Rational design of prodrug-type apoB-targeted siRNA for nuclease resistance improvement without compromising gene silencing potency.

Synthetic siRNA molecules without chemical modifications are easily degraded in the body, and 2'-O-modifications are frequently introduced to enhance stability. However, such chemical modifications tend to impact the gene knockdown potency of siRNA negatively. To circumvent this problem, we previously developed a prodrug-type siRNA bearing 2'-O-methyldithiomethyl (MDTM) groups, which can be converted into unmodified siRNA under the reductive environment in cells. In this study, we developed a nuclease-resistant prodrug-type 2'-O-MDTM siRNA for deployment in future animal experiments. To rationally design siRNA modified with a minimal number of 2'-O-MDTM nucleotide residues, we identified the sites susceptible to nuclease digestion and tolerant to 2'-O-methyl (2'-OMe) modification in the antisense strand of apolipoprotein B-targeted siRNA. Subsequently, we optimized the positions where the 2'-OMe and 2'-O-MDTM groups should be incorporated. siRNA bearing the 2'-O-MDTM and 2'-OMe groups at their respective optimized positions exhibited efficient knockdown potency in vitro and enhanced stability in serum.

Prodrug-Type Phosphotriester Oligonucleotides with Linear Disulfide Promoieties Responsive to Reducing Environment.

Various chemical modifications have been developed to create new antisense oligonucleotides (AONs) for clinical applications. Our previously designed prodrug-type phosphotriester-modified oligonucleotide with cyclic disulfides (cyclic SS PTE ON) can be converted into unmodified ON in an intracellular-mimetic reducing environment. However, the conversion rate of the cyclic SS PTE ON was very low, and the AON with cyclic SS PTE modifications showed much weaker antisense activity than corresponding to the fully phosphorothioate-modified AON. In this study, we synthesized several types of PTE ONs containing linear disulfides (linear SS PTE ONs) and evaluated their conversion rates under reducing conditions. From the results, the structural requirements for the conversion of the synthesized linear SS PTE ONs were elucidated. Linear SS PTE ON with promising promoieties showed a nuclease resistance up to 4.8-fold compared to unmodified ON and a cellular uptake by endocytosis without any transfection reagent. In addition, although the knockdown activity of the linear SS PTE gapmer AON is weaker than that of the fully phosphorothioate-modified gapmer AON, the knockdown activity is slightly stronger than that of the cyclic SS PTE gapmer AON. These results suggest that the conversion rates may be related to the expression of the antisense activity.

Effective gene silencing activity of prodrug-type 2'-O-methyldithiomethyl siRNA compared with non-prodrug-type 2'-O-methyl siRNA.

Small interfering RNAs (siRNAs) are an active agent to induce gene silencing and they have been studied for becoming a biological and therapeutic tool. Various 2'-O-modified RNAs have been extensively studied to improve the nuclease resistance. However, the 2'-O-modified siRNA activities were often decreased by modification, since the bulky 2'-O-modifications inhibit to form a RNA-induced silencing complex (RISC). We developed novel prodrug-type 2'-O-methyldithiomethyl (MDTM) siRNA, which is converted into natural siRNA in an intracellular reducing environment. Prodrug-type 2'-O-MDTM siRNAs modified at the 5'-end side including 5'-end nucleotide and the seed region of the antisense strand exhibited much stronger gene silencing effect than non-prodrug-type 2'-O-methyl (2'-O-Me) siRNAs. Furthermore, the resistances for nuclease digestion of siRNAs were actually enhanced by 2'-O-MDTM modifications. Our results indicate that 2'-O-MDTM modifications improve the stability of siRNA in serum and they are able to be introduced at any positions of siRNA.

Triazole- and Tetrazole-Bridged Nucleic Acids: Synthesis, Duplex Stability, Nuclease Resistance, and in Vitro and in Vivo Antisense Potency.

