EA043057B1 - MEANS THAT ARE MODIFIED DOUBLE-STRANDED RNA - Google Patents

Description

This application claims the benefit of the priority of U.S. Provisional Application No. 62/093919, filed December 18, 2014; U.S. Provisional Application No. 62/083744, filed November 24, 2014; and U.S. Provisional Application No. 62/039507, filed August 20, 2014, all of which are incorporated herein by reference in their entirety.

Field of invention

The present invention relates to RNAi duplex agents with specific motifs that are preferred for suppressing target gene expression, as well as RNAi compositions suitable for therapeutic use. In addition, the present invention provides methods for suppressing the expression of a target gene by administering these RNAi duplex agents, for example, for the treatment of various diseases.

Prerequisites for the creation of the invention

RNA interference or RNAi is a term originally coined by Fire and collaborators to describe the observation that double-stranded RNA (dsRNA) for RNAi can block gene expression (Fire et al. (1998) Nature 391, 806-811; Elbashir et al (2001) Genes Dev 15, 188-200). The short dsRNA drives gene-specific post-transcriptional silencing in many organisms, including vertebrates, and has provided a new tool for studying gene function. RNAi is mediated by the RNA-inducible silencing complex (RISC), a sequence-specific, multicomponent nuclease that degrades messenger RNAs homologous to the silencing initiator. It is known that RISC contains short RNAs (approximately 22 nucleotides in length) derived from a double-stranded RNA initiator, but the protein components of this active complex remain unknown.

Double-stranded RNA (dsRNA) molecules with proper gene silencing properties are needed for drug development based on RNA interference (RNAi). The initial stage of RNAi is the activation of the RNA-induced silencing complex (RISC), which requires the destruction of the sense strand from the dsRNA duplex. It was believed that the sense strand acts as the first substrate for RISC, which is cleaved by the Argonaute 2 protein in the middle of the duplex region. Immediately after the cleaved 5' and 3' sense strand fragments are removed from the Ago2 endonuclease, RISC becomes activated by the antisense strand (Rand et al. (2005) Cell 123, 621).

It has been suggested that endonucleotic cleavage of the target mRNA is impaired when sense strand cleavage is suppressed (Leuschner et al. (2006) EMBO Rep., 7, 314; Rand et al. (2005) Cell 123, 621; Schwarz et al. (2004) Curr. Biol 14, 787). Leuschner et al. showed that insertion of 2'-O-Me-ribose into the cleavage site for Ago2 in the sense strand suppresses RNAi in HeLa cells (Leuschner et al. (2006) EMBO Rep., 7, 314). A similar effect was observed for phosphorothioate modifications, indicating that semantic strand cleavage is required for efficient RNAi also in mammals.

Morrissey et al. used a siRNA duplex containing 2'-P-modified residues, among other sites and modifications, also at the cleavage site for Ago2, and observed silencing activity comparable to that of unmodified siRNAs (Morrissey et al. (2005) Hepatology 41, 1349 ). However, the modification made by Morrissey is not motif-specific, for example, one modification includes 2'-F modifications on all pyrimidine bases in both the sense and antisense strands, as long as a pyrimidine residue is present without any selectivity; therefore, based on these data, it remains unclear whether modification of a specific motif at the semantic strand cleavage site can have any actual effect on gene silencing activity.

Muhonen et al. used a siRNA duplex containing two 2'-P modified residues at the cleavage site for Ago2 on the sense or antisense strand and found that this was acceptable (Muhonen et al. (2007) Chemistry & Biodiversity 4, 858-873). However, the modification made by Muhonen is also sequence specific, for example, in each particular strand, Muhonen modified only either all pyrimidine bases or all purine bases without any selectivity.

Choung et al. used siRNA duplex having intermittent modifications with 2'OMe or various combinations of 2'-F, 2'-OMe and phosphorothioate modifications to stabilize siRNA Surl0058 in serum (Choung et al. (2006) Biochemical and Biophysical Research Communications 342, 919-927). Choung suggested that residues at the antisense strand cleavage site should not be modified with 2'-OMe to increase siRNA stability.

Thus, there is a continuing need for RNA interference duplex agents to increase the gene silencing efficiency of siRNA-based gene therapy agents. The present invention addresses this need.

Brief description of the invention

The present invention provides effective nucleotide or chemical motifs for dsRNA agents, optionally conjugated to at least one ligand, which are preferred for suppressing target gene expression, as well as RNAi compositions suitable for therapeutic use.

- 1 043057

In one aspect, the present invention relates to a double-stranded RNA (dsRNA) agent for suppressing the expression of a target gene. The dsRNA tool contains a sense strand and an antisense strand, with each strand containing 14 to 40 nucleotides. The dsRNA agent is represented by formula (I):

In formula (I), each of Bl, B2, B3, C D, B2', B3' and B4' is independently a nucleotide having a modification selected from the group consisting of 2'-O-alkyl, 2'-substituted alkoxy , 2'-substituted alkyl, 2'-halogen, ΕΝΑ and BNA/LNA. In one embodiment, Bl, B2, B3, B G, B2', B3', and B4' each have 2'-OMe modifications. In one embodiment, each of Bl, B2, B3, B G, B2', B3', and B4' has a 2'-OMe or 2'-E modification. In one embodiment, at least one of Bl, B2, B3, C D, B2', B3', and B4' has a 2'-Ο-Ν-methylacetamido (2'-O-NMA) modification.

