Current status of miRNA-targeting therapeutics and preclinical studies against gastroenterological carcinoma

Expanding knowledge about the crucial roles of microRNAs (miRNAs) in human diseases has led to the idea that miRNAs may be novel, promising therapeutic targets against various pathological conditions. The recent success of a human clinical trial using anti-miR-122 oligonucleotides against chronic hepatitis C virus has paved the way for this approach. In this review, we summarize briefly the current status of clinical trials of miRNA-targeting therapy and several representative preclinical trials against hepato-gastrointestinal carcinoma. In addition, we describe the currently available technologies for modification and delivery of oligonucleotides, which are essential in providing efficient, specific and safe approaches to targeting miRNAs.


Introduction
The expression and functional importance of non-coding RNAs (ncRNAs), such as long ncRNAs and microRNAs, in various human diseases has been reported extensively [1][2][3]. Accordingly, clinical applications of ncRNAs are highly anticipated [4].
MicroRNAs (miRNAs) are small non-coding RNAs first discovered in C. elegans [5]. They are now known to be expressed in most organisms from plants to vertebrates [6]. Many miRNAs are functionally important, acting as oncogenes, tumor suppressors and crucial modulators in intracellular pathways [7]. miRNAs are generated from endogenous transcripts through maturation processing. Long primary miRNAs (pri-miRNAs) ranging in size from several hundred nucleotides (nt) to several kilobases are transcribed from the genome [8]. These are processed into stem-loop precursor miRNAs (pre-miRNAs) of~70 nt by a ribonuclease III (RNase III) known as Drosha and DGCR8/Pasha in the nucleus [8][9][10][11]. After these processing steps, pre-miRNAs are transported to the cytoplasm via exportin-5 [12], where they are recognized by an RNA-induced silencing complex (RISC). This complex is composed of the RNase III Dicer, a double-strand RNA binding protein TRBP, and Argonaute2 (Ago2). Pre-miRNAs are cleaved into mature miRNAs of~22 nt by Dicer [13]. The two RNA strands are separated, and the guide strand for the target mRNA remains associated with Ago2. RISC recognizes the target mRNA based on the complementarity between the guide miRNA and the mRNA transcript [14] within the 3′ untranslated region (UTR) [15]; the target mRNA is subsequently degraded or translationally inhibited [14,16], resulting in post-transcriptional gene silencing [14]. While the molecular mechanisms for gene silencing by miRNA-mRNA targeting require further elucidation, some reports have suggested that miRNAs may sequester target mRNAs into P-bodies [17], where mRNA decay occurs through the initiation of rapid deadenylation [18] or translational repression occurs by inhibiting the binding of ribosomes to the 5′ caps of miRNAs [19].
Alterations in miRNA expression levels contribute to the pathogenesis of human malignancies. The changes result from various mechanisms, including deletions, amplifications or mutations at miRNA loci, epigenetic silencing, dysregulation of transcription factors that are related to the transcription of specific miRNAs, environmental factors such as cigarette smoke and infection, and gene polymorphisms [20][21][22]. Describing specific patterns of miRNA expression levels may be useful for diagnosis, prognosis or evaluating therapeutic response, or in miRNA-targeted therapies that repress or facilitate expression of specific miRNAs.
Many excellent reviews have discussed the aberrant expression and potential biological roles of miRNAs in gastroenterological diseases [23][24][25][26][27][28]. Here, we focus on the recent progress of clinical trials of miRNA-target therapies and representative preclinical trials against gastroenterological carcinoma; additionally, we outline the future work required for clinical utilization of miRNAs.

