Branched oligonucleotides

Abstract

To provide branched oligonucleotides exhibiting efficient and specific tissue distribution, cellular uptake, minimum immune response and off-target effects, without formulation.SOLUTION: Provided herein is a branched oligonucleotide compound comprising two or more nucleic acids, the nucleic acids are connected to one another by one or more moieties selected from a linker, a spacer and a branching point.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] STATEMENT REGARDING 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 No. 62 / 289,268, filed January 31, 2016, and No. 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 various 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 RNA to a wide range of tissues. Currently, small therapeutic RNA can only be effectively delivered to the liver. There remains a need for self-delivering siRNA and general therapeutic oligonucleotides that show 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, which are linked together by one or more moieties selected from a linker, a spacer, and a branch point.

[0007] In certain 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, spacer or branch point at the 3'-end or 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 1-6 from the 3' end or positions 1-7 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 included 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. Particular embodiments of the hydrophobic moiety and the synthesis strategy of the hydrophobically modified branched oligonucleotide compound 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, the 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 the sense strand and the antisense strand independently comprising 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 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 a mismatch, respectively. It has the structure:

[0019] In some 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 each Y is independently a nucleotide that contains a 2'-O-methyl modification; It has the structure:

[0020] In some embodiments, the compound of formula (I) has 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 structure:

[0021] In some embodiments, the compound of formula (I) has 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 a mismatch, respectively. It has the structure:

[0022] In some embodiments, the compound of formula (I) has 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 each Y is independently a nucleotide containing a 2'-O-methyl modification; It has the structure:

[0023] In some embodiments, the compound of formula (I) has 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; each Y is independently a nucleotide that contains a 2'-O-methyl modification; It has the 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 where 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 where 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) 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; each cNA is independently a carrier nucleic acid that contains 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 2 nucleotides. In some embodiments, the nucleotides of the overhang 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 for delivery is selected from the group consisting of brain, liver, skin, kidney, spleen, pancreas, colon, fat, lung, muscle and thymus. [Brief description of the drawings]

[0034] [Figure 1] The structure of di-hsiRNA is shown. Black - 2'-O-methyl, grey - 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 that is essential for tissue distribution, cellular uptake and utility. The structure presented utilizes a teg linker of four monomers. The chemical nature of the linker can be altered without affecting efficacy. It can be adjusted by length, chemical composition (all carbon), saturation or the addition of chemical targeting ligands. [Diagram 2]Chemical synthesis, purification and QC of di-branched siRNAs. [Diagram 3] HPLC and QC of the compounds produced by the method shown in Figure 2 are shown. The 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 potency 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. [Diagram 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. The combination of different linkers, spacers and branch points results 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 10A]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 was normalized to housekeeping genes (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 10B] 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 was normalized to housekeeping genes (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 10C] 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 was normalized to housekeeping genes (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, whereas in (B) the left side of the brain appears only faintly red. [Figure 12] 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 mouse brain. (A) Silencing of Htt mRNA 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] Figure 1 shows widespread distribution and efficacy in the spinal cord after bolus intrathecal injection of di-hsiRNA. 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 healthy subjects in all regions of the spinal cord, 7 days, n=6. Animals were sacrificed 7 days after injection. Tissue punches 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 as a percentage of aCSF control to housekeeping genes, HPRT and grafts. aCSF - artificial CSF. (B) Animals were injected IT lumbar vein with 75 μg Cy3-Chol-hsiRNA, Cy-di-hsiRNA. Chol-hsiRNAs show steep diffusion from the lateral to medial spinal cord. Di-hsiRNA shows widespread distribution throughout the 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 15A] Illustrates the 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 linkers or spacers described above. This can be utilized at the 3' and 5' ends of the sense or antisense strand or combinations thereof; (C) Branched oligonucleotides made with three separate strands. The long double sense strands are synthesized using 3' phosphoramidites and 5' phosphoramidites, allowing for 3'-3' adjacent or 5'-5' adjacent ends. [Figure 15B] Illustrates the 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 linkers or spacers described above. This can be utilized at the 3' and 5' ends of the sense or antisense strand or combinations thereof; (C) Branched oligonucleotides made with three separate strands. The long double sense strands are synthesized using 3' phosphoramidites and 5' phosphoramidites, allowing for 3'-3' adjacent or 5'-5' adjacent ends. [Figure 15C]Illustrates the 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 linkers or spacers described above. This can be utilized at the 3' and 5' ends of the sense or antisense strand or combinations thereof; (C) Branched oligonucleotides made with three separate strands. The long double sense strands are synthesized using 3' phosphoramidites and 5' phosphoramidites, allowing for 3'-3' adjacent or 5'-5' adjacent ends. [Figure 16] 1 shows branched oligonucleotides of the invention having conjugated biologically active moieties. [Figure 17] 1 shows the relationship between phosphorothioate content and stereoselectivity. [Figure 18] Examples of hydrophobic moieties are shown below. [Figure 19] Illustrates internucleotide linkages. [Figure 20] Examples of internucleotide backbone linkages are shown. [Figure 21] Examples of glycosylation products are shown below. [Figure 22A] Di-FM-hsiRNA is shown. (A) Chemical composition of the four byproducts generated from VitD-FM-hsiRNA synthesis and crude reverse-phase HPLC analysis of the original chemical synthesis. (B) Efficacy of the byproducts in HeLa cells after lipid mediated hsiRNA delivery. Cells were treated for 72 h. mRNA was measured using the QuantiGene 2.0 kit (Affimetrix). Data were normalized to the housekeeping gene HPRT and expressed as a percentage of untreated controls. (C) A single, unilateral intrastriatal injection (25 μg) of each hsiRNA byproduct. Images were taken 48 h after injection. [Figure 22B]Di-FM-hsiRNA is shown. (A) Chemical composition of the four byproducts generated from VitD-FM-hsiRNA synthesis and crude reverse-phase HPLC analysis of the original chemical synthesis. (B) Efficacy of the byproducts in HeLa cells after lipid mediated hsiRNA delivery. Cells were treated for 72 h. mRNA was measured using the QuantiGene 2.0 kit (Affimetrix). Data were normalized to the housekeeping gene HPRT and expressed as a percentage of untreated controls. (C) A single, unilateral intrastriatal injection (25 μg) of each hsiRNA byproduct. Images were taken 48 h after injection. [Figure 22C] Di-FM-hsiRNA is shown. (A) Chemical composition of the four byproducts generated from VitD-FM-hsiRNA synthesis and crude reverse-phase HPLC analysis of the original chemical synthesis. (B) Efficacy of the byproducts in HeLa cells after lipid mediated hsiRNA delivery. Cells were treated for 72 h. mRNA was measured using the QuantiGene 2.0 kit (Affimetrix). Data were normalized to the housekeeping gene HPRT and expressed as a percentage of untreated controls. (C) A single, unilateral intrastriatal injection (25 μg) of each hsiRNA byproduct. Images were taken 48 h after injection. [Figure 23A] Figure 23A shows that di-HTT-Cy3 does not efficiently induce silencing in liver or kidney after intrastriatal injection. Figure 23A shows the scatter plot of Htt mRNA expression in liver 1 week after intrastriatal injection of di-HTT-Cy3 compared to negative control (aCSF). Figure 23B shows the scatter plot of Htt mRNA expression in kidney 1 week after intrastriatal injection of di-HTT-Cy3 compared to negative control (aCSF). [Figure 23B] Figure 23A shows that di-HTT-Cy3 does not efficiently induce silencing in liver or kidney after intrastriatal injection. Figure 23A shows the scatter plot of Htt mRNA expression in liver 1 week after intrastriatal injection of di-HTT-Cy3 compared to negative control (aCSF). Figure 23B shows the scatter plot of Htt mRNA expression in kidney 1 week after intrastriatal injection of di-HTT-Cy3 compared to negative control (aCSF). [Figure 24A] The results show that di-HTT efficiently 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 a scatter plot 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 a scatter plot 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 24B] The results show that di-HTT efficiently 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 a scatter plot 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 a scatter plot 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). [Diagram 25] Scatter dot plots measuring the amount of di-HTT-Cy3 in the striatum and cerebral cortex are shown, demonstrating that high amounts of di-HTT-Cy3 are still detectable two weeks after intrastriatal injection.

