Branched Oligonucleotides

Abstract

Provided herein are branched oligonucleotides that exhibit efficient and specific tissue distribution, cellular uptake, minimal immune response, and off-target effects without formulation.

Description

[Technical Field]

[0001] Statement of Federally Funded Research This invention was made with government support under Grant No. 1 R01 GM108803-02 awarded by the National Institutes of Health and with a grant from the CHDI Foundation. The U.S. government has certain rights in this invention.

[0002] Related Applications This application claims priority to U.S. Provisional Patent Application Nos. 62 / 289,268, filed January 31, 2016, and 62 / 317,113, filed April 1, 2016, the contents of which are incorporated herein by reference for all purposes.

[0003] Technical Field The present invention relates to novel branched oligonucleotides designed to achieve unexpectedly high efficacy, uptake and tissue distribution. [Background technology]

[0004] Therapeutic oligonucleotides are simple and effective tools for a variety of applications, including the inhibition of gene function. An example of such inhibition is RNA interference (RNAi). However, the promise of RNAi as a general therapeutic strategy depends on the ability to deliver small RNAs to a wide range of tissues. Currently, small therapeutic RNAs can only be effectively delivered to the liver. There remains a need for self-delivering siRNAs and general therapeutic oligonucleotides that exhibit minimal immune response and off-target effects, efficient cellular uptake without formulation, and efficient and specific tissue distribution. Summary of the Invention

[0005] Thus, the present invention provides branched oligonucleotides ("compounds of the invention") that exhibit unexpected improvements in distribution, in vivo efficacy and safety.

[0006] In a first embodiment, a branched oligonucleotide compound is provided that comprises two or more nucleic acids, the nucleic acids being linked together by one or more moieties selected from a linker, a spacer, and a branch point.

[0007] In some embodiments, the branched oligonucleotide comprises 2, 3, 4, 6, or 8 nucleic acids.

[0008] In branched oligonucleotide embodiments, each nucleic acid is single-stranded and has a 5' end and a 3' end, and each nucleic acid is independently attached to a linker, spacer, or branch point at the 5' end or the 3' end.

[0009] In some embodiments, each single-stranded nucleic acid independently comprises at least 15 consecutive nucleotides. In some embodiments, the antisense strand comprises at least 15, at least 16, at least 17, at least 18, at least 19, or at least 20 consecutive nucleotides and has complementarity to the target.

[0010] In some embodiments, each nucleic acid comprises one or more chemically modified nucleotides. In some embodiments, each nucleic acid consists of chemically modified nucleotides.

[0011] In an embodiment of the branched oligonucleotide, each nucleic acid is double-stranded and comprises a sense strand and an antisense strand, and the sense strand and the antisense strand have a 5'-end and a 3'-end, respectively. In some embodiments, each double-stranded nucleic acid is independently linked to a linker, a spacer, or a branch point at the 3'-end or the 5'-end of the sense strand or the antisense strand.

[0012] In some embodiments, the sense strand and the antisense strand each comprise one or more chemically modified nucleotides. In some embodiments, the sense strand and the antisense strand each comprise chemically modified nucleotides. In some embodiments, the sense strand and the antisense strand both comprise alternating 2'-methoxy-nucleotides and 2'-fluoro-nucleotides. In some embodiments, the nucleotides at positions 1 and 2 from the 5' end of the sense and antisense strands are linked to adjacent nucleotides by phosphorothioate bonds. In some embodiments, the nucleotides at positions 1 to 6 from the 3' end or positions 1 to 7 from the 3' end are linked to adjacent nucleotides by phosphorothioate bonds.

[0013] In some embodiments, the branched oligonucleotide further comprises a hydrophobic moiety. In certain embodiments, the hydrophobic moiety is attached to one or more terminal 5' positions of the branched oligonucleotide compound. The hydrophobic moiety can be contained in one or more 5' phosphate moieties. In certain embodiments, the hydrophobic moiety comprises an alkyl or alkenyl moiety (e.g., an alkyl or alkenyl chain or a saturated or unsaturated fatty acid residue), a vitamin or cholesterol derivative, an aromatic moiety (e.g., phenyl or naphthyl), a lipophilic amino acid, or a combination thereof. Specific embodiments of hydrophobic moieties and a synthetic strategy for hydrophobically modified branched oligonucleotide compounds are shown in Figure 44.

[0014] In branched oligonucleotide embodiments, each linker is independently selected from an ethylene glycol chain, an alkyl chain, a peptide, RNA, DNA, a phosphate, a phosphonate, a phosphoramidate, an ester, an amide, a triazole, and combinations thereof; any carbon or oxygen atom of the linker may optionally be replaced with a nitrogen atom, have a hydroxyl substituent, or have an oxo substituent.

[0015] In a second embodiment, a compound of formula (I): [ka] [During the ceremony, L is selected from an ethylene glycol chain, an alkyl chain, a peptide, RNA, DNA, a phosphate, a phosphonate, a phosphoramidate, an ester, an amide, a triazole, and combinations thereof; Formula (I) may optionally further comprise one or more branch points B and one or more spacers S; Each B is independently a polyvalent organic species or derivative thereof; each S is independently selected from an ethylene glycol chain, an alkyl chain, a peptide, RNA, DNA, a phosphate, a phosphonate, a phosphoramidate, an ester, an amide, a triazole, and combinations thereof; N is an RNA duplex comprising a sense strand and an antisense strand, each of which independently comprises one or more chemical modifications; n is 2, 3, 4, 5, 6, 7, or 8. Provided herein are compounds of the formula:

[0016] In certain embodiments, the compound of Formula (I) has a structure selected from Formulas (I-1) to (I-9) in Table 1. [Table 1]

[0017] In certain embodiments, each antisense strand independently comprises a 5'-terminal group R selected from the group in Table 2. [Table 2]

[0018] In some embodiments, the compound of Formula (I) has the formula (II): [ka] [During the ceremony, each X is independently selected from adenosine, guanosine, uridine, cytidine and chemically modified derivatives thereof; each Y is independently selected from adenosine, guanosine, uridine, cytidine and chemically modified derivatives thereof; - represents a phosphodiester internucleoside linkage; = represents a phosphorothioate internucleoside linkage; and --- each represents a base pair interaction or mismatch, respectively. It has the following structure.

[0019] In certain embodiments, the compound of Formula (I) has Formula (III): [ka] [During the ceremony, X are each independently a nucleotide that includes a 2'-deoxy-2'-fluoro modification; each X is independently a nucleotide containing a 2'-O-methyl modification; Y are each, independently, a nucleotide that includes a 2'-deoxy-2'-fluoro modification; and and each Y is independently a nucleotide containing a 2'-O-methyl modification. It has the following structure.

[0020] In some embodiments, the compound of Formula (I) has the formula (IV): [ka] [During the ceremony, A is an adenosine containing a 2'-deoxy-2'-fluoro modification; A is an adenosine containing a 2'-O-methyl modification; G is a guanosine containing a 2'-deoxy-2'-fluoro modification; G is a guanosine containing a 2'-O-methyl modification; U is a uridine containing a 2'-deoxy-2'-fluoro modification; U is uridine containing a 2'-O-methyl modification; C is a cytidine containing a 2'-deoxy-2'-fluoro modification; and C is a cytidine containing a 2'-O-methyl modification. It has the following structure.

[0021] In some embodiments, the compound of Formula (I) has the formula (V): [ka] [During the ceremony, each X is independently selected from adenosine, guanosine, uridine, cytidine and chemically modified derivatives thereof; each Y is independently selected from adenosine, guanosine, uridine, cytidine and chemically modified derivatives thereof; - represents a phosphodiester internucleoside linkage; = represents a phosphorothioate internucleoside linkage; --- each represents a base pair interaction or mismatch, respectively. It has the following structure.

[0022] In some embodiments, the compound of Formula (I) has the formula (VI): [ka] [During the ceremony, X are each independently a nucleotide that includes a 2'-deoxy-2'-fluoro modification; each X is independently a nucleotide containing a 2'-O-methyl modification; Y are each independently a nucleotide that includes a 2'-deoxy-2'-fluoro modification; and and each Y is independently a nucleotide containing a 2'-O-methyl modification. It has the following structure.

[0023] In some embodiments, the compound of Formula (I) has the formula (VII): [ka] [During the ceremony, A is an adenosine containing a 2'-deoxy-2'-fluoro modification; A is an adenosine containing a 2'-O-methyl modification; G is a guanosine containing a 2'-deoxy-2'-fluoro modification; G is a guanosine containing a 2'-O-methyl modification; U is a uridine containing a 2'-deoxy-2'-fluoro modification; U is uridine containing a 2'-O-methyl modification; C is a cytidine containing a 2'-deoxy-2'-fluoro modification; C is a cytidine containing a 2'-O-methyl modification; Y are each, independently, a nucleotide that includes a 2'-deoxy-2'-fluoro modification; and each Y is independently a nucleotide containing a 2'-O-methyl modification. It has the following structure.

[0024] In an embodiment of the compound of Formula (I), L is L1: [ka] In an embodiment of L1, R has the structure 3 and n is 2.

[0025] In an embodiment of the compound of Formula (I), L is L2: [ka] In an embodiment of L2, R has the structure 3 and n is 2.

[0026] In a third embodiment, the compound of formula (VIII): [ka] [During the ceremony, L is selected from an ethylene glycol chain, an alkyl chain, a peptide, RNA, DNA, a phosphate, a phosphonate, a phosphoramidate, an ester, an amide, a triazole, and combinations thereof; wherein formula (VIII) optionally further comprises one or more branching points B and one or more spacers S; Each B is independently a polyvalent organic species or derivative thereof; each S is independently selected from an ethylene glycol chain, an alkyl chain, a peptide, RNA, DNA, a phosphate, a phosphonate, a phosphoramidate, an ester, an amide, a triazole, and combinations thereof; each cNA is independently a carrier nucleic acid that includes one or more chemical modifications; and n is 2, 3, 4, 5, 6, 7, or 8. Provided herein is a delivery system for therapeutic nucleic acids having the structure:

[0027] In some embodiments, the compound has a structure selected from formulas (VIII-1) to (VIII-9) in Table 3. [Table 3]

[0028] In some embodiments, in the compound of Formula (VIII) (e.g., including Formulas (VIII-1) to (VIII-9)), each cNA independently comprises at least 15 consecutive nucleotides. In some embodiments, each cNA independently comprises chemically modified nucleotides.

[0029] In some embodiments, the delivery system further comprises n therapeutic nucleic acids (NAs), each NA hybridizing to at least one cNA.

[0030] In some embodiments, each NA independently comprises at least 16 consecutive nucleotides. In some embodiments, each NA independently comprises at least 16-20 consecutive nucleotides. In some embodiments, each NA comprises an unpaired overhang of at least two nucleotides. In some embodiments, the overhanging nucleotides are linked by phosphorothioate bonds.

[0031] In some embodiments, each NA is independently selected from the group consisting of DNA, siRNA, antagomir, miRNA, gapmer, mixmer or guide RNA.In some embodiments, each NA is identical.In some embodiments, each NA is not identical.

[0032] In some embodiments, the delivery system further comprising n therapeutic nucleic acids (NAs) has a structure selected from Formula (I), (II), (III), (IV), (V), (VI), (VII) and embodiments thereof described herein.

[0033] In an embodiment of the delivery system, the target of delivery is selected from the group consisting of brain, liver, skin, kidney, spleen, pancreas, colon, fat, lung, muscle, and thymus. [Brief explanation of the drawings]

[0034] [Figure 1] The structure of di-hsiRNA is shown. Black - 2'-O-methyl, gray - 2'-fluoro, red line - phosphorothioate linkage, linker - tetraethylene glycol. Di-hsiRNA is two asymmetric siRNAs joined by a linker at the 3' end of the sense strand. Hybridization with the longer antisense strand creates a protruding, fully phosphorothioated single-stranded region essential for tissue distribution, cellular uptake, and efficacy. The structure shown utilizes a four-monomer teg linker. The chemical properties of the linker can be varied without affecting efficacy. It can be adjusted by length, chemical composition (all carbon), saturation, or the addition of chemical targeting ligands. [Figure 2]Chemical synthesis, purification and QC of di-branched siRNAs. [Figure 3] Figure 2 shows the HPLC and QC analysis of compounds prepared by the method described above. Three major products were identified by mass spectrometry as the sense strand with a TEG (tetraethylene glycol) linker, the di-branched oligo, and the Vit-D (calciferol) conjugate. All products were independently purified by HPLC and tested in vivo. The di-branched oligo was uniquely characterized by unprecedented tissue distribution and efficacy, indicating that the branched structure is essential for tissue retention and distribution. [Figure 4] Figure 1 shows a mass spectrum confirming the mass of the di-branched oligonucleotide. The observed mass of 11683 corresponds to the two sense strands joined by a TEG linker at their 3' ends. [Figure 5] 1 shows the synthesis of branched oligonucleotides using alternative chemical routes. [Figure 6] Examples of amidites, spacers and branching moieties are shown below. [Figure 7] The motif of branched oligonucleotides is shown. The double helix represents the oligonucleotide. Combinations of different linkers, spacers, and branch points result in a variety of branched hsiRNA structures. [Figure 8] Structurally diverse branched oligonucleotides are shown. [Figure 9] 1 shows an asymmetric compound of the invention having four single-stranded phosphorothioate regions.

[0035] [Figure 10]In vitro efficacy data are shown. (A) HeLa cells were transfected (using RNAiMax) with di-branched oligos at the indicated concentrations for 72 hours. (B) Primary cortical mouse neurons were treated with di-branched oligos at the indicated concentrations for 1 week. mRNA was measured using Affymetrix Quantigene 2.0. Data were normalized to a housekeeping gene (PPIB) and graphed as % of untreated cells. (C) HeLa cells were passively transfected (without formulation) with di-siRNA oligos at the indicated concentrations for 1 week. [Figure 11] Brain distribution of di-siRNA or TEG 48 hours after intrastriatal injection. Intrastriatal injection of 2 nmols of (A) di-branched oligo (4 nmol of the corresponding antisense strand) or (B) TEG-oligo alone. N=2 mice per conjugate. Brains were harvested 48 hours later and stained with Dapi (nuclei, blue). Red-oligo. In (A), the left side of the brain appears bright red, while in (B), the left side of the brain appears only faintly red. [Figure 12] Figure 1 shows that a single injection of di-siRNA is detected both ipsilateral and contralateral to the injection site. [Figure 13] Widespread distribution and efficacy of di-hsiRNA in the mouse brain. (A) Htt mRNA silencing in healthy subjects in both the cerebral cortex and striatum 7 days after a single IS injection (25 μg), QuantiGene®. (B) hsiRNA accumulation in tissues 7 days after injection (PNA assay). [Figure 14]This figure shows the widespread distribution and efficacy of di-hsiRNA in the spinal cord after a bolus intrathecal injection. Intrathecal injection of 3 nmol of di-branched oligo (6 nmol of the corresponding antisense HTT strand) in the lumbar region. (A) Htt mRNA silencing in a healthy individual throughout the entire spinal cord, 7 days, n=6. Animals were sacrificed 7 days after injection. Tissue punches were taken from the cervical, thoracic, and lumbar regions of the spinal cord. mRNA was quantified using Affymetrix Quantigene 2.0 according to Coles et al. (2015). Data were normalized to housekeeping genes, HPRT, and grafts as a percentage of the aCSF control. aCSF = artificial CSF. (B) Animals were injected intrathecally with 75 μg of Cy3-Chol-hsiRNA and Cy-di-hsiRNA via the lumbar vein. Chol-hsiRNAs exhibit a steep gradient of diffusion from the lateral to medial sides of the spinal cord. Di-hsiRNA demonstrates widespread distribution throughout the entire spinal cord (all regions). Leica 10x (20 mbar). Image of di-branched oligo in the cervical region of the spinal cord 48 hours after intrathecal injection. Red = oligo, blue = Dapi. (C) Image of di-branched oligo in the liver 48 hours after intrathecal injection. Red = oligo, blue = Dapi.