Review
Current clinical applications of miRNA-targeting therapeutics Anti-miR-122 therapy against chronic hepatitis C Experiments in vitro and in vivo have led to the development of potential new therapies targeting miRNAs. While targeting miRNAs in human clinical trials has focused largely on miRNA signatures as biomarkers for the diagnosis, prognosis, or therapeutic response to traditional treatment [22], two human clinical trials have assessed directed miRNA-targeting as therapeutics, according to ClinicalTrials.gov (http://clinicaltrials.gov) ( Table 1). Both trials are related to gastroenterological diseases.
One of the most extensively studied miRNAs is miR-122, an abundant liver-specific miRNA that plays a critical role in liver function, such as fatty acid and cholesterol metabolism, and in the pathophysiology of liver diseases, such as hepatitis C viral (HCV) replication [11,[31][32][33]. Inhibition of miR-122 with locked-nucleic-acid (LNA)-based anti-miR-122 oligonucleotides complementary to miR-122, caused a long-lasting decrease in total plasma cholesterol in mice [31] and in monkeys [34]. Lanford et al. demonstrated that LNA-based anti-miR-122 oligonucleotides led to the longlasting suppression of HCV viremia and improvement of HCV-induced liver pathology in chimpanzees [33]. In these three cases, no LNA-associated toxicity or histopathological changes were found in mice and nonhuman primates after short-term administration of the oligonucleotides [31,33,34]. These reports indicate that LNA-based anti-miRs can achieve efficient silencing of endogenous miRNA function in mammals and primates, which supports the application of anti-miRNA therapy to other human diseases.
These preclinical results led to the development of miravirsen, a LNA-modified DNA phosphorothioate antisense oligonucleotide against miR-122, as the first miRNAtargeting drug for clinical use [29]. It was developed to target HCV since the stability and propagation of HCV is dependent on a functional interaction between the HCV genome and miR-122 [35]. miR-122 binds to two closely spaced target sites in the highly conserved 5′-untranslated region of the HCV genome, thereby forming an oligomeric miR-122-HCV complex that protects the HCV genome from nucleolytic degradation or from host innate immune responses [35]. The miR-122 binding sites are conserved across all HCV genotypes and subtypes [36]. Miravirsen, an LNA-modified DNA phosphorothioate oligonucleotide complementary to miR-122, is thought to hybridize to the 5′ region of mature miR-122, resulting in sequestration and inhibition of miR-122 [29]. Recently, it was reported that Miravirsen also binds to the stem-loop structure of pri-and pre-miR-122 and inhibits both Dicer-and Drosha-mediated processing of miR-122 precursors [30] (Figure 1).
No harmful events were observed in phase I studies of miravirsen in healthy volunteers; therefore, phase II studies were initiated to evaluate the safety and efficacy of miravirsen in 36 patients with chronic HCV genotype 1 infection. The patients were randomly assigned to receive 5-week subcutaneous injections of placebo or doses of miravirsen at 3, 5 or 7 mg per kilogram of body weight over a 29-day period. Patients who received miravirsen showed a dose-dependent reduction in HCV levels, without major adverse events and with no escape mutations in the miR-122 binding sites of the HCV genome [29].
The success of miravirsen is promising, not only as a novel anti-HCV drug, but also as the first trial of miRNAtargeting therapy. However, caution should be used since miR-122 is generally known as a tumor-suppressive miRNA. Downregulation of miR-122 expression in hepatocellular carcinoma (HCC) is associated with a poor prognosis [37][38][39], and liver tumorigenesis is facilitated in mice lacking miR-122 [40,41]. Nonetheless, as an anti-HCV drug, short-term administration of miravirsen with a 4-week regimen was reversible, and the effects of a 12-week regimen were tested from November 2012-May 2013 (ClinicalTrials.gov Identifier: NCT01727934). Although miravirsen also showed promise for decreasing serum cholesterol levels, we cannot conclude that miravirsen remains free of adverse effects for long-term administration until a long-term trial is completed.

Representative preclinical in vivo experiments
Since their discovery, many novel miRNAs have been identified. As of September 2013, 2,578 mature human miRNA sequences were deposited in miRBase, a public repository hosted by the Sanger Institute (Cambridge, UK). miRNAs have diverse biological functions as they can target mRNAs with weak complementarity. Thus, it is not surprising that most miRNAs are associated with facilitating or suppressing tumors via modulating the expression levels of various tumor-related genes. The PubMed database contains nearly 25,000 miRNA-related articles, many of which aimed to identify novel targets of miRNAs, novel miRNAs, or novel functions of miRNAs in oncogenic/ tumor suppressive function or stabilization. The majority of these in vitro studies need to be confirmed in mice or in non-human primates before their use in clinical trials. In the following section, we describe representative in vivo approaches to targeting miRNAs in hepatogastroenterological cancers, which have evaluated the effects on the initiation, proliferation or growth of tumors in transgenic mice or an in vivo xenograft cancer model ( Figure 2 and Table 2).

