrated to play a causal role in the onset and progression of human cancer (oncomiRNAs) (224–233). Recent findings have nevertheless identified a subclass of miRNAs whose expression is highly associated with the acquisition of metastatic phenotypes and are referred to as miRs endowed with either metastasis-promoting or tumor suppressor inhibitory activities (213,234,235).

Recent data have revealed that miR-25 may act as an onco-miRNA in osteosarcoma, negatively regulating the protein expression of the cell cycle inhibitor, p27. In agreement with this hypothesis restoring the p27 level in miR-25-over-expressing cells was shown to reverse the enhancing effect of miR-25 on Saos-2 and U2OS cell proliferation (236). In addition a recent study published by Siu et al (237), describes miR-96 as a potential target of therapeutics for metastatic prostate cancer, demonstrating the enhanced effects in cellular growth and invasiveness of miR-96 in cell lines (AC1, AC3 and SC1) derived from prostate-specific, Pten/Tp53 double knockout mice and confirmed in tissue samples from prostate cancer patients. miR-96 acts as an oncomiR and metastamiR through TGF-β/mTOR signaling, promoting bone metastasis and contributing to a reduced survival rate in prostate cancer (237). Furthermore Xia et al (238) demonstrated that the overexpression of miR-1908 significantly decreased the expression of PTEN in glioblastoma cells, one of the most frequently mutated tumor suppressors in human cancer, resulting in an increase in proliferation, migration and invasion. Finally Sachdeva et al (239), found that miR-182 targets multiple genes in lung metastasis and regulates intravasation, thus increasing the number of circulating tumor cells (CTCs). Only the simultaneous restoration of miR-182 target genes decreased the number of metastases in vivo, demonstrating that a single miRNA can regulate the metastasis of primary tumors in vivo by the coordinated regulation of multiple genes. Selected examples of oncomiRNAs and metastamiRNAs are presented in Tables II and III. All these miRNAs act by inhibiting tumor suppressor pathways.

Using the development of anticancer therapies as a representative field of investigation, the therapeutic strategy based on miRNA replacement is targeted to pathological cells which downregulate onco-suppressor miRNAs playing a role in controlling the expression of mRNAs encoding key oncoproteins. The downregulation of these oncogene-targeting miRNAs is clearly the key step for oncogene upregulation leading to tumor onset and progression. Table IV presents selected examples of miRNA replacement therapy in cancer research and treatment (90–92,94–97,99).

As a first representative example, Fig. 2A presents the major results obtained by Wu et al (97), who reported that the in vivo restoration of miR-29b may represent an option for lung cancer treatment. To demonstrate the efficacy of this strategy, they developed a cationic lipoplexes (LPs)-based carrier that efficiently delivered miR-29b both in vitro and in vivo. LPs containing miR-29b (LP-miR-29b) efficiently delivered miR-29b to A549 cells and reduced the expression of the key target, CDK6. In a xenograft murine model, in which LPs efficiently accumulated at tumor sites, the systemic delivery of LP-miR-29b increased miR-29b expression in tumors, downregulated CDK6 mRNA expression in tumors and, as shown in the upper panels of Fig. 2A, significantly inhibited tumor growth.

A second example of miRNA replacement therapy has been published by Glover et al (304), who reported that miR-7-5p (miR-7) reduces cell proliferation in vitro and induces G1 cell cycle arrest. The systemic miR-7 administration with delivery vesicles reduced adrenocortical carcinoma (ACC) xenograft growth originating from both ACC cell lines and primary ACC cells. As far as the potential mechanisms of action, miR-7 was demonstrated to target Raf-1 proto-oncogene serine/threonine kinase (RAF1). Additionally, miR-7 therapy in vivo led to the inhibition of cyclin dependent kinase 1 (CDK1) (304). Two other methods have also been used to successfully deliver miR-7 in vivo to treat cancer. In a study by Babae et al (305), a miR-7 mimic was systemically delivered using clinically viable, biodegradable, targeted polyamide nanoparticles. This strategy led to the successful inhibition of tumor growth and vascularisation in a glioblastoma xenograft model system. In an earlier study, Wang et al (306) was able to inhibit glioma xenograft growth and metastasis using a plasmid based miR-7 vector systemically delivered by encapsulation in a cationic liposome formulation.