Antisense oligonucleotides are attractive therapeutic agents for several types of disease. One of the most promising modifications of antisense oligonucleotides is the introduction of bridged nucleic acids. As we report here, we designed novel bridged nucleic acids, triazole-bridged nucleic acid (TrNA), and tetrazole-bridged nucleic acid (TeNA), whose sugar conformations are restricted to N-type by heteroaromatic ring-bridged structures. We then successfully synthesized TrNA and TeNA and introduced these monomers into oligonucleotides. In UV-melting experiments, TrNA-modified oligonucleotides exhibited increased binding affinity toward complementary RNA and decreased binding affinity toward complementary DNA, although TeNA-modified oligonucleotides were decomposed under the annealing conditions. Enzymatic degradation experiments demonstrated that introduction of TrNA at the 3'-terminus rendered oligonucleotides resistant to nuclease digestion. Furthermore, we tested the silencing potencies of TrNA-modified antisense oligonucleotides using in vitro and in vivo assays. These experiments revealed that TrNA-modified antisense oligonucleotides induced potent downregulation of gene expression in liver. In addition, TrNA-modified antisense oligonucleotides showed a tendency for increased liver biodistribution. Taken together, our findings indicate that TrNA is a good candidate for practical application in antisense methodology.

Synthesis, duplex-forming ability, enzymatic stability, and in vitro antisense potency of oligonucleotides including 2'-C,4'-C-ethyleneoxy-bridged thymidine derivatives.

We synthesized thymidine derivatives of 2'-C,4'-C-ethyleneoxy-bridged 2'-deoxyribonucleic acids with an 8'-methyl group ((R)-Me-EoDNA and (S)-Me-EoDNA) and without any substituent (EoDNA). Oligonucleotides including these EoDNAs showed high hybridization abilities with complementary RNA and excellent enzymatic stabilities compared with natural DNA. Moreover, the in vitro antisense potency of oligonucleotides with these EoDNAs and our recently reported methylene-EoDNAs was investigated and compared with that of LNA, which is a practical chemical modification for oligonucleotide-therapeutic agents. The results showed that EoDNAs and methylene-EoDNAs could be promising candidates for antisense technology.

Ca2+ enrichment in culture medium potentiates effect of oligonucleotides.

Antisense and RNAi-related oligonucleotides have gained attention as laboratory tools and therapeutic agents based on their ability to manipulate biological events in vitro and in vivo. We show that Ca(2+) enrichment of medium (CEM) potentiates the in vitro activity of multiple types of oligonucleotides, independent of their net charge and modifications, in various cells. In addition, CEM reflects in vivo silencing activity more consistently than conventional transfection methods. Microscopic analysis reveals that CEM provides a subcellular localization pattern of oligonucleotides resembling that obtained by unassisted transfection, but with quantitative improvement. Highly monodispersed nanoparticles ~100 nm in size are found in Ca(2+)-enriched serum-containing medium regardless of the presence or absence of oligonucleotides. Transmission electron microscopy analysis reveals that the 100-nm particles are in fact an ensemble of much smaller nanoparticles (ϕ ∼ 15 nm). The presence of these nanoparticles is critical for the efficient uptake of various oligonucleotides. In contrast, CEM is ineffective for plasmids, which are readily transfected via the conventional calcium phosphate method. Collectively, CEM enables a more accurate prediction of the systemic activity of therapeutic oligonucleotides, while enhancing the broad usability of oligonucleotides in the laboratory.

Effect of an N-substituent in sulfonamide-bridged nucleic acid (SuNA) on hybridization ability and duplex structure.

A sulfonamide-bridged nucleic acid without an N-substituent (SuNA[NH]) was successfully synthesized. A comparison of the SuNA[NMe]- and SuNA[NH]-modified oligonucleotides revealed that the duplex-forming abilities of the SuNA[NMe]-modified oligonucleotides with complementary DNA and RNA were higher than those of the SuNA[NH]-modified oligonucleotides. The crystal structures of DNA duplexes containing a SuNA[NR] revealed that the helical structures of the two duplexes and hydration patterns around the bridge moiety were different. These results provide insights into hydration patterns and rationale for the high RNA affinity of SuNA-modified oligonucleotides.