miR-221 as an oncogenic miRNA
miR-221 is one of the most frequently and consistently upregulated miRNAs in human cancer, including in HCC, pancreatic, colon, stomach, glioblastoma, kidney, bladder, prostate and thyroid cancer, which indicates its importance in tumorigenesis [48]. miR-221 has multiple gene targets, such as the cyclin-dependent kinase inhibitors p27Kip1 (CDKN1B/p27) [54,55] and CDKN1C/p57 [55], the pro-apoptotic protein B-cell lymphoma 2-modifying factor (Bmf) [56], the inhibitor of the phosphoinositide 3kinase pathway phosphatase and tensin homolog (PTEN) [57], the tissue inhibitor of metalloproteinase 3 (TIMP3) [57], the DNA damage-inducible transcript 4 (DDIT4), and a tumor suppressor that modulates the kinase activity of mammalian target of rapamycin (mTOR) [58]. Callegari et al. demonstrated that miR-221 could promote liver tumorigenicity in a transgenic mouse model of miR-221 overexpression in the liver. This model is characterized by the appearance of spontaneous liver tumors in a fraction of male mice and a strong acceleration of tumor development in 100% of mice treated with diethylnitrosamine (DEN). Ten-day old mice received one intraperitoneal injection of DEN, followed 2 months later by a single intravenous dose of anti-miR-221 oligonucleotide (AMO) diluted in saline solution every 15 days, for a total of three injections. Similar to human HCC, tumors in these mice were characterized by an increase in miR-221 expression and a concomitant inhibition of its target protein-coding genes (CDKN1B/p27 and Bmf, not CDKN1C/p57). As expected, mice that received an in vivo intravenous injection of anti-miR-221 oligonucleotides exhibited a significant reduction in the number and size of liver tumor nodules. This study not only shows that miR-221 can promote liver tumorigenicity, but it also establishes a valuable animal model for preclinical investigations of the use of anti-miRNA approaches to liver cancer therapy [48].

Tumor-suppressive miRNAs
Many studies have used the nude mouse xenograft model to evaluate the effects of a specific miRNA on tumorigenesis, tumor growth or metastasis in vivo. In particular, this model is commonly used to assess xenograft growth or progression after transplantation of cancer cells transfected with specific miRNA-expressing plasmids or empty vectors. miR-7 in hepatocellular carcinoma (HCC) cells miR-7 inhibits HCC cell growth and metastasis in vitro and in vivo. Phosphoinositide 3-kinase catalytic subunit delta (PIK3CD) was first identified as a miR-7 target, and further study suggested that miR-7 might be a key regulator of the PI3K/Akt/mTOR signaling pathway. In a xenograft model, overexpressed miR-7 effectively repressed tumor growth and decreased metastasis to the lung. These findings indicate that miR-7 functions as a tumor suppressor and plays a substantial role in inhibiting the tumorigenesis and reversing the metastasis of HCC through the PI3K/Akt/mTOR-signaling pathway. Given these results, miR-7 may be a potential therapeutic or diagnostic/ prognostic target for treating HCC [49].

miR-520 in hepatocellular carcinoma (HCC) cells
The expression levels of miR-520e were decreased dramatically in HCC cells and clinical HCC tissues resulting from DNA hypermethylation in the upstream region of miR-520e locus, whereas silencing of the expression of miR-520e promoted cell proliferation [50]. Introduction of miR-520e suppresses the growth of HCC cells in vitro by targeting NF-κB-inducing kinase (NIK), which is involved  miR-520e HCC NIK HepG2 cells transfected with miR-520e or a negative control were injected subcutaneously into nude mice. The introduction of miR-520e led to a significant reduction in both the size of tumor volume and the frequency of tumor formation. In addition, direct intratumoral injection with miR-520e oligonucleotides repressed the growth of HCC cells in an in vivo xenograft model. [50] miR-375 HCC AEG-1 Overexpression of miR-375 in liver cancer cells decreased cell proliferation, clonogenicity, migration, and invasion, and induced G1 cell cycle arrest and apoptosis. Direct administration of cholesterol-conjugated 2'-O-methyl-modified miR375 mimics significantly affected the growth of HCC xenografts. [51] miR-25 Colon cancer Smad7 In a xenograft model study, stable overexpression of miR25 in colon cancer cells suppressed tumor growth. [52] miR-217 PDCA KRAS Xenograft tumors of PDAC cells were directly injected with miR217-expressing plasmids or a control vector using in vivo-jet PEI. The results from these assays indicated that miR-217 suppresses tumor cell growth in vivo.

[53]
Oncogenic miRNA: miR-221. Tumor suppressive miRNA: miR-7, 520, 375, 25,217. in NIK/ERK/NF-κB signaling. To determine the effect of miR-520e on HCC cell growth in vivo, HepG2 cells transfected with miR-520e or a negative control were injected subcutaneously into nude mice. The introduction of miR-520e led to a significant reduction in both the size of tumor volume and the frequency of tumor formation. In addition, direct intratumoral injection with miR-520e oligonucleotides repressed the growth of HCC cells in an in vivo xenograft model. This finding provides new insight into the mechanism of hepatocarcinogenesis, indicating the therapeutic potential of miR-520e in the treatment of HCC [50].