[0036] [Figure 15] Illustrated are branched oligonucleotides of the present invention. (A) Branched oligonucleotides formed by annealing three oligonucleotides. The longer linked oligonucleotides can contain cleavable regions in the form of unmodified RNA, DNA, or UNA; (B) Asymmetric branched oligonucleotides with 3' and 5' linkages to the linker or spacer described above. This can be used at the 3' and 5' ends of the sense strand, antisense strand, or a combination thereof; (C) Branched oligonucleotides made with three separate strands. The long double-sense strand is synthesized using 3' and 5' phosphoramidites, allowing for 3'-3' adjacent or 5'-5' adjacent ends. [Figure 16] 1 shows a branched oligonucleotide of the present invention having a conjugated biologically active moiety. [Figure 17] The relationship between phosphorothioate content and stereoselectivity is shown. [Figure 18] Examples of hydrophobic moieties are shown below. [Figure 19] Examples of internucleotide linkages are shown below. [Figure 20] Examples of internucleotide backbone linkages are shown. [Figure 21] Examples of glycosylation products are shown below. [Figure 22] Di-FM-hsiRNA is shown. (A) Crude reverse-phase HPLC analysis of the chemical composition of four byproducts generated from VitD-FM-hsiRNA synthesis and the original chemical synthesis. (B) Efficacy of the byproducts in HeLa cells after lipid-mediated hsiRNA delivery. Cells were treated for 72 hours. mRNA was measured using the QuantiGene 2.0 kit (Affimetrix). Data were normalized to the housekeeping gene HPRT and expressed as a percentage of the untreated control. (C) A single, unilateral intrastriatal injection (25 μg) of each hsiRNA byproduct. Images were taken 48 hours after injection. [Figure 23A-B] This shows that di-HTT-Cy3 does not efficiently induce silencing in the liver or kidney after intrastriatal injection. Figure 23A shows a scatter plot showing Htt mRNA expression in the liver one week after intrastriatal injection of di-HTT-Cy3 compared to a negative control (aCSF). Figure 23B shows a scatter plot showing Htt mRNA expression in the kidney one week after intrastriatal injection of di-HTT-Cy3 compared to a negative control (aCSF). [Figure 24A-B] These results demonstrate that di-HTT effectively silences HTT gene expression in both the striatum and cerebral cortex after intrastriatal injection, and that di-HTT-Cy3 is slightly more effective than di-HTT (unlabeled). Figure 24A shows scatter plots of Htt mRNA expression in the striatum one week after intrastriatal injection of di-HTT, di-HTT-Cy3, or two negative controls (aCSF or di-NTC). Figure 24B shows scatter plots of Htt mRNA expression in the cerebral cortex one week after intrastriatal injection of di-HTT, di-HTT-Cy3, or two negative controls (aCSF or di-NTC). [Figure 25]Scatter dot plots measuring the amount of di-HTT-Cy3 in the striatum and cerebral cortex are shown, demonstrating that significant amounts of di-HTT-Cy3 are still detectable 2 weeks after intrastriatal injection.

[0037] [Figure 26A-B] Figure 26 shows that di-HTT-Cy3 effectively silences HTT mRNA and protein expression in both the striatum and cerebral cortex two weeks after intrastriatal injection. Figure 26A shows a scatter plot measuring Htt mRNA levels in the striatum and cerebral cortex two weeks after intrastriatal injection. Figure 26B shows a scatter plot measuring Htt protein levels in the striatum and cerebral cortex two weeks after intrastriatal injection. [Figure 27A-B] Two weeks after intrastriatal injection, high-dose di-HTT-Cy3 treatment does not result in significant toxicity in vivo, but does result in significant gliosis in vivo. Figure 27A shows scatter plots measuring DARPP32 signals in the striatum and cerebral cortex two weeks after injection of di-HTT-Cy3 or aCSF. Figure 27B shows scatter plots measuring GFAP protein levels in the striatum and cerebral cortex two weeks after injection of di-HTT-Cy3 or aCSF. [Figure 28] Fluorescence imaging showing that intrathecal injection of di-HTT-Cy3 results in a strong and uniform distribution throughout the spinal cord. [Figure 29] A composite fluorescent image of Figure 28B is shown (magnification of the spinal cord). Blue - nuclei, red - di-HTT-Cy3. [Figure 30A-C] Figure 30A shows the widespread distribution of di-HTT-Cy3 48 hours after intraventricular injection. Figure 30A shows fluorescent images of sections of the striatum, cerebral cortex, and cerebellum. Figure 30B shows bright-field images of whole brains injected with control (aCSF) or di-HTT-Cy3. Figure 30C shows fluorescent images of whole brains 48 hours after di-HTT-Cy3 injection. [Figure 31] Two weeks after intracerebroventricular injection, di-HTT-Cy3 accumulates in multiple brain regions. Scatter dot plots show the amount of di-HTT-Cy3 in multiple brain regions. [Figure 32] Figure 32A shows that di-HTT-Cy3 induces Htt gene silencing in multiple brain regions two weeks after intracerebroventricular injection, compared to a negative control injection (aCSF). Scatter dot plots measure Htt mRNA levels in multiple brain regions. Figure 32B shows that di-HTT-Cy3 induces Htt silencing in multiple brain regions two weeks after intracerebroventricular injection, compared to a negative control injection (aCSF). Scatter dot plots measure Htt protein levels in multiple brain regions. [Figure 33] Figure 1 shows that high-dose intracerebroventricular injection causes mild toxicity in vivo. Scatter dot plots measure DARPP32 signal in multiple regions of the brain after di-HTT-Cy3 injection of aCSF. [Figure 34] Scatter dot plots measure DARPP32 signal in multiple regions of the brain after di-HTT-Cy3 injection of aCSF, showing that high-dose intracerebroventricular injection results in significant gliosis in vivo.

[0038] [Figure 35] After intravenous injection, di-HTT-Cy3 is distributed to multiple organs. Fluorescent images show the amount of di-HTT-Cy3 in the heart, kidney, adrenal gland, and spleen after injection of di-HTT-Cy3 or a negative control (PBS). [Figure 36] Figure 1 shows the accumulation of di-HTT-Cy3 in multiple organs after intravenous injection. Scatter dot plots show the amount of di-HTT-Cy3 in multiple tissues. [Figure 37] The structures of hsiRNA and fully metabolized (FM) hsiRNA are shown. [Figure 38A-B] Figure 38A shows a scatter plot measuring HTT mRNA levels up to 12 days after intrastriatal injection. Figure 38B shows a scatter plot measuring HTT mRNA levels up to 28 days after intrastriatal injection. [Figure 39] Figure 1 shows the chemical diversity of single-stranded fully modified oligonucleotides. Single-stranded oligonucleotides can consist of gapmers, mixmers, miRNA inhibitors, SSOs, PMOs, or PNAs. [Figure 40] Di-HTT with a TEG phosphoramidate linker is shown. [Figure 41] Di-HTT with a TEG diphosphate linker is shown. [Figure 42] Variants of di-HTT with either two or four oligonucleotide branches are shown. [Figure 43] 1 shows another variation of di-HTT in structure with two oligonucleotide branches and R2 attached to a linker. [Figure 44] A first strategy for the introduction of hydrophobic moieties into branched oligonucleotide structures is presented. [Figure 45] A second strategy for the introduction of hydrophobic moieties into branched oligonucleotide structures is presented. [Figure 46] A third strategy for the introduction of hydrophobic moieties into branched oligonucleotide structures is presented. DETAILED DESCRIPTION OF THE INVENTION

[0039] Detailed Description The present invention provides branched oligonucleotides ("compounds of the invention") that exhibit unexpected improvements in distribution, in vivo efficacy, and safety. The branched oligonucleotides described herein efficiently and stably deliver small RNAs to multiple regions of the brain and multiple other organs, demonstrating unprecedented delivery efficacy not previously demonstrated with unconjugated small RNAs.

[0040] The compositions described herein enable efficient, stable delivery of siRNA to promote potent silencing of therapeutic target genes, demonstrating therapeutic potential for a number of difficult-to-treat diseases and overcoming existing challenges in the use of RNA therapeutics.

[0041] In a first embodiment, a branched oligonucleotide compound is provided that comprises two or more nucleic acids linked together by one or more moieties selected from a linker, a spacer, and a branch point.

[0042] In various aspects and embodiments herein, branched oligonucleotides are provided as compounds of the present invention. In some embodiments, compounds of the present invention have 2 to 8 oligonucleotides linked by a linker. The linker may be hydrophobic. In certain embodiments, compounds of the present invention have 2 to 3 oligonucleotides. In certain embodiments, these oligonucleotides independently have sufficient chemical stability (e.g., at least 40% of the constituent bases are chemically modified). In certain embodiments, the oligonucleotides have complete chemical stability (i.e., all the constituent bases are chemically modified). In some embodiments, compounds of the present invention include one or more single-stranded phosphorothioated tails, each independently having 2 to 20 nucleotides. In certain embodiments, each single-stranded tail has 8 to 10 nucleotides.

[0043] In certain embodiments, the compounds of the present invention are characterized by three properties: (1) branched structure, (2) metabolic stabilization, and (3) the presence of a single-stranded tail containing a phosphorothioate linker. In certain embodiments, the compounds of the present invention have two or three branches. The increased overall size of the branched structure promotes increased uptake. Also, without being bound by a particular theory of activity, the multiple (e.g., two or three) adjacent branches allow each branch to move in a coordinated manner, thus dramatically increasing the rate of internalization, trafficking, and release.

[0044] The complete metabolic stabilization of the branched oligonucleotides of the present invention results in unexpectedly high in vivo efficacy. Destabilized branched siRNAs lack in vivo efficacy. The presence of a single-stranded tail is necessary for the activity of branched oligonucleotides. The phosphoramidate functional group is important for the function of di-branched oligos.

[0045] In certain embodiments, the compounds of the present invention are characterized by the following properties: (1) two or more branched oligonucleotides, e.g., having an unequal number of 3' and 5' ends; (2) substantially chemically stabilized oligonucleotides (e.g., free of RNA and optionally DNA), e.g., being 40% or more, optimally 100%, chemically modified; and (3) phosphorothioated single-stranded oligonucleotides containing at least three, optimally 5-20, phosphorothioate linkages.

[0046] The compounds of the present invention are provided in a variety of structurally diverse embodiments. For example, as shown in Figure 7, in some embodiments, the oligonucleotides attached at the branch points are single-stranded and consist of miRNA inhibitors, gapmers, mixmers, SSOs, PMOs, or PNAs. These single strands can be attached at their 3' or 5' ends. Combinations of siRNA and single-stranded oligonucleotides can also be used for dual functions. In another embodiment, short oligonucleotides complementary to gapmers, mixmers, miRNA inhibitors, SSOs, PMOs, and PNAs are used to deliver these active single-stranded oligonucleotides and promote their distribution and cellular internalization. The short double-stranded region has a low melting temperature (T) for rapid dissociation of the branched structure after internalization into cells. m approximately 37°C).

[0047] As shown in Figure 16, the di-siRNA compounds of the present invention can contain chemically diverse conjugates. Conjugated bioactive ligands can be used to improve cell specificity and promote membrane binding, internalization, and serum protein binding. Examples of bioactive moieties used in conjugates include DHAg2, DHA, GalNAc, and cholesterol. These moieties can be linked to the di-siRNA via a linker or spacer, or can be added via additional linkers or spacers attached to the other free siRNA termini.

[0048] The presence of branched structures improves the level of tissue retention in the brain by 100-fold compared to unbranched compounds of individual chemical compositions, suggesting a new mechanism of cellular retention and distribution.The compounds of the present invention unexpectedly achieve uniform distribution in the spinal cord and brain.Furthermore, the compounds of the present invention unexpectedly demonstrate efficient systemic delivery to various tissues and extremely high levels of tissue accumulation.

[0049] The compounds of the present invention include various therapeutic oligonucleotides, including ASO, miRNA, miRNA inhibitor, splice switching, PMO, PNA. In some embodiments, the compounds of the present invention further comprise a conjugated hydrophobic moiety and show unexpected silencing and efficacy in vivo and in vitro.

[0050] Non-limiting embodiments of branched oligonucleotide structures are shown in Figures 1, 7-9, 15-17, and 40-45. Non-limiting examples of linkers, spacers, and branch points are shown in Figure 6.

[0051] Variable Nucleic Acids In some embodiments, the branched oligonucleotides comprise 2, 3, 4, 6, or 8 nucleic acids. In some embodiments, the branched oligonucleotides comprise 2 nucleic acids. In another embodiment, the branched oligonucleotides comprise 3 nucleic acids. In another embodiment, the branched oligonucleotides comprise 4 nucleic acids. In another embodiment, the branched oligonucleotides comprise 6 nucleic acids. In another embodiment, the branched oligonucleotides comprise 8 nucleic acids. In another embodiment, the branched oligonucleotides comprise 5 nucleic acids. In another embodiment, the branched oligonucleotides comprise 7 nucleic acids.

[0052] In an embodiment of the branched oligonucleotide, each nucleic acid is single-stranded and has a 5'-end and a 3'-end, and each nucleic acid is independently linked to a linker, spacer, or branch point at the 5'-end or the 3'-end. In one embodiment, each nucleic acid is linked to a linker, spacer, or branch point at the 3'-end. In another embodiment, each nucleic acid is linked to a linker, spacer, or branch point at the 5'-end. In one embodiment, each nucleic acid is linked to each nucleic acid linker. In another embodiment, each nucleic acid is linked to each nucleic acid spacer. In another embodiment, each nucleic acid is linked to each nucleic acid branch point.

[0053] In some embodiments, each single-stranded nucleic acid independently comprises at least 15 consecutive nucleotides.In some embodiments, nucleic acid comprises at least 15, at least 16, at least 17, at least 18, at least 19 or at least 20 consecutive nucleotides, and has complementarity to target.In certain embodiments, complementarity is >95%, >90%, >85%, >80%, >75%, >70%, >65%, >60%, >55% or >50%.In some embodiments, nucleic acid has complete complementarity to target.