Serial incorporation of a monovalent GalNAc phosphoramidite unit into hepatocyte-targeting antisense oligonucleotides.

The targeting of abundant hepatic asialoglycoprotein receptors (ASGPR) with trivalent N-acetylgalactosamine (GalNAc) is a reliable strategy for efficiently delivering antisense oligonucleotides (ASOs) to the liver. We here experimentally demonstrate the high systemic potential of the synthetically-accessible, phosphodiester-linked monovalent GalNAc unit when tethered to the 5'-terminus of well-characterised 2',4'-bridged nucleic acid (also known as locked nucleic acid)-modified apolipoprotein B-targeting ASO via a bio-labile linker. Quantitative analysis of the hepatic disposition of the ASOs revealed that phosphodiester is preferable to phosphorothioate as an interunit linkage in terms of ASGPR binding of the GalNAc moiety, as well as the subcellular behavior of the ASO. The flexibility of this monomeric unit was demonstrated by attaching up to 5 GalNAc units in a serial manner and showing that knockdown activity improves as the number of GalNAc units increases. Our study suggests the structural requirements for efficient hepatocellular targeting using monovalent GalNAc and could contribute to a new molecular design for suitably modifying ASO.

Amido-bridged nucleic acids with small hydrophobic residues enhance hepatic tropism of antisense oligonucleotides in vivo.

High scalability of a novel bicyclic nucleoside building block, amido-bridged nucleic acid (AmNA), to diversify pharmacokinetic properties of therapeutic antisense oligonucleotides is described. N2'-functionalization of AmNA with a variety of hydrophobic groups is straightforward. Combinations of these modules display similar antisense knockdown effects and improve cellular uptake, relative to sequence-matched conventional 2',4'-bridged nucleic acid (2',4'-BNA) in vivo.

Highly Potent GalNAc-Conjugated Tiny LNA Anti-miRNA-122 Antisense Oligonucleotides.

The development of clinically relevant anti-microRNA antisense oligonucleotides (anti-miRNA ASOs) remains a major challenge. One promising configuration of anti-miRNA ASOs called "tiny LNA (tiny Locked Nucleic Acid)" is an unusually small (~8-mer), highly chemically modified anti-miRNA ASO with high activity and specificity. Within this platform, we achieved a great enhancement of the in vivo activity of miRNA-122-targeting tiny LNA by developing a series of N-acetylgalactosamine (GalNAc)-conjugated tiny LNAs. Specifically, the median effective dose (ED50) of the most potent construct, tL-5G3, was estimated to be ~12 nmol/kg, which is ~300-500 times more potent than the original unconjugated tiny LNA. Through in vivo/ex vivo imaging studies, we have confirmed that the major advantage of GalNAc over tiny LNAs can be ascribed to the improvement of their originally poor pharmacokinetics. We also showed that the GalNAc ligand should be introduced into its 5' terminus rather than its 3' end via a biolabile phosphodiester bond. This result suggests that tiny LNA can unexpectedly be recognized by endogenous nucleases and is required to be digested to liberate the parent tiny LNA at an appropriate time in the body. We believe that our strategy will pave the way for the clinical application of miRNA-targeting small ASO therapy.

Drug discovery and development scheme for liver-targeting bridged nucleic acid antisense oligonucleotides.

Antisense oligonucleotides (ASOs) containing bridged nucleic acids (BNAs) have been proven to be very powerful. However, ensuring a reliable discovery and translational development scheme for this class of ASOs with wider therapeutic windows remains a fundamental challenge. We here demonstrate the robustness of our scheme in the context of the selection of ASOs having two different BNA chemistries (2,'4'-BNA/locked nucleic acid [LNA] and amido-bridged nucleic acid [AmNA]) targeting human proprotein convertase subtilisin/kexin type 9 (PCSK9). The scheme features a two-step process, including (1) a unique and sensitive in vitro screening approach, called Ca2+ enrichment of medium (CEM) transfection, and (2) a ligand-targeted drug delivery approach to better reach target tissues, averting unintended accumulation of ASOs. Using CEM screening, we identified a candidate ASO that shows >70% cholesterol-lowering action in monkeys. An N-acetylgalactosamine (GalNAc) ligand then was appended to the candidate ASO to further broaden the therapeutic margin by altering the molecule's pharmacokinetics. The GalNAc conjugate, HsPCSK9-1811-LNA, was found to be at least ten times more potent in non-human primates (compared with the unconjugated counterpart), with reduced nephrotoxicity in rats. Overall, we successfully showed that our drug development scheme is better suited for selecting clinically relevant BNA-based ASOs, especially for the treatment of liver-associated diseases.