miR-375 in hepatocellular carcinoma (HCC) cells
He et al. reported that miR-375 targets astrocyte elevated gene-1 (AEG-1) in HCC and suppresses liver cancer cell growth in vitro and in vivo. Overexpression of miR-375 in liver cancer cells decreased cell proliferation, clonogenicity, migration, and invasion, and induced G1 cell cycle arrest and apoptosis. Direct administration of cholesterol-conjugated 2'-O-methyl-modified miR375 mimics significantly affected the growth of HCC xenografts. These findings indicate that miR-375 targets AEG-1 in HCC and suppresses liver cancer cell growth [51].

miR-25 in colon cancer cells
Li et al. reported that miR-25 was significantly downregulated in human colon cancer tissues, and identified Smad7 as its direct target. In a xenograft model study, stable overexpression of miR25 in colon cancer cells suppressed tumor growth [52]. These results suggest that miR-25 functions as a tumor suppressor by targeting Smad7 in colon cancer, suggesting that miR-25 may serve as a potential therapeutic target for colon cancer therapy.

miR-217 in pancreatic cancer cells
In most studies, cells overexpressing miRNAs or miRNA antagonists are used for xenografts and subsequent analyses. However, Zhao et al. reported another method of evaluating the function of miRNAs in vivo. They investigated the biological role of miR-217 in PDAC cells in vitro and in vivo since miR-217 is frequently down-regulated in pancreatic ductal adenocarcinoma (PDCA) [59,60]. KRAS was identified as a direct target of miR-217 and, concordantly, up-regulation of miR-217 decreased KRAS protein expression and subsequently reduced the constitutive phosphorylation of downstream Akt. To confirm the function of miR-217 in vivo, xenograft tumors of PDAC cells were directly injected with miR217expressing plasmids or a control vector using in vivojet PEI (Polyplus Transfection, Illkirch, France). The results from these assays indicated that miR-217 suppresses tumor cell growth in vivo. Therefore, miR-217 may serve as a therapeutic target for miRNA-based PDAC therapy [53].

Development of effective delivery methods for miRNA-targeting oligonucleotides
The development of effective and safe delivery methods of miRNA-targeting molecules is critical to the success of miRNA-targeting drugs. Currently, modified oligonucleotides and various delivery particles are being used for these purposes (Table 3). Sponges can block a whole family of related miRNAs. Selectable marker or reporter gene in the vector allows to isolate a fraction of cells in which the family of miRNAs is strongly inhibited.
"miRNA sponges" are transcripts that contain multiple, tandem binding sites to the miRNAs of interest. When vectors encoding these sponges are transfected into cells, multiple miRNA targets can be suppressed simultaneously [64]; however, further experiments are necessary to demonstrate the utility of the sponge method in vivo.
In terms of delivery methods, adenovirus-associated vectors are used widely for systemic delivery [47,62]. However, while the toxicity of viral-mediated delivery is rarely reported in in vivo studies [62,68], it remains controversial. Novel nanoparticle-based delivery systems, which may be safer and more amenable [47] than viruses, are being developed. As mentioned earlier, the systemic injection of low-molecular-weight PEI/miRNA mimic complexes has proved useful for intratumoral delivery in xenograft models [69]. Additionally, miRNA-expressing plasmids encapsulated in small multilamellar cationic liposomes (DOTAP/ cholesterol), have been reported as a useful delivery method in the mouse xenograft model [70].
AMOs such as LNA are currently the most promising targeted therapy method in terms of safety, stability and prominence. Continuing studies in vitro and in vivo are undoubtedly critical to discovery of simple, efficient, and safe delivery methods; this will facilitate the development of miRNA-targeting therapies for human disease.

Conclusions
Along with recent discoveries of the diverse effects of miRNAs in biological systems, miRNA-mediated intervention is a promising avenue for the development of novel therapeutics against human diseases. In addition to the current success of anti-miR122 therapy against chronic hepatitis C and the ongoing studies of miR-34 mimics against liver cancers in human clinical trials, the results of preclinical studies will likely lead to human clinical trials in the near future. However, several important issues must be addressed if this knowledge is to be used effectively in clinical trials. These include the delivery method, improved oligonucleotide modification for delivery, and safety. Because of its relative novelty, the safety of oligonucleotide therapy is an important consideration. Since miRNAs generally have diverse effects by targeting multiple mRNAs, undesired outcomes, so called "off-target effects", may be encountered even when a specific miRNA is targeted. Effectiveness should also be considered. Although most current trials target individual miRNAs, targeting multiple miRNAs simultaneously may be necessary because most are considered to function cooperatively [73]. Therefore, although our understanding of miRNAs has increased, further research is needed to transform this knowledge into effective therapeutics against human diseases.