[0037] [Figure 26A] Figure 26 shows that di-HTT-Cy3 effectively silences HTT mRNA and protein expression in both striatum and cerebral cortex after 2 weeks of intrastriatal injection. Figure 26A shows the scatter plots measuring Htt mRNA levels in striatum and cerebral cortex after 2 weeks of intrastriatal injection. Figure 26B shows the scatter plots measuring Htt protein levels in striatum and cerebral cortex after 2 weeks of intrastriatal injection. [Figure 26B]Figure 26 shows that di-HTT-Cy3 effectively silences HTT mRNA and protein expression in both the striatum and cerebral cortex 2 weeks after intrastriatal injection. Figure 26A shows a scatter dot plot measuring Htt mRNA levels in the striatum and cerebral cortex 2 weeks after intrastriatal injection. Figure 26B shows a scatter dot plot measuring Htt protein levels in the striatum and cerebral cortex 2 weeks after intrastriatal injection. [Figure 27A] Two weeks after intrastriatal injection, high-dose di-HTT-Cy3 treatment does not cause high toxicity in vivo, but does cause significant gliosis in vivo. Figure 27A shows a scatter plot measuring DARPP32 signal in the striatum and cerebral cortex two weeks after injection of di-HTT-Cy3 or aCSF. Figure 27B shows a scatter plot measuring GFAP protein amount in the striatum and cerebral cortex two weeks after injection of di-HTT-Cy3 or aCSF. [Figure 27B] Two weeks after intrastriatal injection, high-dose di-HTT-Cy3 treatment does not cause high toxicity in vivo, but does cause significant gliosis in vivo. Figure 27A shows a scatter plot measuring DARPP32 signal in the striatum and cerebral cortex two weeks after injection of di-HTT-Cy3 or aCSF. Figure 27B shows a scatter plot measuring GFAP protein amount 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 FIG. 28B is shown (magnification of the spinal cord). Blue - nuclei, red - di-HTT-Cy3. XXX [Figure 30-1] 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 brightfield 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 30-2] 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 brightfield 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. [Diagram 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 32A] Figure 32A shows that di-HTT-Cy3 induces Htt gene silencing in multiple brain regions 2 weeks after intracerebroventricular injection, compared to 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 2 weeks after intracerebroventricular injection, compared to negative control injection (aCSF). Scatter dot plots measure Htt protein levels in multiple brain regions. [Figure 32B] Figure 32A shows that di-HTT-Cy3 induces Htt gene silencing in multiple brain regions 2 weeks after intracerebroventricular injection, compared to 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 2 weeks after intracerebroventricular injection, compared to negative control injection (aCSF). Scatter dot plots measure Htt protein levels in multiple brain regions. [Diagram 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 following di-HTT-Cy3 injection of aCSF. [Diagram 34] Scatter dot plots measure DARPP32 signal in multiple regions of the brain following di-HTT-Cy3 injection of aCSF, showing that high dose intraventricular injection results in significant gliosis in vivo.

[0038] [Diagram 35] Di-HTT-Cy3 is shown to be distributed in multiple organs after intravenous injection. Fluorescent images show the amount of di-HTT-Cy3 in the heart, kidney, adrenal gland and spleen after injection of di-HTT-Cy3 or negative control (PBS). [Diagram 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 38] This shows that complete metabolic stabilization of hsiRNAs results in more effective gene silencing after intrastriatal injection of hsiRNAHTT or FM-hsiRNAHTT. 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. [Diagram 40] Di-HTT with a TEG phosphoramidate linker is shown. [Diagram 41] Di-HTT with a TEG diphosphate linker is shown. [Diagram 42] Variants of di-HTT having either two oligonucleotide branches or four oligonucleotide branches are shown. [Diagram 43] 1 shows another variation of di-HTT in a structure having two oligonucleotide branches and R2 attached to a linker. [Diagram 44] 1 shows a first strategy for the introduction of hydrophobic moieties into branched oligonucleotide structures. [Diagram 45] A second strategy for the introduction of hydrophobic moieties into branched oligonucleotide structures is presented. [Diagram 46] A third strategy for the introduction of hydrophobic moieties into branched oligonucleotide structures is presented. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[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 shown 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, a branched oligonucleotide is provided that is a compound of the present invention. In some embodiments, the compound of the present invention has 2-8 oligonucleotides linked by a linker. The linker can be hydrophobic. In certain embodiments, the compound of the present invention has 2-3 oligonucleotides. In some embodiments, the oligonucleotides independently have sufficient chemical stability (e.g., at least 40% of the constituent bases are chemically modified). In certain embodiments, the oligonucleotide has complete chemical stability (i.e., all the constituent bases are chemically modified). In some embodiments, the compound of the present invention comprises one or more single-stranded phosphorothioated tails, each independently having 2-20 nucleotides. In certain embodiments, each single-stranded tail has 8-10 nucleotides.