[0054] In an embodiment of the branched oligonucleotide, each nucleic acid is double-stranded and comprises a sense strand and an antisense strand, wherein the sense strand and the antisense strand have a 5'-end and a 3'-end, respectively. In some embodiments, each double-stranded nucleic acid is independently linked to a linker, spacer, or branch point at the 3'-end or 5'-end of the sense strand or the antisense strand. In some embodiments, each nucleic acid is linked to a linker, spacer, or branch point at the 3'-end of the sense strand. In some embodiments, each nucleic acid is linked to a linker, spacer, or branch point at the 3'-end of the antisense strand. In other embodiments, each nucleic acid is linked to a linker, spacer, or branch point at the 5'-end of the sense strand. In other embodiments, each nucleic acid is linked to a linker, spacer, or branch point at the 5'-end of the antisense strand. In some embodiments, each nucleic acid is linked to a linker. In other embodiments, each nucleic acid is linked to a spacer. In other embodiments, each nucleic acid is linked to a branch point.

[0055] In some embodiments, each double-stranded nucleic acid independently comprises at least 15 consecutive nucleotides.In some embodiments, antisense strand comprises at least 15, at least 16, at least 17, at least 18, at least 19 or at least 20 consecutive nucleotides, and has complementarity to target.In certain embodiments, complementarity is >95%, >90%, >85%, >80%, >75%, >70%, >65%, >60%, >55% or >50%.In some embodiments, antisense strand has complete complementarity to target.

[0056] Modified Nucleotides In some embodiments, each nucleic acid comprises one or more chemically modified nucleotides. In some embodiments, each nucleic acid consists of chemically modified nucleotides. In certain embodiments, more than 95%, more than 90%, more than 85%, more than 80%, more than 75%, more than 70%, more than 65%, more than 60%, more than 55%, or more than 50% of each nucleic acid comprises chemically modified nucleotides.

[0057] In some embodiments, the sense strand and the antisense strand each contain one or more chemically modified nucleotides. In some embodiments, the sense strand and the antisense strand each consist of chemically modified nucleotides. In some embodiments, both the sense strand and the antisense strand contain additional 2'-methoxy-nucleotides and 2'-fluoro-nucleotides. In some embodiments, the nucleotides at positions 1 and 2 from the 5' end of the sense strand and the antisense strand are linked to adjacent nucleotides by phosphorothioate bonds. In some embodiments, the nucleotides at positions 1 to 6 from the 3' end or 1 to 7 from the 3' end are linked to adjacent nucleotides by phosphorothioate bonds. In other embodiments, at least five nucleotides are linked to adjacent nucleotides by phosphorothioate bonds.

[0058] In an embodiment of the branched oligonucleotide, each linker is independently selected from an ethylene glycol chain, an alkyl chain, a peptide, RNA, DNA, a phosphate, a phosphonate, a phosphoramidate, an ester, an amide, a triazole, and a combination thereof, wherein any carbon or oxygen of the linker may be optionally replaced with a nitrogen atom, have a hydroxyl functional group, or have an oxo substituent. In one embodiment, each linker is an ethylene glycol chain. In another embodiment, each linker is an alkyl chain. In another embodiment, each linker is a peptide. In another embodiment, each linker is RNA. In another embodiment, each linker is DNA. In another embodiment, each linker is a phosphate. In another embodiment, each linker is a phosphonate. In another embodiment, each linker is a phosphoramidate. In another embodiment, each linker is an ester. In another embodiment, each linker is an amide. In another embodiment, each linker is a triazole. In another embodiment, each linker is a structure selected from the formula in Figure 7.

[0059] Compounds of formula (I) In a second embodiment, a compound of formula (I): [ka] [During the ceremony, L is selected from an ethylene glycol chain, an alkyl chain, a peptide, RNA, DNA, a phosphate, a phosphonate, a phosphoramidate, an ester, an amide, a triazole, and combinations thereof; Formula (I) optionally further comprises one or more branch points B and one or more spacers S; Each B is independently a polyvalent organic species or derivative thereof; each S is independently selected from an ethylene glycol chain, an alkyl chain, a peptide, RNA, DNA, a phosphate, a phosphonate, a phosphoramidate, an ester, an amide, a triazole, and combinations thereof; N is a double-stranded RNA comprising a sense strand and an antisense strand, each of which contains one or more chemical modifications; and n is 2, 3, 4, 5, 6, 7, or 8. The compound of formula (I) is provided.

[0060] In certain embodiments, the compound of Formula (I) has a structure selected from Formulas (I-1) to (I-9) in Table 1. [Table 4]

[0061] In one embodiment, the compound of Formula (I) is Formula (I-1). In another embodiment, the compound of Formula (I) is Formula (I-2). In another embodiment, the compound of Formula (I) is Formula (I-3). In another embodiment, the compound of Formula (I) is Formula (I-4). In another embodiment, the compound of Formula (I) is Formula (I-5). In another embodiment, the compound of Formula (I) is Formula (I-6). In another embodiment, the compound of Formula (I) is Formula (I-7). In another embodiment, the compound of Formula (I) is Formula (I-8). In another embodiment, the compound of Formula (I) is Formula (I-9).

[0062] In an embodiment of the compound of Formula (I), each linker is independently selected from an ethylene glycol chain, an alkyl chain, a peptide, RNA, DNA, a phosphate, a phosphonate, a phosphoramidate, an ester, an amide, a triazole, and combinations thereof, wherein any carbon or oxygen of the linker may be optionally replaced with a nitrogen atom, have a hydroxyl substituent, or have an oxo substituent. In some embodiments of the compound of Formula (I), each linker is an ethylene glycol chain. In other embodiments, each linker is an alkyl chain. In other embodiments of the compound of Formula (I), each linker is a peptide. In other embodiments of the compound of Formula (I), each linker is RNA. In other embodiments of the compound of Formula (I), each linker is DNA. In other embodiments of the compound of Formula (I), each linker is a phosphate. In other embodiments, each linker is a phosphonate. In other embodiments of the compound of Formula (I), each linker is a phosphoramidate. In other embodiments of the compound of Formula (I), each linker is an ester. In another embodiment of the compound of Formula (I), each linker is an amide. In another embodiment of the compound of Formula (I), each linker is a triazole. In another embodiment of the compound of Formula (I), each linker is a structure selected from the formulas of Figure 7.

[0063] In certain embodiments of the compound of Formula (I), B is a polyvalent organic species. In another embodiment of the compound of Formula (I), B is a derivative of a polyvalent organic species. In certain embodiments of the compound of Formula (I), B is a triol or tetraol derivative. In another embodiment, B is a tri- or tetracarboxylic acid derivative. In another embodiment, B is an amine derivative. In another embodiment, B is a tri- or tetra-amine derivative. In another embodiment, B is an amino acid derivative. In another embodiment of the compound of Formula (I), B is selected from the formulae of Figure 6.

[0064] Polyvalent organic species are moieties that contain carbon and a valency of three or more (i.e., points of attachment with moieties such as S, L, or N, as defined above). Non-limiting examples of polyvalent organic species include triols (e.g., glycerol, phloroglucinol, etc.), tetraols (e.g., ribose, pentaerythritol, 1,2,3,5-tetrahydroxybenzene, etc.), tricarboxylic acids (e.g., succinic acid, 1,3,5-cyclohexanetricarboxylic acid, trimesic acid, etc.), tetracarboxylic acids (e.g., ethylenediaminetetraacetic acid, pyromellitic acid, etc.), tertiary amines (e.g., tripropargylamine, triethanolamine, etc.), triamines (e.g., diethylenetriamine, etc.), tetramines, and species that contain combinations of hydroxyl, thiol, amino, and / or carboxyl moieties (e.g., amino acids such as lysine, serine, cysteine, etc.).

[0065] In an embodiment of the compound of Formula (I), each nucleic acid comprises one or more chemically modified nucleotides. In an embodiment of the compound of Formula (I), each nucleic acid consists of chemically modified nucleotides. In certain embodiments of the compound of Formula (I), >95%, >90%, >85%, >80%, >75%, >70%, >65%, >60%, >55%, or >50% of each nucleic acid comprises chemically modified nucleotides.

[0066] In certain embodiments, each antisense strand independently comprises a 5'-terminal group R selected from the group in Table 2. [Table 5]

[0067] In some embodiments, R is R1. In other embodiments, R is R2. In other embodiments, R is R3. In other embodiments, R is R4. In other embodiments, R is R5. In other embodiments, R is R6. In other embodiments, R is R7. In other embodiments, R is R8.

[0068] Structure of Formula (II) In some embodiments, the compound of Formula (I) has the formula (II): [ka] [During the ceremony, Each X is independently selected from adenosine, guanosine, uridine, cytidine, and chemically modified derivatives thereof; each Y is independently selected from adenosine, guanosine, uridine, cytidine, and chemically modified derivatives thereof; - represents a phosphodiester internucleoside linkage; = represents a phosphorothioate internucleoside linkage; and --- each represents a base pair interaction or mismatch, respectively. It has the following structure.

[0069] In certain embodiments, the structure of Formula (II) does not contain any mismatch. In some embodiments, the structure of Formula (II) contains one mismatch. In another embodiment, the compound of Formula (II) contains two mismatches. In another embodiment, the compound of Formula (II) contains three mismatches. In another embodiment, the compound of Formula (II) contains four mismatches. In some embodiments, each nucleic acid is composed of chemically modified nucleotides.

[0070] In certain embodiments, >95%, >90%, >85%, >80%, >75%, >70%, >65%, >60%, >55%, or >50% of the X in the structure of Formula (II) are chemically modified nucleotides. In other embodiments, >95%, >90%, >85%, >80%, >75%, >70%, >65%, >60%, >55%, or >50% of the X in the structure of Formula (II) are chemically modified nucleotides.

[0071] Structure of formula (III) In certain embodiments, the compound of Formula (I) has Formula (III): [ka] [During the ceremony, X are each independently a nucleotide that includes a 2'-deoxy-2'-fluoro modification; each X is independently a nucleotide containing a 2'-O-methyl modification; Y are each independently a nucleotide that includes a 2'-deoxy-2'-fluoro modification; and and each Y is independently a nucleotide containing a 2'-O-methyl modification. It has the following structure.

[0072] In some embodiments, X is selected from the group consisting of 2'-deoxy-2'-fluoro modified adenosine, guanosine, uridine, or cytidine. In some embodiments, X is selected from the group consisting of 2'-O-methyl modified adenosine, guanosine, uridine, or cytidine. In some embodiments, Y is selected from the group consisting of 2'-deoxy-2'-fluoro modified adenosine, guanosine, uridine, or cytidine. In some embodiments, Y is selected from the group consisting of 2'-O-methyl modified adenosine, guanosine, uridine, or cytidine.

[0073] In certain embodiments, the structure of Formula (III) does not contain a mismatch. In certain embodiments, the structure of Formula (III) contains one mismatch. In other embodiments, the structure of Formula (III) contains two mismatches. In other embodiments, the compound of Formula (III) contains three mismatches. In other embodiments, the compound of Formula (III) contains four mismatches.

[0074] Structure of formula (IV) In some embodiments, the compound of Formula (I) has the formula (IV): [ka] [During the ceremony, A is an adenosine containing a 2'-deoxy-2'-fluoro modification; A is an adenosine containing a 2'-O-methyl modification; G is a guanosine containing a 2'-deoxy-2'-fluoro modification; G is a guanosine containing a 2'-O-methyl modification; U is a uridine containing a 2'-deoxy-2'-fluoro modification; U is uridine containing a 2'-O-methyl modification; C is a cytidine containing a 2'-deoxy-2'-fluoro modification; and C is a cytidine containing a 2'-O-methyl modification. It has the following structure.

[0075] Structure of formula (V) In some embodiments, the compound of Formula (I) has the formula (V): [ka] [During the ceremony, Each X is independently selected from adenosine, guanosine, uridine, cytidine, and chemically modified derivatives thereof; each Y is independently selected from adenosine, guanosine, uridine, cytidine, and chemically modified derivatives thereof; - represents a phosphodiester internucleoside linkage; = represents a phosphorothioate internucleoside linkage; and --- each represents a base pair interaction or mismatch, respectively. It has the following structure.

[0076] In certain embodiments, the structure of Formula (V) does not contain any mismatches. In some embodiments, the structure of Formula (V) contains one mismatch. In another embodiment, the compound of Formula (V) contains two mismatches. In another embodiment, the compound of Formula (V) contains three mismatches. In another embodiment, the compound of Formula (V) contains four mismatches. In some embodiments, each nucleic acid is composed of chemically modified nucleotides.

[0077] In certain embodiments, >95%, >90%, >85%, >80%, >75%, >70%, >65%, >60%, >55%, or >50% of the X in the structure of Formula (II) are chemically modified nucleotides. In other embodiments, >95%, >90%, >85%, >80%, >75%, >70%, >65%, >60%, >55%, or >50% of the X in the structure of Formula (II) are chemically modified nucleotides.

[0078] Structure of Formula (VI) In some embodiments, the compound of Formula (I) has the formula (VI): [ka] [During the ceremony, X are each independently a nucleotide that includes a 2'-deoxy-2'-fluoro modification; each X is independently a nucleotide containing a 2'-O-methyl modification; Y are each independently a nucleotide that includes a 2'-deoxy-2'-fluoro modification; and and each Y is independently a nucleotide containing a 2'-O-methyl modification. It has the following structure. (VI)

[0079] In certain embodiments, X is selected from the group consisting of 2'-deoxy-2'-fluoro modified adenosine, guanosine, uridine, or cytidine. In some embodiments, X is selected from the group consisting of 2'-O-methyl modified adenosine, guanosine, uridine, or cytidine. In some embodiments, Y is selected from the group consisting of 2'-deoxy-2'-fluoro modified adenosine, guanosine, uridine, or cytidine. In some embodiments, Y is selected from the group consisting of 2'-O-methyl modified adenosine, guanosine, uridine, or cytidine.

[0080] In certain embodiments, the structure of Formula (VI) does not contain any mismatches. In certain embodiments, the structure of Formula (VI) contains one mismatch. In another embodiment, the compound of Formula (VI) contains two mismatches. In another embodiment, the compound of Formula (VI) contains three mismatches. In another embodiment, the compound of Formula (VI) contains four mismatches.

[0081] Structure of formula (VII) In some embodiments, the compound of Formula (I) has the formula (VII): [ka] [During the ceremony, A is an adenosine containing a 2'-deoxy-2'-fluoro modification; A is an adenosine containing a 2'-O-methyl modification; G is a guanosine containing a 2'-deoxy-2'-fluoro modification; G is a guanosine containing a 2'-O-methyl modification; U is a uridine containing a 2'-deoxy-2'-fluoro modification; U is uridine containing a 2'-O-methyl modification; C is a cytidine containing a 2'-deoxy-2'-fluoro modification; and C is a cytidine containing a 2'-O-methyl modification; Y are each, independently, a nucleotide that includes a 2'-deoxy-2'-fluoro modification; and and each Y is independently a nucleotide containing a 2'-O-methyl modification. It has the following structure.

[0082] Flexible Linker In an embodiment of the compound of Formula (I), L is L1: [ka] In an embodiment of L1, R has the structure 3 and n is 2.

