How to Train Your Dragon: Targeted Delivery of MicroRNA to Cancer Cells In Vivo

The fundamental role of microRNAs (miRNAs) in both initiation and progression of cancer has made them attractive therapeutic targets.¹ miRNAs can simultaneously alter expression of multiple target genes and often disrupt entire signaling networks, resulting in efficient changes in the activity and phenotype of target cells. At the same time, their broad functionality can be a liability that results in difficult-to-control off-target effects and toxicities. A report by Esposito et al. in this issue of Molecular Therapy provides a possible solution to the problem of cell-specific delivery of therapeutic miRNAs and represents an important step in the development of targeted miRNA mimics.² The authors present a novel strategy for the delivery of tumor suppressor let-7g miRNA to specific cancer cells using unformulated oligonucleotides in the form of an aptamer specific for the oncogenic tyrosine kinase receptor, Axl (Figure 1).²

Figure 1.

Novel aptamer–miRNA conjugates and existing strategies for miRNA delivery. The aptamer–micro-RNA (miRNA) molecules (GL21.T-let) are internalized in a cell-specific manner through an oncogenic Axl receptor on the surface of certain cancer cells. By contrast, miRNAs encapsulated in nanoparticles or transferred by viral vectors have limited cellular specificity, which may increase their toxicity and off-target effects. AAV, adeno-associated virus; AV, adenovirus; LV, lentivirus.

miRNAs are small, noncoding RNA molecules acting as negative regulators of gene expression. They bind to partially complementary sequences in the 3′ untranslated regions of target mRNAs, thereby inducing their degradation and/or blocking their translation. Lentiviruses (LVs), adeno­viruses (AVs), and adeno-associated viruses (AAVs) can transfer miRNA sequences,³ which after viral entry into target cells are transcribed and processed into mature miRNA (Figure 1). Despite efficient cellular delivery, viral vectors raise concerns over safety of genomic integration of LVs, which may trigger expression of oncogenes, or excessive immuno­genicity and the transient nature of miRNA expression in the case of AVs and AAVs. Nonviral methods of miRNA delivery instead rely on lipid and polymeric nanoparticles to protect the miRNA from degradation by nucleases and increase their half-life in the circulation. The first miRNA replacement therapy to reach clinical testing is based on the delivery of a tumor suppressor miR-34 mimic in customized liposomes (MRX34, Mirna Therapeutics). The smaller size of miRNA inhibitors and their greater tolerance to chemical modification facilitated the development of naked antagomirs (antisense miRNA). The miR122 inhibitors for treatment of chronic hepatitis C are either single locked nucleic acid molecules, such as Miravirsen (Santaris Pharma), or conjugates of antagomirs to a hepatocyte-specific ligand acting as a targeting moiety, such as GalNac-miR122 (Regulus Therapeutics).⁴ The latter reagent was the first to allow receptor-targeted delivery of miRNA inhibitor specifically into liver cells and recently advanced to phase I clinical trials.

Let-7g belongs to a family of miRNAs known for targeting multiple oncogenes and tumor promoters in solid tumors, such as c-Myc, K-Ras, N-Ras, HMGA2 (ref. 5), as well as STAT3 and NF-κB.⁶ To enable uptake, a let-7g miRNA mimic was linked to a previously generated and characterized RNA aptamer (GL21.T) that binds with high affinity to the oncogenic tyrosine kinase receptor, Axl, which is expressed by various types of solid tumors.⁷ Importantly, the GL21.T aptamer used in this study acted not only as a targeting moiety but also as an Axl antagonist. The resulting aptamer–miRNA conjugate (GL21.T-let) retained the cellular specificity, binding affinity, and inhibitory function of the original GL21.T. To ensure high efficacy of gene targeting, the investigators used an extended sequence of the let-7g miRNA to facilitate processing by Dicer endoribonuclease and loading into the RNA processing machinery.⁸ The refined design of GL21.T-let is likely to allow for dissociation of the diced let-7g miRNA from the ligand–receptor complex, as shown before for cell-targeted Dicer–substrate small interfering RNA (siRNA) conjugates.⁹ In fact, deep sequencing analysis and immunoprecipitation assays using GL21.T-let–treated Axl-positive target cells (A549) confirmed proper Dicer processing of the miRNA mimic with guide strand preference and Ago2 recruitment. As assessed by quantitative real-time PCR, the GL21.T-let conjugate efficiently reduced HMGA2 and N-RAS expression in a Dicer-dependent manner. Blockade of target gene expression was a result of Axl receptor-mediated internalization of the conjugate. The authors demonstrated that Axl-negative MCF7 cancer cells failed to internalize GL21.T-let unless the receptor was forcibly expressed, which resulted in intracellular uptake of the let-7g mimic and target gene silencing. As mentioned earlier, the RNA aptamer part of the conjugate retained its potent and direct inhibitory effect on oncogenic functions of the Axl receptor.