[0043] In certain embodiments, the compounds of the present invention are characterized by three characteristics: (1) branched structure, (2) metabolic stabilization, and (3) the presence of a single-stranded tail that comprises 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 to 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 functionality is important for the function of di-branched oligonucleotides.

[0045] In certain embodiments, the compounds of the invention are characterized by the following properties: (1) two or more branched oligonucleotides, e.g., having unequal numbers 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 3, optimally 5-20 phosphorothioate linkages.

[0046] The compounds of the present invention are provided in various 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. A combination 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 carry these active single-stranded oligonucleotides and promote distribution and cellular internalization. The short double-stranded region provides a low melting temperature (T) for rapid dissociation of the branched structures after internalization into cells. m Approximately 37°C.

[0047] As shown in Figure 16, the di-siRNA compound of the present invention can include 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 by linkers or spacers, or can be added by additional linkers or spacers that are linked to other free siRNA ends.

[0048] The presence of branched structure improves the level of tissue retention in the brain by 100-fold compared with unbranched compounds of individual chemical compositions, suggesting a new mechanism of cell retention and distribution.Compounds of the present invention unexpectedly uniformly distribute in spinal cord and brain.Furthermore, compounds of the present invention unexpectedly show 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 include 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 oligonucleotide comprises 2, 3, 4, 6 or 8 nucleic acids. In some embodiments, the branched oligonucleotide comprises 2 nucleic acids. In another embodiment, the branched oligonucleotide comprises 3 nucleic acids. In another embodiment, the branched oligonucleotide comprises 4 nucleic acids. In another embodiment, the branched oligonucleotide comprises 6 nucleic acids. In another embodiment, the branched oligonucleotide comprises 8 nucleic acids. In another embodiment, the branched oligonucleotide comprises 5 nucleic acids. In another embodiment, the branched oligonucleotide comprises 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 3'-end. In an 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 an 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 some 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 an embodiment, 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 an embodiment, each nucleic acid is linked to a linker, spacer, or branch point at the 3'-end of the sense strand. In an embodiment, each nucleic acid is linked to a linker, spacer, or branch point at the 3'-end of the antisense strand. In another embodiment, each nucleic acid is linked to a linker, spacer, or branch point at the 5'-end of the sense strand. In another embodiment, each nucleic acid is linked to a linker, spacer, or branch point at the 5'-end of the antisense strand. In an embodiment, each nucleic acid is linked to a linker. In another embodiment, each nucleic acid is linked to a spacer. In another embodiment, 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 some 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 is composed of chemically modified nucleotides. In certain embodiments, >95%, >90%, >85%, >80%, >75%, >70%, >65%, >60%, >55% or >50% of each nucleic acid comprises chemically modified nucleotides.

[0057] 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 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 1-6 from the 3' end or positions 1-7 from the 3' end are linked to adjacent nucleotides by phosphorothioate bonds. In other embodiments, at least 5 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, an RNA, a DNA, a phosphate, a phosphonate, a phosphoramidate, an ester, an amide, a triazole, and combinations thereof, where any carbon or oxygen of the linker may be optionally replaced with a nitrogen atom, has a hydroxyl functional group, or has 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 an RNA. In another embodiment, each linker is a 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 of FIG. 7.

[0059] Compounds of formula (I) In a second embodiment, the 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 (1) 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, an RNA, a DNA, a phosphate, a phosphonate, a phosphoramidate, an ester, an amide, a triazole, and combinations thereof, where any carbon or oxygen of the linker may be optionally replaced by a nitrogen atom, has a hydroxyl substituent, or has an oxo substituent. In an embodiment of the compound of formula (I), each linker is an ethylene glycol chain. In another embodiment, each linker is an alkyl chain. In another embodiment of the compound of formula (I), each linker is a peptide. In another embodiment of the compound of formula (I), each linker is an RNA. In another embodiment of the compound of formula (I), each linker is a DNA. In another embodiment of the compound of formula (I), each linker is a phosphate. In another embodiment, each linker is a phosphonate. In another embodiment of the compound of formula (I), each linker is a phosphoramidate. In another embodiment 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 in 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 three or more valencies (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 is composed 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 one embodiment, R is R1. In another embodiment, R is R2. In another embodiment, R is R3. In another embodiment, R is R4. In another embodiment, R is R5. In another embodiment, R is R6. In another embodiment, R is R7. In another embodiment, R is R8.

[0068] Structure of formula (II) In some embodiments, the compound of formula (I) has 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 a mismatch, respectively. It has the structure:

[0069] In certain embodiments, the structure of formula (II) does not contain a 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 some 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 each Y is independently a nucleotide that contains a 2'-O-methyl modification; It has the 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 some embodiments, the structure of formula (III) contains one mismatch. In another embodiment, the structure of formula (III) contains two mismatches. In another embodiment, the compound of formula (III) contains three mismatches. In another embodiment, the compound of formula (III) contains four mismatches.

[0074] Structure of formula (IV) In some embodiments, the compound of formula (I) has 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 structure:

[0075] Structure of formula (V) In some embodiments, the compound of formula (I) has 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 a mismatch, respectively. It has the structure:

[0076] In certain embodiments, the structure of formula (V) does not contain a mismatch. 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 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 each Y is independently a nucleotide that contains a 2'-O-methyl modification; It has the 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 a mismatch. In some 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 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 each Y is independently a nucleotide that contains a 2'-O-methyl modification; It has the 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 where n is 2.