Moreover, Cortez et al (307) revealed a novel function of miR-200c, a member of the miR-200 family, in regulating intracellular reactive oxygen species signaling. They used a lung cancer xenograft model to demonstrate the therapeutic potential of the systemic delivery of miR-200c to enhance radiosensitivity in lung cancer. The results obtained suggest that the antitumor effects of miR-200c result partially from its regulation of the oxidative stress response; they further suggested that miR-200c, in combination with radiation, may represent an effective therapeutic strategy in the future.

Recently, Wu et al (308) reported that the expression of miR-708-5p suppressed lung cancer invasion and metastasis in vitro and in vivo. In particular, it induces apoptosis and suppresses cell migration by inhibiting the cytoplasmic localization of p21, and also weakens the stem cell-like properties of lung cancer cells. In their study, they present the systemic delivery of the PEI/miR-708-5p complexes for miRNA replacement therapy in a mouse model of lung cancer, demonstrating an efficient antitumor activity with no side-effects.

The effects of therapeutic molecules against miRNAs have been the object of very recent studies, in part summarized in Table V (309–316). Of course, the endpoint of the treatment of target cells with molecules against selected miRNAs is the alteration of miRNA-regulated genes. As a first example, Wagenaar et al (317) developed potent and specific single-stranded oligonucleotide inhibitors of miR-21 and used them to verify dependency on miR-21 in a panel of liver cancer cell lines. Treatment with anti-miR-21, but not with a mismatch control anti-miRNA, resulted in the significant derepression of direct targets of miR-21 and led to the loss of viability in the majority of HCC cell lines tested. The robust induction of caspase activity, apoptosis and necrosis was noted in the anti-miR-21-treated HCC cells. Furthermore, the ablation of miR-21 activity resulted in the inhibition of HCC cell migration and in the suppression of clonogenic growth (317).

In another study, using PNAs as anti-miRNA molecules, Fabani et al (318) targeted miR-155, demonstrating the deregulation of mRNA Bat5, Sfp1 and Jarid2. In our laboratory, Brognara et al analyzed the effects of PNAs targeting miR-221 on breast cancer cells (319). In order to maximize uptake in target cells, a polyarginine-peptide (R8) was conjugated, generating an anti-miR-221 PNA displaying very high affinity for RNA and efficient uptake within target cells without the need for transfection reagents. Targeting miR-221 with this PNA molecule resulted in i) a specific decrease in the hybridization levels of miR-221 measured by RT-qPCR, ii) the upregulation of p27^Kip1 mRNA and protein expression, measured by RT-qPCR and western blot analysis, respectively. As regards the in vivo effects of anti-miRNA therapy, Yan et al (320) addressed the potential effects of PNA-anti-miR-21 in vivo on the growth of breast cancer cells. In their experiments, MCF-7 cells treated with PNA-anti-miR-21 or PNA-control were subcutaneously injected into female nude mice and detectable tumor masses were observed in few mice in the MCF/PNA-anti-miR-21 group, while much larger tumors were detected in all mice in the MCF/PNA-control group. Both tumor weight and number showed that MCF/PNA-control cells formed larger tumors more rapidly than MCF/PNA-anti-miR-21 cells in nude mice. As a final example, Cheng et al (57) demonstrated that the PNA anti-miRs with a peptide with a low pH-induced transmembrane structure (pHLIP) target the tumor microenvironment, transport anti-miRs across plasma membranes under acidic conditions, such as those found in solid tumors and effectively inhibit the miR-155 oncomiR in a mouse model of lymphoma.