Programmed Instability of Ligand Conjugation Manifold for Efficient Hepatocyte Delivery of Therapeutic Oligonucleotides.

Ligand-targeted drug delivery (LTDD) has gained more attention in the field of nucleic acid therapeutics. To further elicit the potential of therapeutic oligonucleotides by means of LTDD, we newly developed (R)- and (S)-3-amino-1,2-propanediol (APD) manifold for ligand conjugation. N-acetylgalactosamine (GalNAc)/asialoglycoprotein receptor (ASGPr) system has been shown to be a powerful and robust paradigm of LTDD. Our novel APD-based GalNAc (GalNAcAPD) was shown to have intrinsic chemical instability that could play a role in better manipulation of active drug release. The APD manifold also enables facile production of conjugates through an on-support ligand cluster synthesis. We showed in a series of in vivo studies that while the knockdown activity of antisense oligonucleotides (ASOs) bearing 5'-GalNAcAPD was comparable to the conventional hydroxy-L-prolinol-linked GalNAc (GalNAcHP), 3'-GalNAcAPD elicited ASO activity by more than twice as much as the conventional 3'-GalNAcHP. This was ascribed partly to the GalNAcAPD's ideal susceptibility to nucleolytic digestion, which is expected to facilitate cytosolic internalization of ASO drugs. Moreover, an in vivo/ex vivo imaging study visualized the enhancement effect of monoantennary GalNAcAPD on liver localization of ASOs. This versatile manifold with chemical and biological instability would benefit therapeutic oligonucleotides that target both the liver and extrahepatic tissues.

Calcium-Mediated In Vitro Transfection Technique of Oligonucleotides with Broad Chemical Modification Compatibility.

Oligonucleotide drugs (ODs) have gained increasing attention owing to their promising therapeutic potential. One major obstacle that ODs have been facing is the lack of appropriate in vitro validation systems that can predict in vivo activity and toxicity. We have devised a transfection method called CEM (Ca2+-enrichment method), where the simple enrichment of calcium ion with calcium chloride in culture medium potentiates the activity of various types of naked oligonucleotides including gapmers, siRNA, and phosphorodiamidate morpholino antisense oligonucleotides (PMO) in many cultured cell lines with limited cytotoxicity. We here describe a precise procedure of the method. Besides the benefit of the CEM's predictive power to accurately estimate in vivo activity of ODs of your interest in drug discovery and development settings, this cost-efficient, easy-to-access method can be a robust laboratory technique to modulate gene expressions with ODs with a variety of mechanisms of action.

Effect of modular conjugation strategy for N-acetylgalactosamine-targeted antisense oligonucleotides.

The asialoglycoprotein receptor (ASGPr) and N-acetylgalactosamine (GalNAc) is one of the most reliable receptor-ligand combinations for delivering antisense oligonucleotides (ASOs) to the liver. Here, we show that a modular GalNAc conjugation strategy allows us to reinforce the activity of the parent, naked 2',4'-BNA/LNA gapmer targeting apolipoprotein B. The conjugation partly reduced a possible hepatotoxicity of the parent ASO. The structure-activity study revealed the significance of the metabolic susceptibility of the GalNAc moiety to nucleolytic cleavage that results in exposure of the parent gapmer. The broad usefulness of our delivery strategy of ASOs to the liver has been demonstrated.

Cholesterol-GalNAc Dual Conjugation Strategy for Reducing Renal Distribution of Antisense Oligonucleotides.