To assess potential synergistic or additive effects of the let-7g miRNA mimic together with antagonism of Axl, the authors performed genome-wide expression analysis. A little surprisingly, many of the genes differentially expressed after treatment with the miRNA mimic alone, RNA aptamer alone, and the GL21.T-let conjugate were identical. This may indicate potential functional redundancy between Axl and let-7g. In fact, downstream signaling from the Axl receptor and let-7 miRNAs converges on STAT3 and NF-κB transcription factors, which play an essential role in cell transformation.^6,10 It is possible that combining the Axl-specific aptamer with more functionally distinct miRNA mimics could produce even more pronounced therapeutic effects. Potential candidates include miR-15 and miR-16, which repress BCL2 and MCL1 survival genes,¹¹ or miR-29, which targets epigenetic regulators such as DNA methyltransferases.¹²

The crucial part of this study was to validate the therapeutic effect of GL21.T-let using xenotransplanted human tumors in immunodeficient mice. Nonspecific liver accumulation and rapid kidney clearance of oligonucleotide therapeutics after systemic administration limit their half-life and thereby their therapeutic efficacy. Pioneering work by one of the coauthors demonstrated that chemically modified RNA aptamers can be utilized for siRNA delivery to tumors in vivo.¹³ The GL21.T-let conjugates generated by Esposito et al. were similarly chemically modified to enable intravenous injections. For the ultimate test of reagent efficacy and target specificity, the investigators simultaneously implanted mice with both Axl-positive (A549) and Axl-negative (MCF7) tumors. The systemic administration of GL21.T-let conjugate resulted in the specific delivery of let-7g miRNA to Axl-positive but not to the Axl-negative tumors. In addition, the miRNA mimic was not detectable in liver or spleen, thereby underscoring the remarkable specificity of the miRNA delivery to target cell populations. Further studies should provide more information on tissue distribution and pharmacokinetic properties of the GL21.T-let conjugate. The efficient miRNA delivery correlated closely with reduced expression of target genes in the Axl-positive tumors. Furthermore, repeated treatment (three times a week for 2 weeks) with GL21.T-let inhibited the growth of A549 tumors with no effect on the Axl-negative MCF7 tumors. As suggested by the global gene expression analysis in vitro, the direct inhibitory function of the GL21.T RNA aptamer alone contributed to the combined antitumor effect of the conjugate. The GL21.T-let conjugates were not obviously immunogenic and did not upregulate interferon-dependent genes in liver or spleen. Follow-up studies should evaluate the effect of these oligonucleotides in more complex syngeneic tumor models in immunocompetent mice.

These findings offer a blueprint for the design of cell-specific aptamers for therapy of cancer and other diseases. New or previously selected RNA aptamers could be utilized for delivery of therapeutic miRNAs into specific cancer cells or into various immune cell lineages, provided that the target receptor is efficiently internalized after binding to the aptamer–miRNA conjugate.^13,14,15 The maximum therapeutic efficacy of aptamer–miRNA conjugates will also require empirical testing and selection of designs with optimal functional synergy between both parts of the molecule. The cellular specificity of the aptamer–miRNA is likely to reduce the dose required for pharmacological effect as well as off-target effects and toxicities. Further preclinical studies should verify the in vivo efficacy of this promising approach, which is highly likely to inspire other attempts to harness miRNA for treatment of cancer or other diseases.