[0083] In an embodiment of the structure of formula (II), L has the structure L1. In an embodiment of the structure of formula (III), L has the structure L1. In an embodiment of the structure of formula (IV), L has the structure L1. In an embodiment of the structure of formula (V), L has the structure L1. In an embodiment of the structure of formula (VI), L has the structure L1. In an embodiment of the structure of formula (VII), L has the structure L1.

[0084] In an embodiment of the compound of Formula (I), L is L2: [ka] In an embodiment of L2, R has the structure 3 and n is 2.

[0085] In an embodiment of the structure of formula (II), L has the structure L2. In an embodiment of the structure of formula (III), L has the structure L2. In an embodiment of the structure of formula (IV), L has the structure L2. In an embodiment of the structure of formula (V), L has the structure L2. In an embodiment of the structure of formula (VI), L has the structure L2. In an embodiment of the structure of formula (VII), L has the structure L2.

[0086] delivery system In a third embodiment, the compound of formula (VIII): [ka] [During the ceremony, L is selected from an ethylene glycol chain, an alkyl chain, a peptide, RNA, DNA, a phosphate, a phosphonate, a phosphoramidate, an ester, an amide, a triazole, and combinations thereof; wherein formula (VIII) optionally further comprises one or more branching points B and one or more spacers S; Each B is independently a polyvalent organic species or derivative thereof; each S is independently selected from an ethylene glycol chain, an alkyl chain, a peptide, RNA, DNA, a phosphate, a phosphonate, a phosphoramidate, an ester, an amide, a triazole, and combinations thereof; each cNA is independently a carrier nucleic acid that includes one or more chemical modifications; and n is 2, 3, 4, 5, 6, 7, or 8. Provided herein is a delivery system for therapeutic nucleic acids having the structure:

[0087] In one embodiment of the delivery system, L is an ethylene glycol chain. In another embodiment of the delivery system, L is an alkyl chain. In another embodiment of the delivery system, L is a peptide. In another embodiment of the delivery system, L is RNA. In another embodiment of the delivery system, L is DNA. In another embodiment of the delivery system, L is a phosphate. In another embodiment of the delivery system, L is a phosphonate. In another embodiment of the delivery system, L is a phosphoramidate. In another embodiment of the delivery system, L is an ester. In another embodiment of the delivery system, L is an amide. In another embodiment of the delivery system, L is a triazole.

[0088] In one embodiment of the delivery system, S is an ethylene glycol chain. In another embodiment, S is an alkyl chain. In another embodiment of the delivery system, S is a peptide. In another embodiment, S is RNA. In another embodiment of the delivery system, S is DNA. In another embodiment of the delivery system, S is a phosphate. In another embodiment of the delivery system, S is a phosphonate. In another embodiment of the delivery system, S is a phosphoramidate. In another embodiment of the delivery system, S is an ester. In another embodiment, S is an amide. In another embodiment, S is a triazole.

[0089] In one embodiment of the delivery system, n is 2. In another embodiment of the delivery system, n is 3. In another embodiment of the delivery system, n is 4. In another embodiment of the delivery system, n is 5. In another embodiment of the delivery system, n is 6. In another embodiment of the delivery system, n is 7. In another embodiment of the delivery system, n is 8.

[0090] In certain embodiments, each cNA comprises >95%, >90%, >85%, >80%, >75%, >70%, >65%, >60%, >55% or >50% chemically modified nucleotides.

[0091] In certain embodiments, the compound of formula (VIII) has a structure selected from formulas (VIII-1) to (VIII-9) in Table 3. [Table 6]

[0092] In some embodiments, the compound of Formula (VIII) has the structure of Formula (VIII-1). In some embodiments, the compound of Formula (VIII) has the structure of Formula (VIII-2). In some embodiments, the compound of Formula (VIII) has the structure of Formula (VIII-3). In some embodiments, the compound of Formula (VIII) has the structure of Formula (VIII-4). In some embodiments, the compound of Formula (VIII) has the structure of Formula (VIII-5). In some embodiments, the compound of Formula (VIII) has the structure of Formula (VIII-6). In some embodiments, the compound of Formula (VIII) has the structure of Formula (VIII-7). In some embodiments, the compound of Formula (VIII) has the structure of Formula (VIII-8). In some embodiments, the compound of Formula (VIII) has the structure of Formula (VIII-9).

[0093] In some embodiments, in the compound of Formula (VIII) (e.g., including Formulas (VIII-1) to (VIII-9)), each cNA independently comprises at least 15 consecutive nucleotides. In some embodiments, each cNA independently comprises chemically modified nucleotides.

[0094] In some embodiments, the delivery system further comprises n therapeutic nucleic acids (NAs), wherein each NA is hybridized to at least one cNA. In some embodiments, the delivery system comprises two NAs. In another embodiment, the delivery system comprises three NAs. In another embodiment, the delivery system comprises four NAs. In another embodiment, the delivery system comprises five NAs. In another embodiment, the delivery system comprises six NAs. In another embodiment, the delivery system comprises seven NAs. In another embodiment, the delivery system comprises eight NAs.

[0095] In some embodiments, each NA independently comprises at least 16 consecutive nucleotides. In some embodiments, each NA independently comprises 16 to 20 consecutive nucleotides. In some embodiments, each NA independently comprises 16 consecutive nucleotides. In other embodiments, each NA independently comprises 17 consecutive nucleotides. In other embodiments, each NA independently comprises 18 consecutive nucleotides. In other embodiments, each NA independently comprises 19 consecutive nucleotides. In other embodiments, each NA independently comprises 20 consecutive nucleotides.

[0096] In some embodiments, each NA comprises an unpaired overhang of at least 2 nucleotides. In other embodiments, each NA comprises an unpaired overhang of at least 3 nucleotides. In other embodiments, each NA comprises an unpaired overhang of at least 4 nucleotides. In other embodiments, each NA comprises an unpaired overhang of at least 5 nucleotides. In other embodiments, each NA comprises an unpaired overhang of at least 6 nucleotides. In some embodiments, the nucleotides of the overhang are linked by phosphorothioate bonds.

[0097] In some embodiments, each NA is independently selected from the group consisting of DNA, siRNA, antagomir, miRNAs, gapmers, mixmers, or guide RNA. In some embodiments, each NA is independently DNA. In another embodiment, each NA is independently siRNA. In another embodiment, each NA is independently antagomir. In another embodiment, each NA is independently miRNA. In another embodiment, each NA is independently a gapmer. In another embodiment, each NA is independently a mixmer. In another embodiment, each NA is independently a guide RNA. In some embodiments, each NA is identical. In some embodiments, each NA is not identical.

[0098] In some embodiments, the delivery system further comprising n therapeutic nucleic acids (NAs) has a structure selected from Formula (I), Formula (II), Formula (III), Formula (IV), Formula (V), Formula (VI), Formula (VII), and embodiments thereof described herein. In some embodiments, the delivery system comprises a structure selected from Formula (I), Formula (II), Formula (III), Formula (IV), Formula (V), Formula (VI), Formula (VII), and embodiments thereof described herein, further comprising two therapeutic nucleic acids (NAs). In other embodiments, the delivery system comprises a structure selected from Formula (I), Formula (II), Formula (III), Formula (IV), Formula (V), Formula (VI), Formula (VII), and embodiments thereof described herein, further comprising three therapeutic nucleic acids (NAs). In some embodiments, the delivery system comprises a structure selected from Formula (I), Formula (II), Formula (III), Formula (IV), Formula (V), Formula (VI), Formula (VII), and embodiments thereof described herein, further comprising four therapeutic nucleic acids (NAs). In some embodiments, the delivery system has a structure selected from Formula (I), Formula (II), Formula (III), Formula (IV), Formula (V), Formula (VI), (VII), and embodiments thereof described herein, further comprising five therapeutic nucleic acids (NAs). In some embodiments, the delivery system has a structure selected from Formula (I), Formula (II), Formula (III), Formula (IV), Formula (V), Formula (VI), Formula (VII), and embodiments thereof described herein, further comprising six therapeutic nucleic acids (NAs). In some embodiments, the delivery system has a structure selected from Formula (I), Formula (II), Formula (III), Formula (IV), Formula (V), Formula (VI), Formula (VII), and embodiments thereof described herein, further comprising seven therapeutic nucleic acids (NAs). In some embodiments, the delivery system has a structure selected from Formula (I), Formula (II), Formula (III), Formula (IV), Formula (V), Formula (VI), Formula (VII), and embodiments thereof described herein, further comprising eight therapeutic nucleic acids (NAs).

[0099] In some embodiments, the delivery system comprises a compound wherein R is R 3and n is 2. In another embodiment, the delivery system has a structure selected from Formula (I), Formula (II), Formula (III), Formula (IV), Formula (V), Formula (VI), Formula (VII), further comprising a linker of structure L1 or L2, wherein R is R 3 and n is 2. In another embodiment, the delivery system has a structure selected from Formula (I), Formula (II), Formula (III), Formula (IV), Formula (V), Formula (VI), Formula (VII), further comprising a linker of structure L1, wherein R is R 3 and further comprising a linker of structure L2, wherein n is 2.

[0100] In embodiments of the delivery system, the target of delivery is selected from the group consisting of brain, liver, skin, kidney, spleen, pancreas, colon, fat, lung, muscle, and thymus. In one embodiment, the target of delivery is the brain. In another embodiment, the target of delivery is the striatum of the brain. In another embodiment, the target of delivery is the cerebral cortex of the brain. In another embodiment, the target of delivery is the striatum of the brain. In one embodiment, the target of delivery is the liver. In one embodiment, the target of delivery is the skin. In one embodiment, the target of delivery is the kidney. In one embodiment, the target of delivery is the spleen. In one embodiment, the target of delivery is the pancreas. In one embodiment, the target of delivery is the colon. In one embodiment, the target of delivery is fat. In one embodiment, the target of delivery is the lung. In one embodiment, the target of delivery is muscle. In one embodiment, the target of delivery is the thymus. In one embodiment, the target of delivery is the spinal cord.

[0101] It is to be understood that the methods described herein are not limited to the particular methods and experimental conditions disclosed herein, as such methods and conditions may vary, and the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting.

[0102] Furthermore, unless otherwise indicated, the experiments described herein use conventional molecular and cell biological and immunological techniques that are within the skill of the art.Such techniques are known to those skilled in the art and are described in detail in literature.For example, see Ausubel, et al., ed., Current Protocols in Molecular Biology, John Wiley & Sons, Inc., NY (1987-2008), including all supplements; Molecular Cloning: A Laboratory Manual (Fourth Edition) by MR Green and J. Sambrook and Harlow et al., Antibodies: A Laboratory Manual, Chapter 14, Cold Spring Harbor Laboratory, Cold Spring Harbor (2013, 2nd edition).

[0103] definition Unless otherwise defined herein, scientific and technical terms used herein have the meanings commonly understood by those skilled in the art. In the event of any potential ambiguity, the definitions provided herein supersede any dictionary or external definitions. Unless otherwise required by context, singular terms shall include the plural and plural terms shall include the singular. The use of "or" means "and / or" unless otherwise specified. The use of "comprises" and other forms such as "including" is not limiting.

[0104] As used herein, the term "nucleic acid" refers to an RNA or DNA molecule composed of a chain of ribonucleotides or deoxyribonucleotides, respectively.

[0105] As used herein, the term "therapeutic nucleic acid" refers to a nucleic acid molecule (e.g., a ribonucleic acid) that has partial or complete complementarity with a disease-associated mRNA, interacts with the disease-associated mRNA, and mediates silencing of mRNA expression.

[0106] As used herein, the term "carrier nucleic acid" refers to a nucleic acid molecule (eg, ribonucleic acid) that is complementary to and hybridizes with a therapeutic nucleic acid.

[0107] As used herein, the term "3' end" refers to the end of a nucleic acid that contains an unmodified hydroxyl group on the 3' carbon of the ribose ring.

[0108] As used herein, the term "5' end" refers to the end of a nucleic acid that contains a phosphate group attached to the 5' carbon of the ribose ring.

[0109] As used herein, the term "nucleoside" refers to a molecule consisting of a heterocyclic base and its sugar.

[0110] As used herein, the term "nucleotide" refers to a nucleoside having a phosphate group on either the 3' or 5' sugar hydroxyl group of the nucleoside.

[0111] As used herein, the term "siRNA" refers to small interfering double-stranded RNA that triggers the RNA interference (RNAi) pathway. siRNA molecules can be of various lengths (generally 18-30 base pairs) and contain varying degrees of complementarity to their target mRNA. The term "siRNA" includes two separate duplexes as well as single strands that can form hairpin structures containing the duplex region.

[0112] As used herein, the term "antisense strand" refers to the strand of a double-stranded siRNA that contains a degree of complementarity to a target gene.

[0113] As used herein, the term "sense strand" refers to the strand of a double-stranded siRNA that contains complementarity to the antisense strand.

[0114] As used herein, the term "chemically modified nucleotide" or "nucleotide analog" or "altered nucleotide" or "modified nucleotide" refers to a non-standard nucleotide, including non-naturally occurring ribonucleotides or deoxyribonucleotides. Exemplary nucleotide analogs are modified at any position to alter the specific chemical properties of the nucleotide while maintaining the ability of the nucleotide to perform its intended function. Examples of nucleotide positions that can be derivatized include the 5-position, such as 5-(2-amino)propyluridine, 5-bromouridine, 5-propyneuridine, 5-propenyluridine, etc., as described in Herdewijn, Antisense Nucleic Acid Drug Dev., 2000 Aug. 10(4):297-310; the 6-position, such as 6-(2-amino)propyluridine; and the 8-position of adenosine and / or guanosine, such as 8-bromoguanosine, 8-chloroguanosine, 8-fluoroguanosine, etc. Nucleotide analogs also include deazanucleotides, such as 7-deaza-adenosine; O- and N-modified (e.g., alkylated, e.g., N6-methyladenosine or others known in the art) nucleotides; and other heterocyclic-modified nucleotide analogs.

[0115] Nucleotide analogs may also include modifications to the sugar portion of the nucleotide. For example, the 2'OH group may be replaced with a group selected from H, OR, R, F, Cl, Br, I, SH, SR, NH, NHR, NR, COOR, or OR, where R is a substituted or unsubstituted C-C alkyl, alkenyl, alkynyl, aryl, etc. Other possible modifications include those described in U.S. Patent Nos. 5,858,988 and 6,291,438.

[0116] As used herein, the term "metabolically stabilized" refers to an RNA molecule that contains ribonucleotides that have been chemically modified from a 2'-hydroxyl group to a 2'-O-methyl group.

[0117] As used herein, the term "phosphorothioate" refers to a phosphate group of a nucleotide that has been modified by replacing one or more of the oxygen atoms of the phosphate group with sulfur.

[0118] As used herein, the term "ethylene glycol chain" refers to a carbon chain having the formula ((CH2OH)2).

[0119] The term "alkyl chain" as used herein refers to an acyclic unsaturated hydrocarbon chain. In the context of the present invention, "alkyl chain" includes, but is not limited to, straight chain, branched chain, and cyclic unsaturated hydrocarbons.