[0083] In an embodiment of the structure of formula (II), L has the structure of L1. In an embodiment of the structure of formula (III), L has the structure of L1. In an embodiment of the structure of formula (IV), L has the structure of L1. In an embodiment of the structure of formula (V), L has the structure of L1. In an embodiment of the structure of formula (VI), L has the structure of L1. In an embodiment of the structure of formula (VII), L has the structure of 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 where 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) 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; each cNA is independently a carrier nucleic acid that contains 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 one embodiment, the delivery system further comprises n therapeutic nucleic acids (NAs), where each NA is hybridized to at least one cNA. In one embodiment, the delivery system comprises 2 NAs. In another embodiment, the delivery system comprises 3 NAs. In another embodiment, the delivery system comprises 4 NAs. In another embodiment, the delivery system comprises 5 NAs. In another embodiment, the delivery system comprises 6 NAs. In another embodiment, the delivery system comprises 7 NAs. In another embodiment, the delivery system comprises 8 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 another embodiment, each NA independently comprises 17 consecutive nucleotides. In another embodiment, each NA independently comprises 18 consecutive nucleotides. In another embodiment, each NA independently comprises 19 consecutive nucleotides. In another embodiment, each NA independently comprises 20 consecutive nucleotides.

[0096] In some embodiments, each NA comprises an unpaired overhang of at least 2 nucleotides. In another embodiment, each NA comprises an unpaired overhang of at least 3 nucleotides. In another embodiment, each NA comprises an unpaired overhang of at least 4 nucleotides. In another embodiment, each NA comprises an unpaired overhang of at least 5 nucleotides. In another embodiment, 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 gapmer. In another embodiment, each NA is independently mixmer. In another embodiment, each NA is independently 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 (NA) 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 (NA). In another embodiment, 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 (NA). 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 (NA). 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 5 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 6 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 7 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 8 therapeutic nucleic acids (NAs).

[0099] In one embodiment, the delivery system comprises a compound wherein R is R 3In 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 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 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. 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 in the present invention 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] Moreover, unless otherwise indicated, the experiments described herein use conventional molecular and cell biological and immunological techniques that are within the scope of the art.Such techniques are known to those skilled in the art and are described in detail in the literature.See, for example, 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 case 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 "including" as well as 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 chains 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 has complementarity with and hybridizes to 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 at the 3' carbon of the ribose ring.

[0108] The term "5' end" as used herein 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 that consists of a heterocyclic base and its sugar.

[0110] As used herein, the term "nucleotide" refers to a nucleoside that has 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 various 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 a double-stranded region.

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

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

[0114] The term "chemically modified nucleotide" or "nucleotide analog" or "altered nucleotide" or "modified nucleotide" as used herein refers to a non-standard nucleotide, including non-naturally occurring ribonucleotides or deoxyribonucleotides. Exemplary nucleotide analogs are modified at any position to change the specific chemical properties of the nucleotide but maintain 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; 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 (eg, alkylated, such as N6-methyladenosine or others known in the art) nucleotides; and other heterocyclic modified nucleotide analogs.

[0115] Nucleotide analogs may also include modifications of 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, NH2, NHR, NR2, COOR, or OR, where R is a substituted or unsubstituted C1-C6 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 RNA molecules that contain 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 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] The term "amide," as used herein, refers to an alkyl or aromatic group bound to an aminocarbonyl functional group.

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

[0122] The term "triazole" as used herein refers to a heterocyclic compound having the formula (C2H3N3) that has a five-membered ring of two carbons and three nitrogens that can be varied 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 a 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 that contain 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 invention that contains a branched oligonucleotide structure and includes an siRNA molecule as a therapeutic nucleic acid.

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

[0132] In another embodiment, the "amino acid" has the formula: [ka] It may refer to 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., phosphodiesters and phosphorothioates) 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 the 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 enable the nucleotide to perform its intended function, for example, as described 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 that contains the analog.

[0136] Delivery and Distribution In another embodiment, a method is provided for selectively delivering a nucleic acid as described herein to an organ of a patient, comprising administering to the patient a branched oligonucleotide as 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 tissue retention of modified oligonucleotides. Other conjugates, such as cholesterol, are shown to be retained but rapidly diffuse away from the injection site. The slight hydrophobicity of the two single-stranded phosphorothioate tails aids in 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 ipsilateral and contralateral to the injection site, indicating that diffusion is not limited to the injected hemisphere but occurs across the midline to the uninjected side. Other injection methods, including intracerebroventricular, can also promote bilateral distribution with just a single injection.

[0140] Di-siRNAs show a very unique cellular distribution when injected intrastriatal 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 does not apply to other siRNA conjugates that do not show cell type preference, such as cholesterol.

[0141] Di-siRNA shows localization to fiber tracts in the striatum, but is present in neuronal cell bodies in the cerebral cortex.Transfer to the cerebral cortex may be by diffusion or retrograde transport along the striatal fiber tract.The theory that retrograde transport is partially important is supported by the fact that some areas of the cerebral cortex show sufficient neuronal penetration, while neurons in adjacent areas do not show any 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, a large amount of di-siRNA was 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 di-siRNA structure can efficiently cross the blood-brain barrier.

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

[0145] A single injection of di-siRNA induces strong silencing in both the striatum and cerebral cortex of mouse brain. This level of efficacy has not been shown before for non-conjugated siRNA. Although di-siRNA appears to be visually taken up into fiber tracts in the striatum, the observed efficacy clearly indicates that striatal neurons internalize di-siRNA in large quantities. In the 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 taken from 300 μm brain sections from the striatum and cerebral cortex. The di-siRNA antisense strands present in various brain regions, liver and kidney were quantified using Cy3-labeled complementary PNA to hybridize with the strands and HPLC to quantify the amount of ng of oligo 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 delivery was not hindered by linking two siRNA duplexes. Although di-siRNA was not effective in HeLa cells without transfection, in primary cortical neurons, one phosphorothioate tail was sufficient to induce at least 60% silencing, indicating that phosphorothioation is an effective method for delivering siRNA to primary neurons without formulation.