C1 is a thermostability-disrupting nucleotide placed at the site opposite the seed portion of the antisense strand (ie, located at positions 2-8 from the 5' end of the antisense strand). For example, C1 is at a position on the sense strand that pairs with a nucleotide at positions 2-8 from the 5' end of the antisense strand. In one example, C1 is at position 15 from the 5' end of the sense strand. The C1 nucleotide carries a thermostability-disrupting modification, which may include a modification to remove a nitrogenous base; mismatch with the opposite nucleotide in the duplex; and a sugar modification such as a 2'-deoxy modification or an acyclic nucleotide, eg unlocked nucleic acids (UNA) or glyceric nucleic acid (GNA). In one embodiment, C1 has a thermostability-disrupting modification selected from the group consisting of:

i) mismatch with the opposite nucleotide in the antisense strand;

j) modification to remove a nitrogenous base selected from the group consisting of

and iii) a sugar modification selected from the group consisting of

where B is a modified or unmodified nucleobase;

R ¹ and R ² are independently H, halogen, OR ₃ or alkyl; And

R ₃ is H, alkyl, cycloalkyl, aryl, aralkyl, heteroaryl or sugar.

In one embodiment, the thermostability-damaging modification in C1 is a mismatch selected from the group consisting of G:G, G:A, G:U, G:T, A:A, A:C, C:C, C: U, C:T, U:U, T:T and U:T; and optionally at least one nucleobase in the mismatched pair is a 2'-deoxynucleobase. According to one example, the thermostability-disrupting modification in C1 is GNA or

Each of ΤΙ, TG, T2' and TK' is independently a nucleotide having a modification that gives the nucleotide a steric volume that is less than or equal to the steric volume of the 2'-OM modification. The steric volume means the sum of the steric effects of the modification. Methods for determining the steric effects of a nucleotide modification are known to those skilled in the art. The modification may be at the 2' position of the ribose sugar of the nucleotide, or the modification may be at the nucleotide lacking ribose, an acyclic nucleotide, or at the backbone of the nucleotide at a position that is analogous or similar to the 2' position on the ribose sugar and imparts steric bulk to the nucleotide, which is less than or equal to the steric volume of the 2'-OMe modification. For example, ΤΙ, TG, T2', and T3' are each independently selected from DNA, RNA, LNA, 2'-F, and 2'-F-5'methyl. In one embodiment, T1 is DNA. In one embodiment, the TG is DNA, RNA, or LNA. In one embodiment, T2' is DNA or RNA. In one embodiment, TK' is DNA or RNA.

The length of p ¹ , p ³ and q ¹ is independently 4 to 15 nucleotides.

The length of p ⁵ , q ³ and q ⁷ is independently 1-6 nucleotides.

The length of p ⁴ , q ² and q ⁶ is independently 1-3 nucleotides; alternatively n ⁴ is 0.

The length of q ⁵ is independently 0-10 nucleotides.

-2043057

The length of n ² and q ⁴ is independently 0-3 nucleotides.

Alternatively, n ⁴ is 0-3 nucleotides in length.

In one embodiment, n ⁴ may be 0. In one example, n ⁴ is 0, aq ² and q ⁶ are 1. In another example, n ⁴ is 0, aq ² and q ⁶ are 1, and there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2 and two phosphorothioate modifications of internucleotide bonds within positions 18-23 of the antisense strand (counting from 5' end of the antisense strand).

In one embodiment, each of n ⁴ , q ^{2 ,} and q ⁶ is 1.

According to one embodiment, each of n ² , n ⁴ , q ² , q ⁴ and q ⁶ is 1.

In one embodiment, C1 is at positions 14-17 from the 5' end of the sense strand if the sense strand is 19-22 nucleotides long and n ⁴ is 1. In one embodiment, C1 is at position 15 from the 5' end of the sense strand .

In one embodiment, T3' starts at position 2 from the 5' end of the antisense strand. In one example, T3' is at position 2 from the 5' end of the antisense strand and q ⁶ is 1.

In one embodiment, T1' starts at position 14 from the 5' end of the antisense strand. In one example, T1' is at position 14 from the 5' end of the antisense strand and q ² is 1.

In an exemplary embodiment, T3' starts at position 2 from the 5' end of the antisense strand and T1' starts at position 14 from the 5' end of the antisense strand. In one example, T3' starts at position 2 from the 5' end of the antisense strand and q ⁶ is 1, and T1' starts at position 14 from the 5' end of the antisense strand and q ² is 1.

In one embodiment, T1' and T3' are separated by an 11 nucleotide long region (ie, excluding ΊΤ and T3' nucleotides).