[0120] As used herein, the term "amide" refers to an alkyl or aromatic group attached to an aminocarbonyl functional group.

[0121] As used herein, the terms "internucleoside" and "internucleotide" refer to linkages between nucleosides and between nucleotides, respectively.

[0122] As used herein, the term "triazole" refers to a heterocyclic compound having the formula (C2H3N3), which has a five-membered ring of two carbons and three nitrogens that can vary in position to give rise to multiple isomers.

[0123] As used herein, the term "end group" refers to the group at the end of a carbon chain or nucleic acid.

[0124] As used herein, the term "lipid-soluble amino acid" refers to an amino acid that contains a hydrophobic moiety (eg, an alkyl chain or an aromatic ring).

[0125] As used herein, the term "antagomir" refers to a nucleic acid that can function as an inhibitor of miRNA activity.

[0126] As used herein, the term "gapmer" refers to a chimeric antisense nucleic acid that contains a central block of deoxynucleotide monomers of sufficient length to induce RNase H cleavage. The deoxynucleotide block is flanked by ribonucleotide monomers or ribonucleotide monomers containing modifications.

[0127] As used herein, the term "mixmer" refers to a nucleic acid that comprises a mixture of locked nucleic acid (LNA) and DNA.

[0128] As used herein, the term "guide RNA" refers to a nucleic acid that has complementarity to a specific sequence in the genome adjacent to or one base upstream of a protospacer adjacent motif (PAM) sequence, such as that used in the CRISPR / Cas9 gene editing system.

[0129] As used herein, the term "delivery target" refers to an organ or part of the body to which it is desired to deliver a branched oligonucleotide composition.

[0130] As used herein, the term "di-siRNA" refers to a molecule of the present invention that comprises a branched oligonucleotide structure and includes an siRNA molecule as the therapeutic nucleic acid.

[0131] As used herein, the term "amino acid" refers to a molecule containing an amine and a carboxyl functional group and a side chain (R) characteristic of an amino acid. In certain embodiments, an amino acid has the formula: [ka] It has the following structure.

[0132] In another embodiment, the "amino acid" has the formula: [ka] It may refer to the constituent residues of a peptide or protein having the structure:

[0133] In some embodiments, the amino acid is selected from the group of proteinogenic amino acids. In other embodiments, the amino acid is an L-amino acid or a D-amino acid. In other embodiments, the amino acid is a synthetic amino acid (e.g., a β-amino acid).

[0134] Certain internucleotide linkages described herein (e.g., phosphodiester and phosphorothioate) are understood to contain a formal charge of −1 at physiological pH, which is counterbalanced by a cationic moiety, e.g., an alkali metal such as sodium or potassium, an alkaline earth metal such as calcium or magnesium, or an ammonium or guanidinium ion.

[0135] The phosphate group of a nucleotide can also be modified by replacing one or more oxygen atoms of the phosphate group with sulfur (e.g., phosphorothioate), or by making other substitutions that allow the nucleotide to perform its intended function, as described, for example, in Eckstein, Antisense Nucleic Acid Drug Dev. 2000 Apr. 10(2):117-21, Rusckowski et al. Antisense Nucleic Acid Drug Dev. 2000 Oct. 10(5):333-45, Stein, Antisense Nucleic Acid Drug Dev. 2001 Oct. 11(5):317-25, Vorobjev et al. Antisense Nucleic Acid Drug Dev. 2001 Apr. 11(2):77-85, and U.S. Patent Application No. 5,684,143. Some of the modifications referenced above (eg, phosphate group modifications) preferably reduce the rate of in vivo or in vitro hydrolysis of, for example, a polynucleotide containing the analog.

[0136] Delivery and Distribution In another aspect, there is provided a method for selectively delivering a nucleic acid described herein to an organ of a patient, comprising administering to the patient a branched oligonucleotide described herein such that the nucleic acid is selectively delivered. In one embodiment, the organ is the liver. In another embodiment, the organ is the kidney. In another embodiment, the organ is the spleen. In another embodiment, the organ is the heart. In another embodiment, the organ is the brain. In another embodiment, the nucleic acid.

[0137] The compositions described herein facilitate simple, effective, non-toxic delivery of metabolically stable oligonucleotides (e.g., siRNAs) and potent silencing of therapeutic targets in a range of tissues in vivo.

[0138] As shown in Figure 11, di-siRNA is distributed throughout the injected hemisphere of the mouse brain after intrastriatal injection. Single-stranded, unconjugated siRNA can silence mRNA in primary neurons, and the di-siRNA structure is essential for enhanced tissue distribution and retention of modified oligonucleotides. Other conjugates, such as cholesterol, are retained but rapidly diffuse away from the injection site. The slight hydrophobicity of the two single-stranded phosphorothioate tails aids tissue retention while also allowing widespread and uniform distribution throughout the injected ipsilateral hemisphere of the brain.

[0139] As shown in Figure 12, a single injection of di-siRNA was detected both ipsilaterally and contralaterally to the injection site, indicating that diffusion is not limited to the injected hemisphere but also occurs across the midline to the uninjected hemisphere. Other injection methods, including intracerebroventricular, can also promote bilateral distribution with just a single injection.

[0140] Di-siRNAs exhibit a highly distinctive cellular distribution when injected intrastriatally into the brain. Fluorescently labeled di-siRNAs appear to preferentially localize to neurons in the cerebral cortex. This selective feature is specific to these compounds and not to other siRNA conjugates, such as cholesterol, that do not exhibit cell type preference.

[0141] Di-siRNAs show localization to fiber tracts in the striatum but are present within neuronal cell bodies in the cerebral cortex. Transport to the cerebral cortex may be by diffusion or retrograde transport along striatal fiber tracts. The theory that retrograde transport is partially important is supported by the fact that some regions of the cerebral cortex show significant neuronal penetration, while neurons in neighboring regions show no di-siRNA uptake.

[0142] A single therapeutically relevant brain injection of di-siRNA results in widespread distribution of di-siRNA throughout the brain. The distribution volumes shown in Figures 31-32 are unprecedented in the prior art and represent a promising therapeutic delivery system for di-siRNA.

[0143] After a single intravenous injection, di-siRNA shows widespread distribution throughout the body. As shown in Figure 37, large amounts of di-siRNA were detected in the liver, skin, brain, kidney, spleen, pancreas, colon, fat, lung, muscle, and thymus of mice. The finding that di-siRNA is present in the brain after intravenous injection also indicates that the di-siRNA structure efficiently crosses the blood-brain barrier.

[0144] Silencing In some embodiments, compounds of the present invention promote about 90% striatal silencing and about 65% cortical silencing in vivo after a single injection in the brain without any signs of toxicity.In some embodiments, compounds of the present invention show about 60% silencing over the entire spinal cord area after intrathecal injection.

[0145] A single injection of di-siRNA induces robust silencing in both the striatum and cerebral cortex of the mouse brain. This level of efficacy has not been previously demonstrated with unconjugated siRNA. While di-siRNA appears to be visually taken up by fiber tracts within the striatum, the observed efficacy clearly indicates that striatal neurons internalize di-siRNA in large quantities. In this experiment, 2 nmol of di-siRNA (4 nmol of the corresponding antisense HTT strand) was injected intrastriatally. Seven days after injection, animals were sacrificed. Tissue punches were collected from 300 μm brain sections of the striatum and cerebral cortex. The presence of di-siRNA antisense strands in various brain regions, liver, and kidney was quantified using Cy3-labeled complementary PNAs to hybridize with the strands and HPLC to quantify the amount of oligonucleotide per mg of tissue.

[0146] As shown in Figure 10, di-siRNA showed comparable efficacy compared to single duplexes after lipid-mediated transfection in HeLa cells, indicating that RISC transduction was not hindered by linking two siRNA duplexes. Although di-siRNA was not effective in HeLa cells without transfection, a single phosphorothioate tail was sufficient to induce at least 60% silencing in primary cortical neurons, demonstrating that phosphorothioation is an effective method for delivering siRNA to primary neurons without formulation.

[0147] As shown in Figure 13, a single administration of di-siRNA induces robust silencing in both the striatum and cerebral cortex of the mouse brain. 63X images of pyramidal neurons containing Cy3-labeled di-branched oligos show that di-siRNA appears to be visually taken up into fiber tracts in the striatum, while the clearly observed efficacy indicates that striatal neurons are absorbing the di-siRNA in large amounts.

[0148] As shown in Figure 14, di-siRNA exhibits strong and uniform silencing throughout the spinal cord after intrathecal injection. A single injection of di-siRNA in the lumbar region of the spinal cord silences mRNA to a similar extent in the cervical, thoracic, and lumbar regions, demonstrating a uniform and long-range distribution.

[0149] As shown in Figure 24, Cy3-labeled di-siRNA induces strong silencing in both the striatum and cerebral cortex of the mouse brain. The mRNA expression levels indicate that the addition of Cy3 to the branched oligonucleotide composition enhances silencing compared to unlabeled di-siRNA.

[0150] As shown in Figures 23-24, a single intrastriatal injection resulted in silencing in both the cerebral cortex and striatum of the mouse brain, but not in the liver or kidney, demonstrating that branched oligonucleotides can specifically target organs of interest.

[0151] As shown in Figure 26, a single injection of di-siRNA continues to maintain strong silencing in both the striatum and cerebral cortex of the mouse brain for two weeks after injection. Di-siRNA is stable and effective in vivo for at least two weeks.

[0152] As shown in Figures 33 and 34, a single injection of therapeutically relevant di-siRNA induces significant silencing in multiple brain regions. This is the first example of widespread siRNA silencing in the brain after a single therapeutically relevant injection. These results indicate that di-siRNA is an effective option for RNA therapy.

[0153] Modified RNA silencing agents In certain embodiments of the present invention, the above-mentioned RNA silencing agent of the present invention (or any part thereof) can be modified to further improve the activity of the agent.For example, the above-mentioned RNA silencing agent can be modified with any of the following modifications.Modifications can be used in part to further enhance target discrimination, enhance agent stability (for example, prevent degradation), promote cellular uptake, enhance target efficacy, improve binding (for example, to target) efficacy, improve patient tolerance to the agent and / or reduce toxicity.

[0154] 1) Modifications to enhance target discrimination In certain embodiments, the RNA silencing agent of the present invention can be substituted with destabilizing nucleotides to enhance single nucleotide target discrimination (see U.S. Patent Application No. 11 / 698,689, filed January 25, 2007, and U.S. Provisional Application No. 60 / 762,225, filed January 25, 2006, which are incorporated herein by reference). Such modifications can be sufficient to abolish the specificity of the RNA silencing agent for non-target mRNA (e.g., wild-type mRNA) without significantly affecting the specificity of the RNA silencing agent for target mRNA (e.g., gain-of-function mutant mRNA).

[0155] In a preferred embodiment, the RNA silencing agent of the present invention is modified by introducing at least one universal nucleotide into its antisense strand. A universal nucleotide contains a base moiety that can indiscriminately base pair with any of the four standard nucleotide bases (e.g., A, G, C, U). Universal nucleotides are preferred because they have relatively little effect on the stability of the duplex formed by the double-stranded RNA or the guide strand of the RNA silencing agent and the target mRNA. Exemplary universal nucleotides are selected from nucleotides having an inosine base moiety or an inosine analog base moiety selected from the group consisting of deoxyinosine (e.g., 2'-deoxyinosine), 7-deaza-2'-deoxyinosine, 2'-aza-2'-deoxyinosine, PNA-inosine, norphorino-inosine, LNA-inosine, phosphoramidate-inosine, 2'-O-methoxyethyl-inosine, and 2'-OMe-inosine. In a particularly preferred embodiment, the universal nucleotide is an inosine residue or a naturally occurring analog thereof.

[0156] In certain embodiments, the RNA silencing agent of the present invention is modified by introducing at least one destabilizing nucleotide within 5 nucleotides from the specificity-determining nucleotide (i.e., the nucleotide that recognizes disease-related polymorphism).For example, the destabilizing nucleotide can be introduced within 5 nucleotides, 4 nucleotides, 3 nucleotides, 2 nucleotides or 1 nucleotide from the specificity-determining nucleotide.In exemplary embodiments, the destabilizing nucleotide is introduced within 3 nucleotides from the specificity-determining nucleotide (i.e., there are two destabilizing nucleotides between the destabilizing nucleotide and the specificity-determining nucleotide).In the RNA silencing agent having a double strand or a part of a strand (e.g., siRNA and shRNA), the destabilizing nucleotide can be introduced into the strand or part of a strand that does not contain the specificity-determining nucleotide.In a preferred embodiment, the destabilizing nucleotide is introduced into the same strand or part of a strand that contains the specificity-determining nucleotide.

[0157] 2) Modifications to enhance efficacy and specificity In certain embodiments, the RNA silencing agent of the present invention can be modified to facilitate enhanced efficacy and specificity in mediated RNAi through asymmetric design rules (see U.S. Patent Nos. 8,309,704, 7,750,144, 8,304,530, 8,329,892 and 8,309,705). Such modifications facilitate the introduction of the sense strand of the antisense strand of siRNA (for example, siRNA produced from the siRNA or shRNA designed using the method of the present invention) into RISC in favor of the sense strand, thereby increasing and improving the effectiveness of target cleavage and silencing.Preferably, the asymmetry of the RNA silencing agent is enhanced by reducing the base pair strength between the 5' end of the antisense strand (AS5') and the 3' end of the sense strand (S3') of the RNA silencing agent, compared with the binding strength or base pair strength between the 3' end of the antisense strand (AS3') and the 5' end of the sense strand (S5') of the RNA silencing agent.

[0158] In one embodiment, the asymmetry of the RNA silencing agent of the present invention can be strengthened so that there are fewer G:C base pairs between the 5' end of the first or antisense strand and the 3' end of the 3' sense strand portion than between the 3' end of the first or antisense strand and the 5' end of the sense strand portion. In another embodiment, the asymmetry of the RNA silencing agent of the present invention can be strengthened so that there is at least one mismatched base between the 5' end of the first or antisense strand and the 3' end of the 3' sense strand portion. Preferably, the mismatched base is selected from the group consisting of G:A, C:A, C:U, G:G, A:A, C:C, and U:U. In another embodiment, the asymmetry of the RNA silencing agent of the present invention can be strengthened so that there is at least one wobble base pair, for example, G:U, between the 5' end of the first or antisense strand and the 3' end of the 3' sense strand portion. In another embodiment, the asymmetry of the RNA silencing agent of the present invention can be strengthened so that there is at least one base pair containing a rare nucleotide, for example, inosine (I). Preferably, the base pair is selected from the group consisting of I:A, I:U and I:C. In yet another embodiment, the asymmetry of the RNA silencing agent of the present invention can be enhanced by having at least one base pair containing a modified nucleotide. In a preferred embodiment, the modified nucleotide is selected from the group consisting of 2-amino-G, 2-amino-A, 2,6-diamino-G and 2,6-diamino-A.