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

[0148] As shown in Figure 14, di-siRNA shows 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 silenced mRNA to a similar extent in the cervical, thoracic and lumbar regions, showing 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 mouse brain. The mRNA expression levels show 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 did not result in significant silencing in the liver or kidney, indicating 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 is modified with any of the following modifications.Modifications can be useful in part to further enhance target discrimination, enhance the stability of the agent (e.g., prevent degradation), promote cellular uptake, enhance targeting effectiveness, improve the effectiveness of binding (e.g., to target), improve the patient's 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 to non-target mRNA (e.g. wild-type mRNA) without significantly affecting the specificity of the RNA silencing agent to 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 the introduction of at least one universal nucleotide in its antisense strand. A universal nucleotide comprises 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 a relatively small 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, norpholino-inosine, LNA-inosine, phosphoramidate-inosine, 2'-O-methoxyethyl-inosine and 2'-OMe-inosine. In a particular 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 the disease-associated 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 an exemplary embodiment, 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 an 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 may be modified to facilitate enhanced efficacy and specificity in mediated RNAi by 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 (e.g., the siRNA produced from the siRNA or shRNA designed using the method of the present invention) into RISC in favor of the sense strand, thus increasing and improving the efficacy of target cleavage and silencing. Preferably, the asymmetry of the RNA silencing agent is enhanced by reducing the base pair strength of the antisense strand 5' end (AS5') and the sense strand 3' end (S3') of the RNA silencing agent, when compared with the binding strength or base pair strength of the antisense strand 3' end (AS3') and the sense strand 5' end (S5') of the RNA silencing agent.

[0158] In one embodiment, the asymmetry of the RNA silencing agent of the present invention can be enhanced such 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 there are G:C base pairs 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 enhanced such 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 enhanced such that there is at least one wobble base pair, such as 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 enhanced such that there is at least one base pair that includes a rare nucleotide, such as 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 such that there is at least one base pair that comprises 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 be composed of purine nucleotides, particularly adenosine or guanosine nucleotides.Alternatively, the substitution of pyrimidine nucleotides with modified analogs, for example, the 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 an internal nucleotide with a modified nucleotide 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 one embodiment, the sense strand and / or antisense strand are modified by replacing at least one internal nucleotide.In another embodiment, 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 strand and / or the antisense strand are modified by substitution of at least 5%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95% or more of the internal nucleotides. In yet another embodiment, the sense strand and / or the antisense strand are modified by substitution of all internal nucleotides.

[0161] In a preferred embodiment of the present invention, the RNA silencing agent comprises at least one modified nucleotide analogue.Nucleotide analogue can be located at the position where target specific silencing activity, for example RNAi-mediated activity or translation suppression activity, is not substantially affected, for example, at the 5'-end and / or 3'-end region of siRNA molecule.In particular, 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 that is linked to adjacent ribonucleotides is replaced by a modified group, for example, a phosphothioate group. In exemplary sugar-modified ribonucleotides, the 2'OH group is replaced by a group selected from H, OR, R, halo, SH, SR, NH2, NHR, NR2, or ON (wherein R is C1-C6 alkyl, alkenyl, or alkynyl, and halo is F, Cl, Br, or I).

[0163] In a specific embodiment, 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 a specific embodiment, 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 within 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, such that the binding moiety is a 2'-O-methyl oligonucleotide.

[0164] In an exemplary embodiment, the RNA silencing agent of the present invention comprises locked nucleic acid (LNA). LNA is resistant to nuclease activity (highly stable) and comprises 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 comprise 2'-O,4'-C-ethylene-bridged nucleic acids with possible modifications such as 2'-deoxy-2''-fluorouridine. In addition, LNA increases the specificity of oligonucleotides by constraining the sugar moiety to a 3'-endo configuration, thereby pre-organizing 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 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 more resistant to nuclease digestion and provides improved binding specificity to molecules (Nielsen, et al., Science, (2001), 254: 1497-1500).

[0166] Nucleic acid modified ribonucleotides, i.e. nucleotides that contain at least one non-natural nucleobase instead of naturally occurring nucleobases, are also preferred.The base can be modified to block the activity of adenosine deaminase.Exemplary modified nucleobases 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; deaza nucleotides, such as 7-deazaadenosine; O- and N-alkylated nucleotides, 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 half-life in the body.Thus, the present invention includes RNA silencing agents with 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 the corresponding siRNAs, or enhance the cell targeting activity of the resulting siRNA derivatives, and is useful for tracking siRNA derivatives in cells compared to the 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 a U in the sense or antisense strand, particularly the sense strand, or providing a 3' overhang, such as a 2'OMe portion at the 3' end (where 3' end means the 3' atom or most 3' portion of the molecule, e.g., most 3'P or 2' position as indicated by the context); (b) modifications of the backbone, such as modifications by substituting S for O in a phosphate backbone, e.g., providing U or A or both, phosphorothioate modifications, particularly in the antisense strand, e.g., substituting S for P; (c) substitution of U with a C5 amino linker; (d) substitution of G for A (wherein the sequence change is preferably located in the sense strand and not 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 and not the antisense strand, or those in which the antisense strand has few such modifications. Further exemplary modifications include 3' overhangs, e.g., the use of a methylated P at the 3' end; combinations 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; modifications having a 3' alkyl; modifications having a 3' overhang, e.g., a non-basic pyrrolidone at the 3' end; modifications 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, for example 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 compound of the present invention is conjugated to a lipophilic moiety. In some embodiments, the lipophilic moiety is a ligand that comprises 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 dye (e.g., Cy3).