Recently, some studies have reported nephrotoxicity associated with a certain class of antisense oligonucleotides (ASOs) in humans. One possibility for reducing the potential nephrotoxicity of ASOs is to alter their pharmacokinetics. In this study, we investigated the effect of a ligand conjugation strategy on the renal accumulation of ASOs. We selected two ligands, cholesterol and N-acetylgalactosamine (GalNAc), with the purpose of reducing renal distribution and liver targeting, and then designed a series of cholesterol-GalNAc dual conjugated ASOs. The gene-silencing activity of the cholesterol-GalNAc dual conjugated ASO in the liver was slightly lower than that of a GalNAc-conjugated ASO. On the other hand, the renal distribution of the cholesterol-GalNAc dual conjugated ASO was considerably decreased compared with the GalNAc-conjugated ASO, as we expected. As dual conjugation was successful in reducing the renal distribution of ASO, it should be an effective strategy for reducing the nephrotoxic potential of ASOs.

Antisense Oligonucleotides Targeting Y-Box Binding Protein-1 Inhibit Tumor Angiogenesis by Downregulating Bcl-xL-VEGFR2/-Tie Axes.

Y-box binding protein-1 (YB-1), involved in cancer progression and chemoradiation resistance, is overexpressed in not only cancer cells but also tumor blood vessels. In this study, we investigated the potential value of amido-bridged nucleic acid (AmNA)-modified antisense oligonucleotides (ASOs) targeting YB-1 (YB-1 ASOA) as an antiangiogenic cancer therapy. YB-1 ASOA was superior to natural DNA-based ASO or locked nucleic acid (LNA)-modified YB-1 ASO in both knockdown efficiency and safety, the latter assessed by liver function. YB-1 ASOA administered i.v. significantly inhibited YB-1 expression in CD31-positive angiogenic endothelial cells, but not in cancer cells, in the tumors. With regard to the mechanism of its antiangiogenic effects, YB-1 ASOA downregulated both Bcl-xL/VEGFR2 and Bcl-xL/Tie signal axes, which are key regulators of angiogenesis, and induced apoptosis in vascular endothelial cells. In the xenograft tumor model that had low sensitivity to anti-VEGF antibody, YB-1 ASOA significantly suppressed tumor growth; not only VEGFR2 but also Tie2 expression was decreased in tumor vessels. In conclusion, YB-1/Bcl-xL/VEGFR2 and YB-1/Bcl-xL/Tie signal axes play pivotal roles in tumor angiogenesis, and YB-1 ASOA may be feasible as an antiangiogenic therapy for solid tumors.

Development of Antisense Drugs for Dyslipidemia.

Abnormal elevation of low-density lipoprotein (LDL) and triglyceride-rich lipoproteins in plasma as well as dysfunction of anti-atherogenic high-density lipoprotein (HDL) have both been recognized as essential components of the pathogenesis of atherosclerosis and are classified as dyslipidemia. This review describes the arc of development of antisense oligonucleotides for the treatment of dyslipidemia. Chemically-armed antisense candidates can act on various kinds of transcripts, including mRNA and miRNA, via several different endogenous antisense mechanisms, and have exhibited potent systemic anti-dyslipidemic effects. Here, we present specific cutting-edge technologies have recently been brought into antisense strategies, and describe how they have improved the potency of antisense drugs in regard to pharmacokinetics and pharmacodynamics. In addition, we discuss perspectives for the use of armed antisense oligonucleotides as new clinical options for dyslipidemia, in the light of outcomes of recent clinical trials and safety concerns indicated by several clinical and preclinical studies.

Evaluation of the effects of chemically different linkers on hepatic accumulations, cell tropism and gene silencing ability of cholesterol-conjugated antisense oligonucleotides.