[0159] 3) RNA silencing agents with improved stability The RNA silencing agent of the present invention can be modified to improve stability in serum or growth medium for cell culture.To enhance stability, 3'-residues can be stabilized against degradation, for example, they can be selected to consist of purine nucleotides, particularly adenosine or guanosine nucleotides.Alternatively, substitution of pyrimidine nucleotides with modified analogs, for example, substitution of uridine with 2'-deoxythymidine, is tolerated and does not affect the efficiency of RNA interference.

[0160] In a preferred embodiment, the present invention relates to an RNA silencing agent comprising a first strand and a second strand modified by replacing internal nucleotides with modified nucleotides so as to enhance in vivo stability compared to the corresponding unmodified RNA silencing agent.As defined herein, an "internal" nucleotide is a nucleotide occurring at any position other than the 5'-end or 3'-end of a nucleic acid molecule, polynucleotide, or oligonucleotide.Internal nucleotides can be present in a single-stranded molecule or in one strand of a double-stranded or double-stranded molecule.In some embodiments, the sense strand and / or antisense strand are modified by replacing at least one internal nucleotide.In other embodiments, the sense strand and / or antisense strand are modified by replacing at least 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, or more internal nucleotides. In another embodiment, the sense and / or antisense strands are modified by at least 5%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95% or more internal nucleotide substitutions. In yet another embodiment, the sense and / or antisense strands are modified by all internal nucleotide substitutions.

[0161] In a preferred embodiment of the present invention, the RNA silencing agent comprises at least one modified nucleotide analogue.The nucleotide analogue can be located at a position where the target-specific silencing activity, for example, RNAi-mediated activity or translational repression activity, is not substantially affected, for example, at the 5'-end and / or 3'-end of the siRNA molecule.In particular, the end can be stabilized by incorporating modified nucleotide analogue.

[0162] Exemplary nucleotide analogs include sugar-modified and / or backbone-modified ribonucleotides (i.e., modifications to the phosphate-sugar backbone). For example, the phosphodiester bond of natural RNA can be modified to include at least one nitrogen or sulfur heteroatom. In exemplary backbone-modified ribonucleotides, the phosphoester group linked to adjacent ribonucleotides is replaced with a modified group, for example, a phosphothioate group. In exemplary sugar-modified ribonucleotides, the 2'OH group is replaced with a group selected from H, OR, R, halo, SH, SR, NH, NHR, NR, or ON (wherein R is C-C alkyl, alkenyl, or alkynyl, and halo is F, Cl, Br, or I).

[0163] In specific embodiments, the modification is 2'-fluoro, 2'-amino and / or 2'-thio modification. Particularly preferred modifications include 2'-fluoro-cytidine, 2'-fluoro-uridine, 2'-fluoro-adenosine, 2'-fluoro-guanosine, 2'-amino-cytidine, 2'-amino-uridine, 2'-amino-adenosine, 2'-amino-guanosine, 2,6-diaminopurine, 4-thio-uridine and / or 5-amino-allyl-uridine. In certain embodiments, all 2'-fluoro ribonucleotides are uridine and cytidine. Further exemplary modifications include 5-bromo-uridine, 5-iodo-uridine, 5-methyl-cytidine, ribo-thymidine, 2-aminopurine, 2'-amino-butyryl-pyrene-uridine, 5-fluoro-cytidine and 5-fluoro-uridine. 2'-deoxy-nucleotides and 2'-OMe nucleotides can be used in the modified RNA silencing agent moiety of the present invention. Additional modified residues include deoxy-abasic, inosine, N3-methyl-uridine, N6,N6-dimethyl-adenosine, pseudouridine, purine ribonucleoside and ribavirin. In a particularly preferred embodiment, the 2' moiety is a methyl group, so that the binding moiety is a 2'-O-methyl oligonucleotide.

[0164] In an exemplary embodiment, the RNA silencing agent of the present invention comprises a locked nucleic acid (LNA). LNA is resistant to nuclease activity (highly stable) and contains sugar-modified nucleotides with single-nucleotide discrimination for mRNA (Elmen et al., Nucleic Acids Res., (2005), 33(1): 439-447; Braasch et al. (2003) Biochemistry 42:7967-7975, Petersen et al. (2003) Trends Biotechnol 21:74-81). These molecules include 2'-O,4'-C-ethylene-bridged nucleic acids, including possible modifications such as 2'-deoxy-2''-fluorouridine. Furthermore, LNA increases the specificity of oligonucleotides by constraining the sugar moiety to a 3'-endo configuration, thereby pre-arranging the nucleotide for base pairing and increasing the melting temperature of oligonucleotides by up to 10°C per base.

[0165] In another exemplary embodiment, the RNA silencing agent of the present invention comprises a peptide nucleic acid (PNA). PNA comprises modified nucleotides in which the sugar-phosphate moiety of the nucleotide is replaced with a neutral 2-aminoethylglycine moiety, which can form a polyamide backbone that is highly resistant to nuclease digestion and confers improved molecular binding specificity (Nielsen, et al., Science, (2001), 254: 1497-1500).

[0166] Also preferred is nucleic acid modified ribonucleotide, that is, nucleotide that contains at least one non-natural nucleic acid base instead of naturally occurring nucleic acid base.The base can be modified to block the activity of adenosine deaminase.Exemplary modified nucleic acid bases include but are not limited to uridine and / or cytidine modified at 5-position, such as 5-(2-amino)propyluridine, 5-bromouridine; adenosine and / or guanosine modified at 8-position, such as 8-bromoguanosine; deazanucleotide, such as 7-deazaadenosine; O- and N-alkylated nucleotide, such as N6-methyladenosine.It should be noted that the above modifications can be combined.

[0167] In other embodiments, crosslinking can be used to change the pharmacokinetics of RNA silencing agents, for example, to increase their half-life in the body.Therefore, the present invention includes RNA silencing agents having two crosslinked complementary strands of nucleic acid.The present invention also includes RNA silencing agents that are conjugated or not (e.g., at the 3' end) to other moieties (e.g., non-nucleic acid moieties such as peptides), organic compounds (e.g., dyes), etc.).Modifying siRNA derivatives in this way can improve cellular uptake compared to corresponding siRNAs, or enhance the cell targeting activity of the resulting siRNA derivatives, which is useful for tracking siRNA derivatives in cells compared to corresponding siRNAs, or improve the stability of siRNA derivatives.

[0168] Other exemplary modifications include: (a) 2' modifications, such as providing a 2'OMe portion of U in the sense strand or antisense strand, particularly the sense strand, or providing a 3' overhang, such as providing a 2'OMe portion at the 3' end (3' end means the 3' atom or most 3' portion of the molecule, for example, most 3'P or 2' position as indicated by the context); (b) modifications of the backbone, such as substituting S for O in a phosphate backbone, for example, U or A or both, particularly providing phosphorothioate modifications in the antisense strand, for example, substituting S for P; (c) substituting a C5 amino linker for U; (d) substituting G for A (the sequence change is preferably located in the sense strand rather than the antisense strand); and (d) modifications at the 2', 6', 7', or 8' positions. Exemplary embodiments are those in which one or more of these modifications are present in the sense strand rather than the antisense strand, or in which the antisense strand has few such modifications. Still other exemplary modifications include a 3' overhang, e.g., the use of a methylated P at the 3' end; a combination of 2' modifications, e.g., providing a 2'OMe moiety and a modification of the backbone, e.g., a modification by substituting P with S, e.g., providing a phosphorothioate modification, or a 3' overhang, e.g., the use of a methylated P at the 3' end; a modification having a 3' alkyl; a modification having a 3' overhang, e.g., a non-basic pyrrolidone at the 3' end; a modification having naproxen, ibuprofen, or other moieties that exhibit degradation at the 3' end.

[0169] 4) Modifications to enhance cellular uptake In other embodiments, the compounds of the invention are modified with chemical moieties, e.g., to enhance cellular uptake by target cells (e.g., neuronal cells). Thus, the invention includes RNA silencing agents conjugated or unconjugated (e.g., at the 3' end) to other moieties (e.g., non-nucleic acid moieties such as peptides), organic compounds (e.g., dyes), etc. Conjugation can be achieved by methods known in the art, for example, by using the methods of Lambert et al., Drug Deliv. Rev.: 47(1), 99-112 (2001) (describing nucleic acids incorporated into polyalkylcyanoacrylate (PACA) nanoparticles); Fattal et al., J. Control Release 53(1-3):137-43 (1998) (describing nucleic acids bound to nanoparticles); Schwab et al., Ann. Oncol. 5 Suppl. 4:55-8 (1994) (describing nucleic acids bound to intercalating agents, hydrophobic groups, polycations, or PACA nanoparticles); and Godard et al., Eur. J. Biochem. 232(2):404-10 (1995) (describing nucleic acids bound to nanoparticles).

[0170] In certain embodiments, the compounds of the present invention are conjugated to a lipophilic moiety. In some embodiments, the lipophilic moiety is a ligand containing a cationic group. In another embodiment, the lipophilic moiety is bound to one or both strands of the siRNA. In an exemplary embodiment, the lipophilic moiety is bound to one end of the sense strand of the siRNA. In another exemplary embodiment, the lipophilic moiety is bound to the 3' end of the sense strand. In certain embodiments, the lipophilic moiety is selected from the group consisting of cholesterol, vitamin D, DHA, DHAg2, EPA, vitamin E, vitamin K, vitamin A, folic acid, or cationic dyes (e.g., Cy3).

[0171] 5) Bound Ligand Other substances can be attached to the compounds of the present invention. For example, ligands can be attached to RNA silencing agents to improve stability, thermodynamics of hybridization with target nucleic acids, targeting to specific tissues or cell types, or cell permeability via endocytosis-dependent or -independent mechanisms. Ligands and related modifications can also increase sequence specificity, thereby reducing off-site targeting. The attached ligand can contain one or more modified bases or sugars that can function as intercalators. These are preferably located in intercalation regions, such as bulges, of the RNA silencing agent / target strand. Intercalators can be aromatic, for example, polycyclic aromatic or heterocyclic aromatic compounds. Polycyclic intercalators can have stacking capabilities and can include two, three, or four fused ring structures. The universal bases described herein can be included in the ligand. In some embodiments, the ligand can contain a cleavage group that contributes to target gene inhibition by cleaving the target nucleic acid. The cleavage group can be, for example, bleomycin (e.g., bleomycin-A5, bleomycin-A2, or bleomycin-B2), pyrene, phenanthroline (e.g., O-phenanthroline), polyamine, tripeptide (e.g., lys-tyr-lys tripeptide), or metal ion. The metal ion chelating group can include, for example, Lu(III) or EU(III) macrocyclic complexes, Zn(II) 2,9-dimethylphenanthroline derivatives, Cu(II) terpyridine, or acridine, which can promote selective cleavage of target RNA at the bulge region with free metal ions such as Lu(III). In some embodiments, a peptide ligand can be attached to an RNA silencing agent, for example, at the bulge region, to promote cleavage of target RNA. For example, 1,8-dimethyl-1,3,6,8,10,13-hexaazacyclotetradecane (cyclam) can be conjugated to a peptide (e.g., via an amino acid derivative) to promote target RNA cleavage.The conjugated ligand can be an aminoglycoside ligand, which provides the RNA silencing agent with improved hybridization properties or improved sequence specificity. Exemplary aminoglycosides include glycosylated polylysine, galactosylated polylysine, neomycin B, tobramycin, kanamycin A, and acridine conjugates of aminoglycosides (e.g., neo-N-acridine, neo-S-acridine, neo-C-acridine, tobra-N-acridine, and kana-AN-acridine). The use of an acridine analog can increase sequence specificity. For example, neomycin B has a higher affinity for RNA compared to DNA, but lower sequence specificity. The acridine analog, neo-5-acridine, has increased affinity for the HIV Rev response element (RRE). In some embodiments, a guanidine analog (guanidinoglycoside) of an aminoglycoside ligand is conjugated to the RNA silencing agent. In a guanidinoglycoside, the amine group of an amino acid is replaced with a guanidine group. The attachment of a guanidine analog can enhance the cell permeability of the RNA silencing agent. The attached ligand can be a polyarginine peptide, peptoid, or peptidomimetic, which can enhance the cellular uptake of the oligonucleotide agent.

[0172] Exemplary ligands can be directly or indirectly coupled to the ligand-conjugated carrier by a linkage between them, preferably covalently.In exemplary embodiments, the ligand is coupled to the carrier by a linkage between them.In exemplary embodiments, the ligand changes the distribution, targeting or life span of the RNA silencing agent that is incorporated.In exemplary embodiments, the ligand provides enhanced affinity to selected target, for example, molecule, cell or cell type, compartment, for example, cell or organ compartment, tissue, organ or body region, compared with species that do not have such ligand.

[0173] Exemplary ligands can improve transport, hybridization, and specificity properties, and can also improve the nuclease resistance of the resulting natural or modified RNA silencing agent or polymer molecule containing any combination of monomers and / or natural or modified ribonucleotides described herein. Ligands generally include therapeutic modifications, such as modifications to enhance uptake; diagnostic compounds or reporter groups, such as for monitoring distribution; cross-linking agents; nuclease-resistant moieties; and natural or unnatural nucleic acid bases. Common examples include lipophilic substances, lipids, steroids (e.g., uvaol, hecogenin, diosgenin), terpenes (e.g., triterpenes, such as sarsasapogenin, friedelin, epifriedelanol-derivatized lithocholic acid), vitamins (e.g., folic acid, vitamin A, biotin, pyridoxal), carbohydrates, proteins, protein-binding agents, integrin-targeting molecules, polycationic substances, peptides, polyamines, and peptidomimetics. Ligands can include naturally occurring substances (e.g., human serum albumin (HSA), low-density lipoprotein (LDL) or globulins); carbohydrates (e.g., dextran, pullulan, chitin, chitosan, inulin, cyclodextrin or hyaluronic acid); amino acids or lipids. Ligands can also be recombinant or synthetic molecules, such as synthetic polymers, e.g., synthetic polyamino acids. Examples of polyamino acids include polylysine (PLL), poly-L-aspartic acid, poly-L-glutamic acid, styrene-maleic anhydride copolymer, poly(L-lactide-co-glycolide) copolymer, divinyl ether-maleic anhydride copolymer, N-(2-hydroxypropyl)methacrylamide copolymer (HMPA), polyethylene glycol (PEG), polyvinyl alcohol (PVA), polyurethane, poly(2-ethylacrylic acid), N-isopropylacrylamide polymer or polyphosphazine.Examples of polyamines include polyethyleneimine, polylysine (PLL), spermine, spermidine, polyamines, pseudopeptide-polyamines, peptidomimetic polyamines, dendrimeric polyamines, arginine, amidine, protamine, cationic lipids, quaternary salts of cationic porphyrin polyamines or alpha-helical peptides.