[0171] 5) Bound Ligand Other substances can be attached to the compounds of the present invention. For example, ligands attached to RNA silencing agents to improve stability, thermodynamics of hybridization with target nucleic acid, targeting to specific tissues or cell types, or cell permeability by endocytosis-dependent or -independent mechanisms. Ligands and related modifications can also increase sequence specificity and, as a result, reduce off-site targeting. Attached ligands can include modified bases or sugars that can function as one or more intercalators. These are preferably present in an intercalation region, such as a bulge, 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 forms. Universal bases as described herein can be included in the ligand. In some embodiments, the ligand can include 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, and can promote selective cleavage of target RNA at the bulge portion by 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., by amino acid derivatives) 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 acridine analogs can increase sequence specificity. For example, neomycin B has a high affinity for RNA compared to DNA, but low 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 may enhance the cell permeability of the RNA silencing agent. The attached ligand may be a polyarginine peptide, peptoid or peptidomimetic, which may enhance the cellular uptake of the oligonucleotide agent.

[0172] Exemplary ligands can be coupled to the ligand-conjugate carrier directly or indirectly, preferably covalently, through a link between them.In exemplary embodiments, the ligands are coupled to the carrier through a link between them.In exemplary embodiments, the ligands change the distribution, targeting or life span of the RNA silencing agent incorporated.In exemplary embodiments, the ligands provide enhanced affinity for selected targets, such as molecules, cells or cell types, compartments, such as cell or organ compartments, tissues, organs or body regions, compared to species that do not have such ligands.

[0173] Exemplary ligands can improve the transport, hybridization and specificity properties, and can also improve the nuclease resistance of the resulting natural or modified RNA silencing agent or polymer molecule comprising 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 resistance-conferring moieties; and natural or non-natural nucleic acid bases. Common examples include lipophilic substances, lipids, steroids (e.g., uvaol, hecogenin, diosgenin), terpenes (e.g., triterpenes, such as sarsasapogenin, friedelin, epifriedelanol-derived 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 may 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 may 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 copolymers, poly(L-lactide-co-glycolide) copolymers, divinyl ether-maleic anhydride copolymers, N-(2-hydroxypropyl) methacrylamide copolymers (HMPA), polyethylene glycol (PEG), polyvinyl alcohol (PVA), polyurethanes, poly(2-ethylacrylic acid), N-isopropylacrylamide polymers or polyphosphazines.Examples of polyamines include polyethyleneimine, polylysine (PLL), spermine, spermidine, polyamines, pseudopeptide-polyamines, peptidomimetic polyamines, dendrimeric polyamines, arginine, amidines, protamines, cationic lipids, quaternary salts of cationic porphyrin polyamines or alpha-helical peptides.

[0174] The ligand may 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 a particular cell, such as a renal cell. The targeting group may be thiotropin, melanotropin, lectin, glycoprotein, surfactant protein A, mucin carbohydrate, polyvalent lactose, polyvalent galactose, N-acetyl-galactosamine, N-acetyl-glucosamine, polyvalent mannose, polyvalent fucose, glycosylated polyamino acids, polyvalent galactose, transferrin, bisphosphonates, polyglutamates, polyaspartates, lipids, cholesterol, steroids, bile acids, folic acid, vitamin B12, biotin, or an RGD peptide or RGD peptide mimetic. Other examples of ligands are dyes, intercalating agents (e.g. acridine and substituted acridines), 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), lipophilic molecules such as cholesterol (and its thio analogues), cholic acid, cholanic acid, lithocholic acid, adamantane acetic 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, bornenol, 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 peptide may be an alkylating agent, a phosphate, an amino, a mercapto, a PEG (e.g., PEG-40K), a MPEG, [MPEG]2, a polyamino, an alkyl, a substituted alkyl, a radiolabeled marker, an enzyme, a hapten (e.g., biotin), a transport / absorption enhancer (e.g., aspirin, naproxen, vitamin E, folic acid), a synthetic ribonuclease (e.g., imidazole, bisimidazole, histamine, imidazole cluster, acridine-imidazole conjugate, Eu3+ complex of a tetraazamacrocycle), a dinitrophenyl, HRP, or AP.

[0175] The ligand can be a protein, e.g., a glycoprotein, or a peptide, e.g., a molecule with specific affinity for a co-ligand, or an antibody, e.g., an antibody that binds to a particular cell type, such as cancer cells, endothelial cells, or bone cells. Ligands can also include hormones and hormone receptors. They can include lipids, lectins, carbohydrates, vitamins, cofactors, non-peptide species such as multivalent lactose, multivalent galactose, N-acetyl-galactosamine, N-acetyl-glucosamine, multivalent mannose, or multivalent fucose. The ligand can be, for example, lipopolysaccharide, p38 MAP kinase activator, or NF-κB activator.

[0176] The ligand can be a drug that can increase the uptake of the RNA silencing agent into cells by disrupting the substrate, for example, the cell's cytoskeleton, for example, by disrupting the cell's 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 uptake of the RNA silencing agent into cells, 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, for example, human serum albumin (HSA). HSA-binding ligands allow the distribution of the conjugate to target tissues, for example, non-kidney 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. The lipid or lipid-based ligand can (a) increase the resistance of the conjugate to degradation, (b) increase targeting or transport to the target cell or cell membrane, and / or (c) can be used to adjust binding to serum proteins, such as HSA. The lipid-based ligand can be used to regulate, for example, control the binding of the conjugate to target tissue. 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 the conjugate to the kidney. In a preferred embodiment, the lipid-based ligand binds to HSA. The lipid-based ligand can bind to HSA with sufficient affinity so that it is preferably distributed to non-renal tissues. However, it is preferred 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 renal 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.Exemplary agents are peptides 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 lipid-soluble and lipophilic phase.