EMT is a powerful process in tumor invasion, metastasis and tumorigenesis, and describes the molecular reprogramming and phenotypic changes that are characterized by a transition from polarized immotile epithelial cells to motile mesenchymal cells (Fig. 3). This process is characterized by the loss of polarity and cell-cell contacts by the differentiated epithelial cells, with deep alterations occurring at the level of tight junctions and desmosomes. The breach of the basement membrane is a following step, leading to the invasion of blood and/or lymphatic vessels by these mesenchymal differentiated cancer cells, which at the end of the process, causes migration, often accompanied by drug resistance (Fig. 3). It is now well-known that several miRNAs are important regulators of EMT. Some of these are miR-7, miR-17/20, miR-22, miR-30, miR-200 and its family members. Most of these miRNAs potentiate EMT, while some well-characterized miRNAs play a suppressive role in EMT. For instance, the metastasis suppressor role of the miR-200 members is strongly associated with the inhibition of EMT. This is well described in the published review by Zhang and Ma (321), and in the studies by Zaravinos et al (322) and Kiesslich et al (323), showing the most recent advances regarding the influence of miRNAs in EMT and the regulatory effects they exert on major signaling pathways in various types of cancer (Fig. 3). In Caski cervical cancer cells, the oncomiR-155 acts as a tumor suppressor and suppresses EGF-induced EMT, decreasing migration/invasion capacities, inhibiting cell proliferation and enhancing the chemosensitivity to DDP in humans (324). Chang et al (279) demonstrated that the overexpression of miR-20a in gallbladder carcinoma cells induced EMT and promoted metastasis via the direct inhibition of Smad7, correlating this miRNA with local invasion, distant metastasis and a poor prognosis in patients with gallbladder carcinoma.

In the ovarian surface epithelium, EMT is considered the key regulator of the post-ovulatory repair process and it can be triggered by a range of environmental stimuli. The aberrant expression of the miR-200 family (miR-200a, miR-200b, miR-200c, miR-141 and miR-429) in ovarian cancer, and its involvement in the initiation and progression of ovarian cancer have been well demonstrated. The miR-200 family members seem to be strongly associated with EMT and to have a metastasis suppressor role. miRNA signatures can accurately distinguish ovarian cancer from the normal ovary and can be used as diagnostic tools to predict the clinical response to chemotherapy. Recent evidence suggests a growing list of novel miRNAs (miR-187, miR-34a, miR-506, miRNA-138, miR-30c, miR-30d, miR-30e-3p, miR-370 and miR-106a, among others) that are also implicated in ovarian cancer-associated EMT, either enhancing or suppressing it. MicroRNA-based gene therapy provides a prospective antitumor approach for integrated cancer therapy (325).

As regards the molecular targets of EMT-regulating miRNAs, several are known and validated. Among these, transcription factors play a very important role. For instance, Gao et al (326) identified SOX2 as a key player in EMT, by examining the effects of its overexpression. They demonstrated that SOX2-overexpressing Eca-109 cells exhibited an enhanced cell migration/invasion capacity. Moreover, these cells exhibited characteristics of EMT, that is, a significantly suppressed expression of the epithelial cell marker with a concomitant enhancement in the expression of mesenchymal markers. An increased expression of Slug in SOX2-overexpressing cells suggested the involvement of this transcription factor in SOX2-regulated metastasis. Finally, the expression levels of STAT3/HIF-1α were found to be upregulated in SOX2-expressing cells, and the blockade of these transcription factors resulted in the inhibition of Slug expression at both the protein and mRNA level.

Of interest, is also the finding that miR-221/222, which are involved in EMT as positive regulators, can be transcriptionally controlled by Slug. This was demonstrated by Lambertini et al (327), who showed that Slug silencing significantly decreased the level of miR-221, strongly suggesting that miR-221 is a Slug target gene. This was further confirmed by the characterization of a specific region of the miR-221 promoter that is transcriptionally active and is bound by the transcription factor Slug in vivo.

On the other hand, various miRNAs have been reported to directly target EMT-promoting transcription factors. For instance Qiu et al (328) found that miR-139-5p functions as a suppressor of EMT in HCC and metastasis by targeting ZEB1 and ZEB2, and that it may be a therapeutic target for metastatic HCC. In conclusion, miRNAs targeting and miRNA mimicking strategies are both expected to be suitable for the control of EMT.

A very important step in tumor dissemination and metastasis is neoangiogenesis. This is a very complex process in which several proteins and protein networks participate, for instance interleukin (IL)-8, vascular endothelial growth factor (VEGF), basic fibroblast growth factor (bFGF), angiopoietins and matrix metalloproteinases (MMPs). As far as the expression of the IL-8 gene is concerned, the increase in IL-8 gene expression from the healthy brain to low-grade glioma (LGG) can be explained by alterations in the regulatory networks associated with IL-8 gene transcription. Among these, the nuclear factor-κB (NF-κB) network should be proposed, since i) NF-κB is one of the major transcription factors involved in IL-8 gene regulation (329); ii) NF-κB is a marker of glioma onset and progression (330–333); iii) miR-16 inhibits glioma cell growth through the suppression of the NF-κB signaling pathway (334). In addition to transcription factors, miRNAs can directly modulate pro-angiogenic factors. For instance, the increased IL-8 gene expression in high-grade glioma (HGG; with respect to LGG) may be associated with decrease of its inhibitory miRNA, miR-93, at least in a subset of HGG patients. The decrease in miR-93 expression in these HGG patients, in addition to IL-8, may lead to the post-transcriptional upregulation of VEGF, monocyte chemoattractant protein-1 (MCP-1) and platelet-derived growth factor (PDGF)-bb, well recognized markers of the late tumor stages of gliomas (335–337). However, it should be mentioned that HGG samples are highly heterogeneous with respect to miR-93 levels, suggesting the involvement of multiple regulatory pathways in controlling the level of IL-8 gene expression.