[0174] The ligand can also include a targeting group, such as a cell or tissue targeting agent, such as a lectin, glycoprotein, lipid, or protein, such as an antibody, which binds to specific cells, such as kidney cells. The targeting group can be tiotropin, melanotropin, lectin, glycoprotein, surfactant protein A, mucin hydrocarbon, polyvalent lactose, polyvalent galactose, N-acetyl-galactosamine, N-acetyl-glucosamine, polyvalent mannose, polyvalent fucose, glycosylated polyamino acids, polyvalent galactose, transferrin, bisphosphonate, polyglutamate, polyaspartate, lipid, cholesterol, steroid, bile acid, folic acid, vitamin B12, biotin, or an RGD peptide or RGD peptidomimetic. Other examples of ligands include dyes, intercalating agents (e.g., acridine and substituted acridine), crosslinkers (e.g., psoralens, mitomycin C), porphyrins (TPPC4, texaphyrin, sapphyrin), polycyclic aromatic hydrocarbons (e.g., phenazine, dihydrophenazine, phenanthroline, pyrene), lys-tyr-lys tripeptides, aminoglycosides, guanidinium aminoglycosides, artificial endonucleases (e.g., EDTA), lipid-soluble molecules, e.g., cholesterol (and its thioanalogues), cholic acid, cholanic acid, lithocholic acid, adamantaneacetic acid, 1-pyrenebutyric acid, dihydrotestosterone, glycerol (e.g., esters (e.g., mono-, bis-, or tris-fatty acid esters, e.g., C 10 , C 11 , C 12 , C 13 , C 14 , C 15 , C 16 , C 17 , C 18 , C 19 or C20 fatty acids) and their ethers, e.g., C 10 , C 11 , C 12 , C 13 , C 14 , C 15 , C 16 , C 17 , C 18 , C 19 , or C 20 alkyl; e.g., 1,3-bis-O(hexadecyl)glycerol, 1,3-bis-O(octadecyl)glycerol, geranyloxyhexyl group, hexadecylglycerol, borneol, menthol, 1,3-propanediol, heptadecyl group, palmitic acid, stearic acid (e.g., glyceryl distearate), oleic acid, myristic acid, O3-(oleoyl)lithocholic acid, O3-(oleoyl)cholenic acid, dimethoxytrityl or phenoxazine) and peptide conjugates (e.g., Antennapedia peptide, Tat peptide The nucleotide sequence may be a nucleotide sequence of ...

[0175] Ligands can be proteins, e.g., glycoproteins, or peptides, e.g., molecules with specific affinity for a co-ligand, or antibodies, e.g., antibodies that bind to specific cell types such as cancer cells, endothelial cells, or bone cells. Ligands can also include hormones and hormone receptors. They can also include lipids, lectins, carbohydrates, vitamins, cofactors, and non-peptide species such as multivalent lactose, multivalent galactose, N-acetyl-galactosamine, N-acetyl-glucosamine, multivalent mannose, or multivalent fucose. Ligands can be, for example, lipopolysaccharides, p38 MAP kinase activators, or NF-κB activators.

[0176] The ligand can be a drug that can increase the cellular uptake of the RNA silencing agent by disrupting the substrate, for example, the cytoskeleton, for example, by disrupting cellular microtubules, microfilaments, and / or filaments. The drug can be, for example, taxon, vincristine, vinblastine, cytochalasin, nocodazole, jasplakinolide, latrunculin A, phalloidin, swinholide A, indanocine, or myoservin. The ligand can increase the cellular uptake of the RNA silencing agent, for example, by activating an inflammatory response. Exemplary ligands with such effects include tumor necrosis factor alpha (TNFα), interleukin-1β, or gamma interferon. In some embodiments, the ligand is a lipid or lipid-based molecule. Such lipid or lipid-based molecules preferably bind to serum proteins, such as human serum albumin (HSA). HSA-binding ligands allow the distribution of the conjugate to target tissues, such as non-renal target tissues of the body. For example, the target tissue can be the liver, including liver parenchymal cells. Other molecules that can bind to HSA can also be used as ligands. For example, neproxin or aspirin can be used. Lipids or lipid-based ligands can (a) increase the resistance of conjugates to degradation, (b) increase targeting or transport to target cells or cell membranes, and / or (c) can be used to adjust binding to serum proteins, such as HSA. Lipid-based ligands can be used to regulate, for example, control the binding of conjugates to target tissues. For example, lipids or lipid-based ligands that bind more strongly to HSA are less likely to be targeted to the kidney and therefore less likely to be excreted from the body. Lipids or lipid-based ligands that do not bind strongly to HSA can be used to target conjugates to the kidney. In a preferred embodiment, lipid-based ligands bind to HSA. Lipid-based ligands can bind to HSA with sufficient affinity so that they are preferably distributed in non-renal tissues. However, it is preferable that the affinity is not so strong that HSA-ligand binding is irreversible.In another preferred embodiment, the lipid-based ligand binds weakly or not at all to HSA such that the conjugate is preferably distributed to the kidney. Other moieties that target kidney cells can also be used in place of or in addition to the lipid-based ligand.

[0177] In another embodiment, the ligand is a moiety, for example, a vitamin that is taken up by target cells, for example, proliferating cells. These are particularly useful for treating disorders characterized by unwanted cell proliferation, for example, malignant or non-malignant, for example, cancer cells. Exemplary vitamins include vitamins A, E, and K. Other exemplary vitamins include B vitamins, for example, folic acid, B12, riboflavin, biotin, pyridoxal, or other vitamins or nutrients that are taken up by cancer cells. HSA and low-density lipoprotein (LDL) are also included.

[0178] In another embodiment, the ligand is a cell-penetrating agent, preferably a helical cell-penetrating agent. Preferably, the agent is amphipathic. An exemplary agent is a peptide such as tat or antennapedia. If the agent is a peptide, it can be modified, including the use of peptidomimetics, invertomers, non-peptides, pseudopeptides or pseudopeptides, and D-amino acids. The helical agent is preferably an α-helical agent, and preferably has a lipophilic and lipophilic phase.

[0179] The ligand can be a peptide or peptidomimetic. Peptidomimetics (also referred to herein as oligopeptidomimetics) are molecules capable of folding into specific three-dimensional structures similar to natural peptides. Attachment of peptides and peptidomimetics to oligonucleotide agents can affect the pharmacokinetic distribution of RNA silencing agents, for example, by enhancing cellular recognition and uptake. The peptide or peptidomimetic moiety can be about 5 to 50 amino acids in length, e.g., about 5, 10, 15, 20, 25, 30, 35, 40, 45, or 50 amino acids in length. The peptide or peptidomimetic can be, for example, a cell-penetrating peptide, a cationic peptide, an amphipathic peptide, or a hydrophobic peptide (e.g., composed primarily of Tyr, Trp, or Phe). The peptide moiety can be a dendrimeric peptide, a conformation-directed peptide, or a cross-linked peptide. The peptide moiety can be an L-peptide or a D-peptide. In another alternative, the peptide moiety can include a hydrophobic membrane translocating sequence (MTS). Peptides or peptidomimetics can be encoded by random sequences of DNA, such as peptides identified from phage display libraries or one-bead-one-compound (OBOC) combinatorial libraries (Lam et al., Nature 354:82-84, 1991). In exemplary embodiments, the peptide or peptidomimetic linked to the RNA silencing agent by the incorporated monomer unit is a cell-targeting peptide, such as an arginine-glycine-aspartic acid (RGD)-peptide or RGD mimic. The peptide portion can range in length from about 5 amino acids to 40 amino acids. The peptide portion can have structural modifications, for example, to increase stability or direct conformational properties. Any of the structural modifications described below can be utilized. [Example]

[0180] Example 1 Chemical synthesis of di-RNA and vitamin D-conjugated hsiRNA The di-siRNA used in in vitro and in vivo efficacy evaluations was synthesized as follows. As shown in Figure 2, triethylene glycol was reacted with acrylonitrile to introduce a protected amine functional group. The branch point was then added as a tosylated solketal, and the nitrile was reduced to yield a primary amine that was subsequently attached to vitamin D (calciferol) via a carbamate linker. The ketal was then hydrolyzed to release a cis-diol selectively protected at the primary hydroxyl with a dimethoxytrityl (DMTr) protecting group and succinylated with succinic anhydride. The resulting moiety was attached to a solid support, followed by solid-phase oligonucleotide synthesis and deprotection to yield three products: VitD, the capped linker, and the di-siRNA. The products of the synthesis were then analyzed as described in Example 4.

[0181] Example 2 Alternative synthetic route 1 As shown in Figure 5, the monophosphoamidate linker approach involves the following steps: monoazidotetraethylene glycol is loaded as a tosylated solketal at the branch point. The ketal is then removed to release a cis-diol, selectively protected at the primary hydroxyl with a dimethoxytrityl (DMTr) protecting group, which is reduced to a primary amine with triphenylphosphine, which is immediately protected with a monomethoxytrityl (MMTr) protecting group. The remaining hydroxyl is succinylated with succinic anhydride and coupled to a solid support (LCAA CPG). Oligonucleotide synthesis and deprotection yield one major product, a di-siRNA with both phosphate and phosphoamidate linkages. This example highlights an alternative and straightforward synthetic route for producing only phosphate and phosphoamidate linkers.

[0182] Example 3 Alternative synthetic route 2 A second alternative synthetic approach was developed to prepare diphosphate-containing moieties. As shown in Figure 5, the diphosphoate linker approach involves the following steps: starting with solketal-modified tetraethylene glycol, removing the ketal and selectively protecting the two primary hydroxyls with dimethoxytrityl (DMTr). The remaining hydroxyls are extended with silyl-protected 1-bromoethanol. The TBDMS is removed, succinylated, and attached to a solid support. This is followed by solid-phase oligonucleotide synthesis and deprotection to produce the di-siRNA diphosphate-containing linker.

[0183] Example 4 Quality control of chemical synthesis of di-siRNA and vitamin D-conjugated hsiRNA. HPLC To evaluate the quality of the chemical synthesis of di-siRNA and vitamin D-conjugated hsiRNA, analytical HPLC was used to identify and quantify the synthesized products. Three major products were identified: siRNA sense strand capped with a triethylene glycol (TEG) linker, di-siRNA, and vitamin D-conjugated hsiRNA sense strand (Figure 3). Each product was isolated by HPLC and used in subsequent experiments. The chemical structures of the three major synthesized products are shown in Figure 3. HPLC conditions: 5-80% B (15 min), Buffer A (0.1 M TEAA + 5% ACN), Buffer B (100% ACN).

[0184] Mass spectrum Further quality control was performed by mass spectrometry to confirm the nature of the di-siRNA complex. The product was observed to have a mass of 11683 m / z, corresponding to the two sense strands of siRNA linked at the 3' end by a TEG linker (Figure 4). In this particular example, the siRNA sense strand was designed to target the huntingtin gene (Htt). The chemical synthesis method described in Example 1 successfully produced the desired product, a di-branched siRNA complex targeting the huntingtin gene. LC-MS conditions: 0-100% B (7 min), 0.6 mL / min. Buffer A (25 mM HFIP, 15 mM DBA in 20% MeOH), Buffer B (20% MeOH in Buffer A).

[0185] Example 5 Efficacy and cellular uptake of by-products from the chemical synthesis of di-siRNA and vitamin D-conjugated hsiRNA To evaluate the Htt gene silencing efficacy of each HPLC-isolated byproduct from the chemical synthesis of di-siRNA and vitamin D-conjugated hsiRNA, HeLa cells were treated with each isolated compound via lipid-mediated transfection. Huntingtin mRNA expression was assessed using Affymetrix Quantigene 2.0 and normalized to a housekeeping gene (PPIB). All four byproducts resulted in significant Htt gene silencing 72 hours after transfection (Figure 22). Fluorescently labeled byproducts were delivered to mice via intrastriatal injection, and cellular uptake was tested in vivo by measuring uptake via fluorescent imaging. The di-siRNA product showed dramatically increased uptake in the hemisphere of injected mice compared with the other three byproducts (Figure 22). Of the four byproducts resulting from the chemical synthesis reaction, di-siRNA demonstrated both effective gene silencing and high cellular uptake in vivo.

[0186] Example 6 In vitro efficacy of di-branched siRNA structures To determine the in vitro efficacy of di-branched siRNA (di-siRNA), di-siRNA targeting Htt was transfected into HeLa cells using a lipid-mediated delivery system. HeLa cells were transfected with branched oligonucleotides at various concentrations using RNAiMax. HTT mRNA expression was measured 72 hours after transfection. Di-siRNA resulted in significant silencing of the HTT gene, similar to the effect produced by single double-stranded siRNA in HeLa cells (Figure 10).

[0187] To determine the cellular uptake and gene silencing efficacy in primary cortical neurons without lipid-mediated delivery, cells were passively treated with Htt-di-siRNA at various concentrations for one week. HTT mRNA expression was measured and normalized to the housekeeping gene PPIB. As shown in Figure 10, the di-siRNA structure resulted in significant silencing of the Htt gene, demonstrating that the di-branched siRNA structure can be efficiently delivered to neurons without lipid formation. This indicates that the di-branched structure of the siRNA complex does not interfere with RISC transduction and gene silencing of known effective siRNAs.

[0188] Example 7 Methods for administration of di-siRNA and vitamin D-conjugated hsiRNA To evaluate the efficacy of branched oligonucleotide delivery and activity in neurons in vivo, di-HTT-Cy3 was delivered to mice via intrastriatal (IS) injection. Di-HTT-Cy3 localized and accumulated throughout the injected hemisphere of the brain, whereas a single branched HTT-siRNA (triethylene glycol-conjugated siRNA (TEG-siRNA)) showed very low accumulation throughout the injected hemisphere (Figure 11). A single IS injection of Htt-di-siRNA resulted in significant gene silencing one week after injection (Figure 13), and the level of gene silencing was maintained for two weeks after injection (Figure 26). Further experiments showed that a single IS injection of di-HTT-Cy3 did not result in significant toxicity two weeks after injection (Figure 27A). Htt-di-siRNA resulted in significant gliosis (Figure 27B), which is expected when the Htt gene is silenced in neurons. Furthermore, di-HTT-Cy3 did not accumulate in the liver or kidney 2 weeks after IS injection (Figure 13), and Htt mRNA was not significantly silenced in the liver or kidney after IS injection (Figure 23). Compared with TEG-siRNA alone, the double-branched structure of di-siRNA improved distribution and neuronal uptake; therefore, the size and / or structure of the siRNA complex is important for efficacy. IS injection of Htt-di-siRNA resulted in a significant and stable reduction of Htt that remained localized in the brain, a level of efficacy never before demonstrated with unconjugated siRNA.

[0189] Example 8 Other Administration Routes1 To evaluate the efficacy of delivery and activity of the branched oligonucleotides in the spinal cord, di-HTT-Cy3 was delivered to mice via intrathecal (IT) injection into the lumbar region of the spinal cord. As shown in Figures 14 and 28-29, one week after injection, di-HTT-Cy3 accumulates in the spinal cord. IT injection also resulted in significant Htt mRNA silencing in the cervical, thoracic, and lumbar regions of the spinal cord (Figure 14). IT injection of di-HTT-Cy3 successfully resulted in significant gene silencing in the spinal cord.