Cholesterol conjugation of oligonucleotides is an attractive way to deliver the oligonucleotides specifically to the liver. However cholesterol-conjugated antisense oligonucleotides (ASOs) mainly accumulate in non-parenchymal cells (NPCs) such as Kupffer cells. In this study, to increase the hepatic accumulation of cholesterol-conjugated ASOs, we prepared a variety of linkers for cholesterol conjugation to anti-Pcsk9 ASOs and examined their effects on pharmacological parameters. Hepatic accumulation of ASO was dramatically increased with cholesterol conjugation. The increase in hepatic accumulation depended largely on the linker chemistry of each cholesterol-conjugated ASO. In addition to hepatic accumulation, the cell tropism of each cholesterol-conjugated ASO tended to depend on their linker. Although a linker bearing a disulfide bond accumulated mainly in NPCs, hexamethylene succinimide linker accumulated mainly in hepatocytes. To estimate the benefits of releasing ASO from the conjugated cholesterol in hepatocyte, we designed another linker based on hexamethylene succinimide, which has a phosphodiester bond between the linker and the ASO. The cholesterol-conjugated ASO bearing such a phosphodiester bond showed a significantly improved Pcsk9 mRNA inhibitory effect compared to its counterpart, cholesterol-conjugated ASO with a phosphorothioate bond, while the hepatic accumulation of both cholesterol-conjugated ASOs was comparable, indicating the effectiveness of removing the conjugated cholesterol for ASO activity. In toxicity analysis, some of the linkers induced lethal toxicities when they were injected at high concentrations (>600µM). These toxicities were attributed to decreased platelet levels in the blood, suggesting an interaction between cholesterol-conjugated ASO and platelets. Our findings may provide a guideline for the design of molecule-conjugated ASOs.

Locked nucleic acid antisense inhibitor targeting apolipoprotein C-III efficiently and preferentially removes triglyceride from large very low-density lipoprotein particles in murine plasma.

A 20-mer phosphorothioate antisense oligodeoxyribonucleotide having locked nucleic acids (LNA-AON) was used to reduce elevated serum triglyceride levels in mice. We repeatedly administered LNA-AON, which targets murine apolipoprotein C-III mRNA, to high-fat-fed C57Bl/6J male mice for 2 weeks. The LNA-AON showed efficient dose-dependent reductions in hepatic apolipoprotein C-III mRNA and decreased serum apolipoprotein C-III protein concentrations, along with efficient dose-dependent reductions in serum triglyceride concentrations and attenuation of fat accumulation in the liver. Through precise lipoprotein profiling analysis of sera, we found that serum reductions in triglyceride and cholesterol levels were largely a result of decreased serum very low-density lipoprotein (VLDL)-triglycerides and -cholesterol. It is noteworthy that larger VLDL particles were more susceptible to removal from blood than smaller particles, resulting in a shift in particle size distribution to smaller diameters. Histopathologically, fatty changes were markedly reduced in antisense-treated mice, while moderate granular degeneration was frequently seen the highest dose of LNA-AON. The observed granular degeneration of hepatocytes may be associated with moderate elevation in the levels of serum transaminases. In conclusion, we developed an LNA-based selective inhibitor of apolipoprotein C-III. Although it remains necessary to eliminate its potential hepatotoxicity, the present LNA-AON will be helpful for further elucidating the molecular biology of apolipoprotein C-III.

Evaluation of multiple-turnover capability of locked nucleic acid antisense oligonucleotides in cell-free RNase H-mediated antisense reaction and in mice.

The multiple-turnover ability of a series of locked nucleic acid (LNA)-based antisense oligonucleotides (AONs) in the RNase H-mediated scission reaction was estimated using a newly developed cell-free reaction system. We determined the initial reaction rates of AONs under multiple-turnover conditions and found that among 24 AONs tested, AONs with melting temperatures (Tm) of 40°C-60°C efficiently elicit multiple rounds of RNA scission. On the other hand, by measuring Tm with two 10-mer RNAs partially complementary to AONs as models of cleaved 5' and 3' fragments of mRNA, we found that AONs require adequate binding affinity for efficient turnover activities. We further demonstrated that the efficacy of a set of 13-mer AONs in mice correlated with their turnover efficiency, indicating that the intracellular situation where AONs function is similar to multiple-turnover conditions. Our methodology and findings may provide an opportunity to shed light on a previously unknown antisense mechanism, leading to further improvement of the activity and safety profiles of AONs.