One of the better described examples of tumor suppressor miRNAs is miR-124. This miRNA has been found to play a significant role in several types of cancer (168–173,338). Specifically, miR-124 expression is reportedly downregulated in the cells and tissues of esophageal cancer (339), breast cancer (340), renal cell carcinoma (341) and CRC (172). Accordingly, the ectopic expression of miR-124 by target tumor cells inhibits tumor-related parameters in experimental model systems mimicking prostate cancer, medulloblastoma, hepatocellular carcinoma, gastric cancer, glioma, osteosarcoma and CRC.

For instance, Taniguchi et al (164) recently demonstrated that the ectopic expression of miR-124 induced apoptosis and autophagy in colon cancer cells. In addition, miR-124 was demonstrated to target polypyrimidine tract-binding protein 1 (PTB1), which is a splicer of pyruvate kinase muscles 1 and 2 (PKM1 and PKM2), and to induce the switching of PKM isoform expression from PKM2 to PKM1 (164). In addition to this study, Lu et al (342) demonstrated that miR-124a expression was downregulated in human glioma tissues, and that its expression level negatively correlated with the pathological grade of the glioma. The restoration of miR-124a inhibited glioma cell proliferation and invasion in vitro.

Furthermore, they found that miR-124a directly targeted and suppressed IQ motif containing GTPase activating protein 1 (IQGAP1), a well-known regulator of actin dynamics and cell motility (342). Taken together all these data clearly demonstrate that miR-124a is an important tumor suppressor miRNA which is downregulated in cancer cells; accordingly antitumor effects can be achieved following the administration of miR-124, pre-miR-124 or a variety of miR-124 mimics to cancer cells.

Finally, the translational relevance of the role of miR-124 in antitumor drug sensitivity is suggested by the finding that the increased miR-124 expression correlates with an improved breast cancer prognosis, specifically in patients receiving chemotherapy. This finding suggests that miR-124 may potentially be used as a therapeutic agent to improve the efficacy of chemotherapy, including that based on DNA-damaging agents via ATM interactor (ATMIN)- and poly(ADP-ribose) polymerase 1 (PARP1)-mediated mechanisms (343).

A second example of possible miRNA replacement therapy is based on the inhibition of IL-8 and VEGF by the transfection of tumor target cells with pre-miR-93. This was performed in human glioma cell lines (U251 and T98G), as well as on the SK-N-AS neuroblastoma cell line.

The first conclusion of this research activity is that the miRNA, miR-93, is involved in the control of the expression of the IL-8 gene in the glioma U251 and in the neuroblastoma SK-N-AS cell lines (344,345). The effects of these treatments were analyzed by RT-qPCR (looking at the IL-8 mRNA content) or by Bio-plex analysis (looking at IL-8 protein secretion). In addition, Fabbri et al (344) found that the transfection of target cells with pre-miR-93 led to the downregulation of VEGF (see the results depicted in Fig. 4A), suggesting that, as shown in Fig. 4B, miR-93 has effects on the growth of gliomas [by interfering with growth factors, including PDGF, fibroblast growth factor (FGF), granulocyte-macrophage colony-stimulating factor (GM-CSF) and granulocyte-colony stimulating factor(G-CSF)], as well as on neoangiogenesis.