[0190] Example 9 Other administration routes2 To evaluate the efficacy of branched oligonucleotide delivery and activity throughout the brain in clinically relevant experiments, di-HTT-Cy3 was delivered to mice via intracerebroventricular (ICV) injection. Both 2 days and 2 weeks after injection, di-siRNA accumulated throughout the brain (Figures 30-31). ICV administration of di-HTT-Cy3 also significantly silenced Htt mRNA and protein expression 2 weeks after injection (Figure 32). Further experiments demonstrated that ICV delivery did not result in significant toxicity 2 weeks after injection (Figure 33). However, ICV injection of di-HTT-Cy3 resulted in gliosis in multiple regions of the brain, presumably due to silencing of the Htt gene (Figure 34). ICV injection avoids the blood-brain barrier by administering di-siRNA directly into the cerebrospinal fluid (CSF), and this injection is used to treat brain diseases.

[0191] This result is significant for the therapeutic potential of branched oligonucleotides, as ICV injection is a therapeutically relevant injection for neurological disorders. The efficacy and stability of branched oligonucleotides after ICV administration indicates that the invention described herein can be used as a therapy for a variety of difficult-to-treat neurological disorders, including Huntington's disease.

[0192] Example 10 Other routes of administration3 To evaluate the efficacy of di-siRNA delivery and activity throughout the body, di-HTT-Cy3 was administered to mice via intravenous (IV) injection. Mice were injected with 20 mg / kg di-HTT-Cy3 on two consecutive days (40 mg / kg total) and sacrificed 24 hours after the final injection. As shown in Figures 35-36, after IV delivery, di-siRNA accumulated in many organs (including liver, kidney, spleen, pancreas, lung, fat, muscle, thymus, colon, and skin). Di-siRNA also accumulated in the brain, demonstrating the ability of di-siRNA to cross the blood-brain barrier, an unpredictable result when using therapeutic siRNA. IV injection demonstrates that the di-siRNA construct is effective and functional in a wide variety of cell types throughout the body.

[0193] Example 11 Determination of toxicity and gliosis toxicity Because elevated DARPP32 indicates neuronal death (Jin, H., et al. DARPP-32 to quantify intracerebral hemorrhage-induced neuronal death in basal ganglia. Transl Stroke Res. 4(1): 130-134. 2013), we assessed DARPP32 protein levels in brain tissue to assess the toxicity level in the brain after di-HTT-Cy3 injection. Mice were treated with 2 nmol of di-HTT-Cy3 (4 nmol of the corresponding antisense HTT chain) via intravenous or intravenous injection. Animals were sacrificed 14 days after injection, and tissue punches were taken from 300 μm brain sections derived from various brain regions. DARPP32 protein was quantified by immunoblotting. Artificial cerebrospinal fluid (aCSF) was used as a negative control. Neither high-dose di-HTT-Cy3IS injection nor ICV injection resulted in significant toxicity (Figures 27 and 33).

[0194] gliosis To assess the level of gliosis in the brain after di-HTT-Cy3 injection, GFAP protein levels were assessed after high-dose di-HTT-Cy3. Mice were treated with 2 nmol of di-HTT-Cy3 (4 nmol of the corresponding antisense HTT chain) via IS or ICV injection. 14 days after injection, mice were sacrificed, and tissue punches were taken from 300 μm brain sections derived from various brain regions. GFAP protein was quantified by immunoblotting. Artificial cerebrospinal fluid (aCSF) was used as a negative control. Both IS and ICV injection of high-dose di-HTT-Cy3 resulted in significant gliomas (Figures 27 and 34). However, glioma induction is likely due to near-complete silencing of the huntingtin gene.

[0195] Example 12 Determination of di-HTT-Cy3 in vivo efficacy Distribution and Accumulation To determine the efficacy of branched oligonucleotide distribution in vivo, mice were treated with di-HTT-Cy3 by intravenous, intrathecal, intravenous, or intravenous injection as described in Examples 7-10. In all examples, 2 nmol of di-HTT-Cy3 (4 nmol of the corresponding antisense HTT strand) was injected, and accumulation was quantified using Cy3-labeled peptide nucleic acid (PNA) to hybridize with the sense strand. HPLC analysis was then used to quantify ng of di-HTT-Cy3 per mg of tissue. Artificial cerebrospinal fluid (aCSF) was used as a negative control.

[0196] In fluorescence imaging experiments, brain sections were stained with DAPI (blue) imaged using the Cy3 channel to detect the accumulation of di-HTT-Cy3 (red).

[0197] Silencing To determine the in vivo silencing efficacy of branched oligonucleotides, mice were treated with di-HTT-Cy3 via intravenous, intrathecal, or intravenous injection as described above in Examples 7-10. In all examples, mice were injected with 2 nmol of di-HTT-Cy3 (4 nmol of the corresponding antisense HTT strand) and silencing of Htt mRNA was quantified using Affymetrix Quantigene 2.0 as described in Coles, A. et al., A High-Throughput Method for Direct Detection of Therapeutic Oligonucleotide-Induced Gene Silencing In Vivo. Nucl Acid Ther. 26 (2), 86-92, 2015. Data were normalized to the housekeeping control, HPRT, and artificial cerebrospinal fluid (aCSF) was used as a negative control.

[0198] Example 13 Incorporation of hydrophobic moieties in branched oligonucleotide structures: Strategy 1 In some embodiments, short-chain hydrophobic alkylenes or alkanes (Hy) bearing unprotected hydroxyl groups (or amines) that have been phosphitylated with 2-cyanoethoxy-bis(N,N-diisopropylamino)phosphine (or other suitable phosphitylation reagents) are used to prepare the corresponding lipophilic phosphoramidites. These lipophilic phosphoramidites can be added to the terminal positions of branched oligonucleotides using conventional oligonucleotide synthesis conditions. This strategy is shown in Figure 44.

[0199] Example 14 Incorporation of hydrophobic moieties in branched oligonucleotide structures: Strategy 2 In another example, short / small aromatic planar molecules (Hy) bearing unprotected hydroxyl groups (or amines) with or without a positive charge that can be phosphitylated using 2-cyanoethoxy-bis(N,N-diisopropylamino)phosphine (or any other suitable phosphitylation reagent) are used to prepare the corresponding aromatic hydrophobic phosphoramidites. The aromatic moiety carries a positive charge. These lipophilic phosphoramidites can be added to the terminal positions of branched oligonucleotides using conventional oligonucleotide synthesis conditions. This strategy is shown in Figure 45.

[0200] Example 15 Incorporating hydrophobic moieties into branched oligonucleotide structures: Strategy 3 To introduce biologically relevant hydrophobic moieties, short lipophilic peptides were prepared by sequential peptide synthesis on a solid support or in solution (the latter described herein). Any positive charge reduces the overall net charge of the oligonucleotide; therefore, to increase hydrophobicity, a short (1–10) amino acid chain can also contain positively charged or polar amino acid moieties. Once a peptide of appropriate length is prepared, it should be capped with acetic anhydride or another short-chain fatty acid to increase hydrophobicity and mask the free amine. The carbonyl protecting group is then removed to yield the coupled 3-aminopropan-1-ol, which allows phosphitylation of the free hydroxyl (or amine). This amino acid phosphoramidite can then be added to the terminal 5' position of the branched oligonucleotide using conventional oligonucleotide synthesis conditions. This strategy is shown in Figure 46.

Claims

1. 1. A composition for use in mediating RNA silencing in neurons in the brain of a subject, comprising a compound of formula (I): 【Chemical 1】 [During the ceremony, L has an ethylene glycol or polyethylene glycol chain consisting of one or more ethylene glycols, and L is attached to N by a phosphate, phosphonate, or phosphoramidate group; N is an siRNA comprising a duplex of sense and antisense strands and unpaired overhangs of 2 to 6 nucleotides linked by phosphorothioate linkages, wherein both the sense and antisense strands comprise alternating 2'-O-methyl-modified nucleotides and 2'-deoxy-2'-fluoro-modified nucleotides; the sense strand and the antisense strand each have a 5' end and a 3' end; n is 2; L connects the 3' ends of each of the two sense strands, the 5' ends of each of the two sense strands, the 3' ends of each of the two antisense strands, or the 5' ends of each of the two antisense strands; the sense strand and the antisense strand each consist of chemically modified nucleotides, where the chemical modification is a 2'-O-methyl modification or a 2'-deoxy-2'-fluoro modification; and The nucleotides at positions 1 and 2 from the 5' end of the sense strand and the antisense strand are linked to adjacent nucleotides by phosphorothioate bonds; wherein the antisense strand is complementary to a target mRNA in a neuron. A composition comprising a compound of formula (I).

2. 2. A composition for use according to claim 1, wherein the compound of formula (I) is of formula (II): 【Chemistry 2】 [During the ceremony, The top strand is the antisense strand and the bottom strand is the sense strand; each X is independently selected from a chemically modified adenosine, a chemically modified guanosine, a chemically modified uridine, and a chemically modified cytidine, wherein the chemical modification is a 2'-O-methyl modification or a 2'-deoxy-2'-fluoro modification, and the antisense strand comprises alternating 2'-O-methyl-modified nucleotides and 2'-deoxy-2'-fluoro-modified nucleotides; each Y is independently selected from a chemically modified adenosine, a chemically modified guanosine, a chemically modified uridine, and a chemically modified cytidine, wherein the chemical modification is a 2'-O-methyl modification or a 2'-deoxy-2'-fluoro modification, and the sense strand comprises alternating 2'-O-methyl-modified nucleotides and 2'-deoxy-2'-fluoro-modified nucleotides; - represents a phosphodiester internucleoside linkage; = represents a phosphorothioate internucleoside linkage; and Each --- represents an individual base pair interaction; n and L are defined as in formula (I); and where R is the 5' terminal group of the antisense strand, and 【Chemistry 3】 selected from the group consisting of A composition having the structure:

3. 2. The composition for use according to claim 1, wherein the compound of formula (I) is a compound of formula (III): 【Chemistry 4】 [During the ceremony, The top strand is the antisense strand and the bottom strand is the sense strand; each X is independently a 2'-deoxy-2'-fluoro-modified nucleotide; each X is independently a 2'-O-methyl-modified nucleotide; each Y is independently a 2'-deoxy-2'-fluoro-modified nucleotide; each Y is independently a 2'-O-methyl-modified nucleotide; - represents a phosphodiester internucleoside linkage; = represents a phosphorothioate internucleoside linkage; and Each --- represents an individual base pair interaction; n and L are defined as in formula (I); and R is the 5' terminal group of the antisense strand, and 【Chemistry 5】 selected from the group consisting of A composition having the structure:

4. 4. The composition for use according to claim 3, wherein R in formula (III) is R ３ : 【Chemistry 6】 The composition.

5. 2. A composition for use according to claim 1, wherein the compound of formula (I) is of formula (IV): 【Chemistry 7】 [During the ceremony, The top strand is the antisense strand and the bottom strand is the sense strand; A is a 2'-deoxy-2'-fluoro modified adenosine; A is a 2'-O-methyl-modified adenosine; G is a 2'-deoxy-2'-fluoro-modified guanosine; G is a 2'-O-methyl-modified guanosine; U is 2'-deoxy-2'-fluoro-modified uridine; U is a 2'-O-methyl modified uridine containing a 2'-O-methyl modification of the sugar portion of the nucleotide; C is a 2'-deoxy-2'-fluoro-modified cytidine containing a 2'-deoxy-2'-fluoro modification; C is a 2'-O-methyl-modified cytidine; - represents a phosphodiester internucleoside linkage; = represents a phosphorothioate internucleoside linkage; and Each --- represents an individual base pair interaction; n and L are defined as in formula (I); and R is the 5' terminal group of the antisense strand, and 【Chemistry 8】 selected from the group consisting of A composition having the structure:

6. 6. The composition for use according to claim 5, wherein R is R ３ : 【Chemistry 9】 The composition.

7. 2. A composition for use according to claim 1, wherein the compound of formula (I) is of formula (V): 【Chemistry 10】 [During the ceremony, The top strand is the antisense strand and the bottom strand is the sense strand; each X is independently selected from a chemically modified adenosine, a chemically modified guanosine, a chemically modified uridine, and a chemically modified cytidine, wherein the chemical modification is a 2'-O-methyl-modified or a 2'-deoxy-2'-fluoro-modified, and the antisense strand comprises alternating 2'-O-methyl-modified nucleotides and 2'-deoxy-2'-fluoro-modified nucleotides; each Y is independently selected from a chemically modified adenosine, a chemically modified guanosine, a chemically modified uridine, and a chemically modified cytidine, wherein the chemical modification is a 2'-O-methyl-modified or a 2'-deoxy-2'-fluoro-modified, and the sense strand comprises alternating 2'-O-methyl-modified nucleotides and 2'-deoxy-2'-fluoro-modified nucleotides; - represents a phosphodiester internucleoside linkage; = represents a phosphorothioate internucleoside linkage; and Each --- represents an individual base pair interaction; n and L are defined as in formula (I); and R is the 5' terminal group of the antisense strand, and 【Chemistry 11】 selected from the group consisting of A composition having the structure:

8. 2. A composition for use according to claim 1, wherein the compound of formula (I) is of formula (VI): 【Chemistry 12】 [During the ceremony, The top strand is the antisense strand and the bottom strand is the sense strand; each X is independently a 2'-deoxy-2'-fluoro-modified nucleotide; each X is independently a 2'-O-methyl-modified nucleotide; each Y is independently a 2'-deoxy-2'-fluoro-modified nucleotide; and each Y is independently a 2'-O-methyl-modified nucleotide; - represents a phosphodiester internucleoside linkage; = represents a phosphorothioate internucleoside linkage; and Each --- represents an individual base pair interaction; n and L are defined as in formula (I); and R is the 5' terminal group of the antisense strand, and 【Chemistry 13】 selected from the group consisting of A composition having the structure:

9. 2. The composition for use according to claim 1, wherein the compound of formula (I) is of formula (VII): 【Chemistry 14】 [During the ceremony, The top strand is the antisense strand and the bottom strand is the sense strand; A is a 2'-deoxy-2'-fluoro-modified adenosine; A is a 2'-O-methyl-modified adenosine; G is a 2'-deoxy-2'-fluoro-modified guanosine; G is a 2'-O-methyl-modified guanosine; U is 2'-deoxy-2'-fluoro-modified uridine; U is 2'-O-methyl-modified uridine; C is a 2'-deoxy-2'-fluoro-modified cytidine; C is a 2'-O-methyl-modified cytidine; each Y is independently a 2'-deoxy-2'-fluoro-modified nucleotide; each Y is independently a 2'-O-methyl-modified nucleotide; - represents a phosphodiester internucleoside linkage; = represents a phosphorothioate internucleoside linkage; and Each --- represents an individual base pair interaction; n and L are defined as in formula (I); and R is the 5' terminal group of the antisense strand, and 【Chemistry 15】 selected from the group consisting of A composition having the structure:

10. 10. The composition for use according to claim 9, wherein R is R ３ : 【Chemistry 16】 The composition.

11. A composition for use according to any one of claims 1 to 10, wherein L of the compound of formula (I) is L1: 【Chemistry 17】 ;or L2: 【Chemistry 18】 A composition having the structure:

Images