[0179] The ligand can be a peptide or peptidomimetic. Peptidomimetics (also referred to herein as oligopeptidomimetics) are molecules that can fold into specific three-dimensional structures that resemble natural peptides. Binding of peptides and peptidomimetics to oligonucleotide agents affects the pharmacokinetic distribution of RNA silencing agents, for example, by enhancing cellular recognition and uptake. The peptide or peptidomimetic moiety can be about 5-50 amino acids long, for example, about 5, 10, 15, 20, 25, 30, 35, 40, 45 or 50 amino acids long. 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-restricting 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). The peptide or peptidomimetic can be encoded by a random sequence of DNA, such as peptides identified from a phage display library or one bead one compound (OBOC) combinatorial library (Lam et al., Nature 354:82-84, 1991). In an exemplary embodiment, 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 structural modifications described below can be utilized. EXAMPLES

[0180] Example 1 Chemical synthesis of di-RNA and vitamin D-conjugated hsiRNA The di-siRNA used in the 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 give a primary amine that was then attached to vitamin D (calciferol) via a carbamate linker. The ketal was then hydrolyzed to release a cis-diol that was 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 give three products; VitD, capped linker, and 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 includes the following steps: monoazidotetraethylene glycol has a branch point loaded as a tosylated solketal. 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 phosphate and phosphoamidate linkages. This example highlights an alternative and direct route of synthesis to produce 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 a solketal-modified tetraethylene glycol, the ketal is removed and the two primary hydroxyls are selectively protected 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 chemical synthesis of di-siRNA and vitamin D-conjugated hsiRNA, analytical HPLC was used to identify and quantify the synthesized products. Three main products were identified: siRNA sense strand capped with triethylene glycol (TEG) linker, di-siRNA and vitamin D-conjugated siRNA sense strand (Figure 3). Each product was isolated by HPLC and used in subsequent experiments. The chemical structures of the three main products synthesized are shown in Figure 3. HPLC conditions: 5-80% B (15 min), Buffer A (0.1M 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 method of chemical synthesis described in Example 1 was successful in producing the desired product of 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 chemical synthesis of di-siRNA and vitamin D-conjugated hsiRNA To evaluate the Htt gene silencing efficacy of each HPLC-isolated by-product from the chemical synthesis of di-siRNA and vitamin D-conjugated hsiRNA, HeLa cells were treated with each isolated compound by lipid-mediated transfection. Huntingtin mRNA expression was evaluated by Affymetrix Quantigene 2.0 and normalized to a housekeeping gene (PPIB). All four by-products resulted in significant Htt gene silencing 72 hours after transfection (Figure 22). Cellular uptake was tested in vivo by delivering each fluorescently labeled by-product to mice via intrastriatal injection and measuring uptake by fluorescent imaging. The di-siRNA product showed dramatically increased uptake in the hemisphere of injected mice compared to the other three by-products (Figure 22). For the four by-products resulting from the chemical synthesis reaction, di-siRNA showed 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 not delivered via lipids, cells were passively treated with Htt-di-siRNA for one week at various concentrations. 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, indicating that the di-branched siRNA structure was 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 by 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 extremely low accumulation throughout the injected hemisphere of the brain (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 does not accumulate in the liver or kidney 2 weeks after IS injection (Figure 13) and Htt mRNA is not significantly silenced in the liver or kidney after IS injection (Figure 23). The double-branched structure of di-siRNA improves distribution and neuronal uptake when compared to TEG-siRNA alone; therefore, the size and / or structure of the siRNA complex is important for efficacy. IS injection of Htt-di-siRNA results in a significant and stable reduction of Htt that continues to localize to the brain, a level of efficacy never shown for unconjugated siRNA.

[0189] Example 8 Other routes of administration 1 To evaluate the efficacy of delivery and activity of branched oligonucleotides in the spinal cord, di-HTT-Cy3 was delivered to mice by 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 routes of administration 2 In clinically relevant experiments, to evaluate the efficacy of branched oligonucleotide delivery and activity throughout the brain, di-HTT-Cy3 was delivered to mice by intracerebroventricular (ICV) injection. Both 2 days and 2 weeks after injection, di-siRNA accumulated throughout the brain (Figures 30-31). ICV of di-HTT-Cy3 also significantly silenced Htt mRNA and protein expression 2 weeks after injection (Figure 32). Further experiments showed 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, which is assumed to be a result of silencing the Htt gene (Figure 34). ICV injection administers di-siRNA directly into the cerebrospinal fluid (CSF) to avoid the blood-brain barrier, and this injection is used to treat diseases of the brain.

[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 may 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 by intravenous (IV) injection. Mice were injected with 20 mg / kg di-HTT-Cy3 on two consecutive days (total of 40 mg / kg) and sacrificed 24 hours after the last 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, indicating the ability of di-siRNA to cross the blood-brain barrier, an unpredictable result using therapeutic siRNA. IV injection indicates 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 Since 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), to evaluate the toxicity level in the brain after injection of di-HTT-Cy3, the protein amount of DARPP32 was evaluated in brain tissue. Mice were treated with 2 nmol of di-HTT-Cy3 (4 nmol of the corresponding antisense HTT chain) by IS or ICV injection. 14 days after injection, animals were sacrificed and tissue punches were taken from 300 μm brain sections derived from different regions of the brain. DARPP32 protein was quantified by immunoblot. 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 di-HTT-Cy3 (4 nmol of the corresponding antisense HTT chain) by IS or ICV injection. Mice were sacrificed 14 days after injection and tissue punches were taken from 300 μm brain sections from different regions of the brain. GFAP protein was quantified by immunoblot. 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, induction of gliomas is believed to be the result of 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 IS, ICV, intrathecal or IV 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 a 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 fluorescent imaging experiments, brain sections were stained with DAPI (blue) imaged using the Cy3 channel to detect accumulation of di-HTT-Cy3 (red).

[0197] Silencing To determine the efficacy of branched oligonucleotide silencing in vivo, mice were treated with di-HTT-Cy3 by IS, ICV, intrathecal or IV injection as described above in Examples 7-10. In all examples, 2 nmol of di-HTT-Cy3 (4 nmol of the corresponding antisense HTT strand) was injected 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 a 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 hydrophobic alkylenes or alkanes (Hy) bearing unprotected hydroxyl groups (or amines) that are phosphitylated with 2-cyanoethoxy-bis(N,N-diisopropylamino)phosphine (or other suitable phosphitylation reagent) 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 Incorporating hydrophobic moieties into branched oligonucleotide structures: Strategy 2 In another example, short / small aromatic planar molecules (Hy) bearing unprotected hydroxyl groups (or amines) with or without 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 bears 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 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 give 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. A branched oligonucleotide compound comprising two or more nucleic acids, said nucleic acids being linked together by one or more moieties selected from a linker, a spacer and a branch point.