Gliomas, as other tumors, express miR-221 at high levels, promoting malignant progression through activation of the Akt pathway and the inhibition of p27^Kip1 (346–349). In addition miR-221 mediates the downregulation of other genes, such as PUMA (258), intercellular adhesion molecule 1 (ICAM-1) (350), TIMP metallopeptidase inhibitor 3 (TIMP3) (351) and phosphatase and tensin homolog (PTEN) (352), and may thus be associated with cancer onset and progression (353). Therefore, miR-221 appears to be a specific target for the treatment of gliomas (354,355). Zhang et al (354) reported that the co-suppression of the miR-221/222 cluster suppressed human glioma cell growth by affecting p27^Kip1 expression in vitro and in vivo. In our own laboratory, we have also examined the effects of a PNA against miR-221 and showed that it is able to induce a sharp decrease in miR-221 biological activity. The employed PNA carryed an Arg(8) peptide to facilitate PNA uptake by target cells. Two studies were published on this specific issue. In the first study by Brognara et al (319), we demonstrated that targeting miR-221 induced a sharp increase in the expression of the miR-221 target p27^Kip1 mRNA in a breast cancer cell line (319). In a more recent study of ours, Brognara et al (56) demonstrated that the PNA against miR-221 can be internalized by glioma cells when linked to a Arg(8) tail (R8), leading to the inhibition of miR-221 functions, associated with the increased expression of p27^Kip1 in U251 and T98G cells. In addition, the expression of another miR-221 target gene, TIMP3, was upregulated following treatment of the T98G cells with R8-PNA-a221. These data support the concept that targeting miR-221 with antagomiR molecules may provide novel options for developing protocols for the treatment of gliomas. This is supported by the finding that the treatment of all the glioma cells lines with R8-PNA-a221 induced the activation of the early apoptotic pathway (56).

Several tumors express upregulated levels of several miRNAs, suggesting that a possible limit to anti-mRNA therapeutics may be the requirement of the co-targeting of several miRNAs to obtain the programmed biological effects. Moreover, an important anti-miRNA strategy may be associated with the obvious need for the co-targeting of different miRNAs belonging to the same miRNA family.

One of the most interesting results obtained to date using miRNA therapeutics is the formal demonstration that, when used in combination with antitumor drugs, satisfactory therapeutic effects may be achieved (359). This has been demonstrated using both miRNA mimicking approaches, as well as anti-miRNA molecules.

Xue et al (370) verified the biological activity of novel lung-targeting nanoparticles capable of delivering miRNA mimics and siRNAs to lung adenocarcinoma cells in vitro and to tumors in a genetically engineered mouse model of lung cancer based on the activation of oncogenic Kirsten rat sarcoma viral oncogene homolog (Kras) and the loss of p53 function. The therapeutic delivery of miR-34a, a p53-regulated tumor suppressor miRNA, restored the miR-34a levels in lung tumors, specifically downregulated miR-34a target genes, and attenuated tumor growth. The delivery of siRNAs targeting Kras reduced Kras gene expression and MAPK signaling, increased apoptosis and inhibited tumor growth. The combination of miR-34a and siRNA targeting Kras improved the therapeutic responses as compared to those observed with either small RNA alone, leading to tumor regression. Furthermore, nanoparticle-mediated small RNA delivery plus conventional, cisplatin-based chemotherapy prolonged survival in this model compared to chemotherapy alone. These findings demonstrate that RNA combination therapy is possible in a model of lung cancer and provide preclinical support for the use of small RNA therapies in patients who have cancer (370). A second example is that published by Nishimura et al (371) who first demonstrated that the siRNA-mediated silencing of EphA2, an ovarian cancer oncogene, resulted in the reduction of tumor growth. Second, they presented evidence that the additional inhibition of EphA2 by an miRNA further ‘boosts’ its antitumor effects. They identified miR-520d-3p as a tumor suppressor upstream of EphA2. The restoration of miR-520d-3p prominently decreased EphA2 protein levels, and suppressed tumor growth and migration/invasion both in vitro and in vivo. The dual inhibition of EphA2 in vivo using nanoliposomes loaded with miR-520d-3p and EphA2 siRNA exhibited synergistic antitumor efficiency and greater therapeutic efficacy than either monotherapy alone. These data emphasize the feasibility of combined miRNA-siRNA therapy, and will have broad implications for innovative gene silencing therapies for cancer and other diseases.