2. The compound of claim 1 , comprising 2, 3, 4, 6 or 8 nucleic acids.

3. 2. The compound of claim 1, wherein each nucleic acid is single stranded and has a 5' end and a 3' end, and each nucleic acid is independently attached at the 5' end or the 3' end to a linker, a spacer and a branch point.

4. The compound of any one of claims 1 to 3, wherein each single-stranded nucleic acid independently comprises at least 15 contiguous nucleotides.

5. The compound according to any one of claims 1 to 4, wherein each nucleic acid comprises one or more chemically modified nucleotides.

6. The compound of claim 5 , wherein each nucleic acid is composed of chemically modified nucleotides.

7. The compound of claim 1 , wherein each nucleic acid is double-stranded and comprises a sense strand and an antisense strand, each of which comprises a 5′ end and a 3′ end.

8. The compound of claim 7 , wherein the sense strand and the antisense strand each contain one or more chemically modified nucleotides.

9. The compound of claim 8 , wherein the sense strand and the antisense strand each comprise chemically modified nucleotides.

10. 10. The compound of claim 9, wherein both the sense and antisense strands comprise alternating 2'-methoxy and 2'-fluoro nucleotides.

11. The compound according to any one of claims 7 to 10, wherein 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.

12. The compound according to any one of claims 7 to 11, wherein each double-stranded nucleic acid is independently linked at the 3'-end or 5'-end of the sense strand or the antisense strand.

13. 13. The compound of any one of claims 7 to 12, wherein the antisense strand comprises at least 16, at least 17, at least 18, at least 19, or at least 20 consecutive nucleotides and has complementarity to the target.

14. The compound according to any one of claims 7 to 13, wherein the nucleotide at positions 1 to 6 from the 3' end or positions 1 to 7 from the 3' end is linked to an adjacent nucleotide by a phosphorothioate bond.

15. The compound of any one of claims 7 to 14, wherein the compound further comprises a hydrophobic moiety attached to the terminal 5' position of the branched oligonucleotide compound.

16. 16. The compound of claim 15, wherein the hydrophobic moiety comprises an alkyl, alkenyl or aryl moiety, a vitamin or cholesterol derivative, a fat-soluble amino acid, or a combination thereof.

17. 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 atom of the linker may optionally be replaced with a nitrogen atom, may have a hydroxyl substituent, or may have an oxo substituent; A compound according to any one of claims 1 to 16.

18. Formula (I): 【Chemistry 1】 [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. Compound.

19. Formulae (I-1) to (I-9): 【Table 1】 20. The compound of claim 18 having a structure selected from:

20. The antisense strand is 【Chemistry 2】 20. The compound of claim 18, comprising a 5' terminal group R selected from the group consisting of:

21. Formula (II): 【Chemistry 3】 [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 a mismatch, respectively.

20. The compound of claim 18 having the structure:

22. Formula (II): 【Chemistry 4】 [During the ceremony, Each X is independently a nucleotide that contains a 2'-deoxy-2'-fluoro modification; Each X is independently a nucleotide containing a 2'-O-methyl modification; Each Y is independently a nucleotide that includes a 2'-deoxy-2'-fluoro modification; and each Y is independently a nucleotide that contains a 2'-O-methyl modification; 21. The compound of claim 20 having the structure:

23. Formula (IV): 【Chemistry 5】 [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 a 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.

23. The compound of claim 22 having the structure:

24. Formula (V): 【Chemistry 6】 [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 a mismatch, respectively.

20. The compound of claim 18 having the structure:

25. Formula (VI): 【Chemistry 7】 [During the ceremony, Each X is independently a nucleotide that contains a 2'-deoxy-2'-fluoro modification; Each X is independently a nucleotide containing a 2'-O-methyl modification; Each Y is independently a nucleotide that includes a 2'-deoxy-2'-fluoro modification; and each Y is independently a nucleotide that contains a 2'-O-methyl modification; 25. The compound of claim 24 having the structure:

26. Formula (VII): 【Chemistry 8】 [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 a 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; Each Y is independently a nucleotide that includes a 2'-deoxy-2'-fluoro modification; and each Y is independently a nucleotide that contains a 2'-O-methyl modification; 25. The compound of claim 24 having the structure:

27. L is L1: 【Chemistry 9】 The compound of any one of claims 18 to 26, having the structure:

28. R is R ３ and n is 2.

29. L is L2: 【Chemistry 10】 The compound of any one of claims 18 to 26, having the structure:

30. R is R ３ and n is 2.

31. Formula (VIII): 【Chemistry 11】 [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) 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; Each cNA is independently a carrier nucleic acid that contains one or more chemical modifications; and n is 2, 3, 4, 5, 6, 7 or 8. A delivery system for therapeutic nucleic acids having the structure of:

32. Formulas (VIII-1) to (VIII-9): 【Table 2】 32. The delivery system of claim 31 having a structure selected from:

33. The delivery system of claim 31 , wherein each cNA independently comprises at least 15 consecutive nucleotides.

34. The delivery system of claim 31 , wherein each cNA is independently composed of a chemically modified nucleotide.

35. The delivery system of claim 31, further comprising n therapeutic nucleic acids (NAs), wherein each NA is hybridized to at least one cNA.

36. The delivery system of claim 35, wherein each NA independently comprises at least 16 consecutive nucleotides.

37. The delivery system of claim 35, wherein each NA independently comprises 16 to 20 consecutive nucleotides.

38. The delivery system of claim 35, wherein each NA comprises an unpaired overhang of at least two nucleotides.

39. 36. The delivery system of claim 35, wherein the nucleotides of the overhang are linked by phosphorothioate linkages.

40. 36. The delivery system of claim 35, wherein each NA is independently selected from the group consisting of DNA, siRNA, antagomir, miRNA, gapmer, mixmer or guide RNA.

41. 36. The delivery system of claim 35, wherein each NA is identical.

42. 36. The delivery system of claim 35, wherein each NA is not identical.

43. A delivery system according to claim 35, having a structure according to any one of claims 16 to 28.

44. 32. The delivery system of claim 31, wherein the target of delivery is selected from the brain, liver, skin, kidney, spleen, pancreas, colon, fat, lung, muscle and thymus.

Images