A further example in this very exciting field of investigation was reported by Hu et al (372), studying Bcl-2, a prominent member of the Bcl-2 family of proteins that regulate the induction of apoptosis. They investigated the effect of Bcl-2 siRNAs combined with miR-15a oligonucleotides on the growth of Raji cells. Following transfection of these combined reagents, the protein and mRNA levels of Bcl-2 were markedly decreased. The growth of the cells was significantly inhibited compared with the cells transfected with Bcl-2 siRNA or miR-15a alone and the apoptotic rate significantly increased. These results suggest that the combination of Bcl-2 siRNA and miR-15a oligonucleotides increases the apoptosis of Raji cells, and strongly support the concept that the combination of Bcl-2 siRNA and miR-15a may be a useful approach in the treatment of lymphoma.

Finally, an example of possible combined treatment is shown in Fig. 6, which indicates that the co-treatment of U251 cells with PNAs targeting miR-221 or miR-222 in the presence of pre-miR-124 transfection leads to a much higher level of apoptosis as opposed to singularly administered reagents (Fabbri et al, unpublished data).

MicroRNA therapeutics in cancer are based on targeting or mimicking miRNAs involved in cancer onset, progression, angiogenesis, EMT and metastasis. This strategy has been proposed several years ago and is based on the well-recognized fact that miRNAs play a key role in the post-transcriptional control of gene expression by the sequence-selective targeting of mRNAs and are key players in several biological functions and pathological processes, including cancer. In this respect, several studies have conclusively demonstrated that miRNAs are deeply involved in tumor onset and progression, either behaving as tumor-promoting miRNAs (oncomiRNAs and metastamiRNAs) or as tumor suppressor miRNAs. In general, miRNAs able to promote cancer target mRNAs coding for tumor suppressor proteins, whereas miRNAs exhibiting tumor suppressor properties usually target mRNAs coding oncoproteins. This has a very important implication in diagnosis and/or prognosis, including the recent discovery that the pattern of circulating cell-free miRNAs in serum allows us to perform molecular analyses on these non-invasive liquid biopsies. This research field has confirmed that cancer-specific miRNAs are present in extracellular body fluids, and may play a very important role in the crosstalk between cancer cells and surrounding normal cells. Interestingly, the evidence of the presence of miRNAs in serum, plasma and saliva supports their potential as an additional set of biomarkers for cancer.

This review has focused on the most promising examples potentially leading to the development of anticancer, miRNA-based therapeutic protocols. The inhibition of miRNA activity can be readily achieved by the use of miRNA inhibitors and oligomers, including RNA, DNA, DNA analogues (miRNA antisense therapy), small molecule inhibitors, miRNA sponges or through miRNA masking. On the contrary, the enhancement of miRNA function (miRNA replacement therapy) can be achieved by the use of modified miRNA mimetics and plasmids or lentiviral vectors carrying miRNA sequences. However, we should carefully consider that a single miRNA can target several mRNAs (not only tumor-associated mRNAs) and a single mRNA may contain in the 3′UTR sequence several signals for miRNA recognition. In this case, antagomiRNA-based therapy should be designed to target multiple miRNAs. MicroRNA targeting and mimicking is further complicated by the facts that, since their discovery and first characterization, the number of miRNA sequences deposited in the miRBase databases is increasing, and research studies on miRNAs in cancer have confirmed the very high complexity of the networks constituted by miRNAs and RNA targets.

One possible approach includes the combination strategies based on the co-administration of anticancer agents, as shown by the observation that i) the combined administration of different antagomiR molecules induces greater antitumor effects and ii) some anti-miR molecules can sensitize drug-resistant tumor cell lines to drug treatment. In this review, we approached two additional issues: i) the combination of miRNA replacement therapy with drug administration and ii) the combination of antagomiR and miRNA replacement therapy. One of the solid results emerging from different independent studies is the demonstration that miRNA replacement therapy can enhance the antitumor effects of the antitumor drugs.

The second important conclusion of the reviewed studies is that the combination of anti-miRNA and miRNA replacement strategies may lead to excellent results, in terms of antitumor effects. This possible combined strategy is in its infancy and very few studies are available in the literature. Proof-of-principle data are presented as examples of possible combined treatments in Fig. 6. Our data indicate that the co-treatment of U251 glioblastoma cells with PNAs targeting miR-221 or miR-222 in the presence of pre-miR-124 transfection leads to a much higher level of apoptosis as opposed to singularly administered reagents. These data further extend the possible combined antitumor treatment based on antitumor drugs and antagomiR-molecules, and present the very novel possibility of combining antagomiR and miRNA replacement therapies.