In one embodiment, T1' is at position 14 from the 5' end of the antisense strand. In one example, T1' is at position 14 from the 5' end of the antisense strand and q ² is 1, and the modification is at the 2' position or positions in the ribose-free, acyclic nucleotide, or nucleotide backbone, which gives it less steric volume than that of ribose with 2'-OMe modification.

In one embodiment, T3' is at position 2 from the 5' end of the antisense strand. In one example, T3' is at position 2 from the 5' end of the antisense strand and q ⁶ is 1, and modification at the 2' position or positions in a ribose-free, acyclic nucleotide, or nucleotide backbone confers less steric volume than ribose with 2'-OMe modification or equal to it.

In one embodiment, T1 is located at the cleavage site of the sense strand. In one example, T1 is at position 11 from the 5' end of the sense strand if the sense strand is 19-22 nucleotides long and n ² is 1. In an exemplary embodiment, T1 is at the cleavage site of the sense strand at position 11 from the 5' end sense strand if the length of the sense strand is 19-22 nucleotides and n ² is equal to 1.

In one embodiment, T2' starts at position 6 from the 5' end of the antisense strand. In one example, T2' is at positions 6-10 from the 5' end of the antisense strand and q ⁴ is 1.

In an exemplary embodiment, T1 is at the cleavage site of the sense strand, eg, position 11 from the 5' end of the sense strand, if the sense strand is 1922 nucleotides long and n ² is 1; ΊΤ is at position 14 from the 5' end of the antisense strand and q ² is 1, with the modification at T1' being at the 2' position of the ribose sugar or at positions in the ribose-free, acyclic nucleotide, or nucleotide backbone that gives it lower steric volume than ribose with 2'-OMe modification; T2' is at positions 6-10 from the 5' end of the antisense strand and q ⁴ is 1; and T3' is at position 2 from the 5' end of the antisense strand and q ⁶ is 1, with the T3' modification being at the 2' position or at positions in a nucleotide lacking ribose, an acyclic nucleotide, or a nucleotide backbone, giving it a smaller steric volume than ribose with 2'-OMe modification. or equal to it.

In one embodiment, T2' starts at position 8 from the 5' end of the antisense strand. In one example, T2' starts at position 8 from the 5' end of the antisense strand and q ⁴ is 2.

In one embodiment, T2' starts at position 9 from the 5' end of the antisense strand. In one example, T2' is at position 9 from the 5' end of the antisense strand and q ⁴ is 1.

In one embodiment, B1' is 2'-OMe or 2'-F, q ¹ is 9, ΊΤ is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 1, B3' is 2'-OMe or 2'-F, q ⁵ is 6, T3' is 2'-F , q ⁶ is 1, B4' is 2'-OMe and q ⁷ is 1; there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18- 23 of the antisense strand (counting from the 5' end of the antisense strand).

In one embodiment, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 1, B3' is 2'-OMe or 2'- F, q ⁵ is 6, T3' is 2'-F, q ⁶ is 1, B4' is 2'-OMe and q ⁷ is 1; there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18- 23 of the antisense strand (counting from the 5' end of the antisense strand).

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 2, B3' is 2'-OMe or 2'-F, q ⁵ is 5, T3' is 2'-F, q ⁶ is 1, B4' is 2'-OMe and q ⁷ is 1.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 2, B3' is 2'-OMe or 2'-F, q ⁵ is 5, T3' is 2'-F, q ⁶ is 1, B4' is 2'-OMe and q ⁷ is 1; there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18- 23 antisense strand (counting from the 5' end of the antisense strand).

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 6, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 7, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 2, B3' is 2'-OMe or 2'-F, q ⁵ is 5, T3' is 2'-F, q ⁶ is 1, B4' is 2'-OMe and q ⁷ is 1.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 6, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 7, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 2, B3' is 2'-OMe or 2'-F, q ⁵ is 5, T3' is 2'-F, q ⁶ is 1, B4' is 2'-OMe and q ⁷ is 1; there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18- 23 of the antisense strand (counting from the 5' end of the antisense strand).

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 1, B3' is 2'-OMe or 2'-F, q ⁵ is 6, T3' is 2'-F, q ⁶ is 1, B4' is 2'-OMe and q ⁷ is 1.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 1, B3' is 2'-OMe or 2'-F, q ⁵ is 6, T3' is 2'-F, q ⁶ is 1, B4' is 2'-OMe and q ⁷ is 1; there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18- 23 antisense strand (counting from the 5' end of the antisense strand).

- 4 043057

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 5, T2' is 2'-F, q ⁴ is 1, B3' is 2'-OMe or 2'-F, q ⁵ is 5, T3' is 2'-F, q ⁶ is 1, B4' is 2'-OMe and q ⁷ is 1; optionally, at least 2 additional TTs are at the 3' end of the antisense strand.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 5, T2' is 2'-F, q ⁴ is 1, B3' is 2'-OMe or 2'-F, q ⁵ is 5, T3' is 2'-F, q ⁶ is 1, B4' is 2'-OMe and q ⁷ is 1; optionally, at least 2 additional TTs are at the 3' end of the antisense strand; in this case, there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18- 23 of the antisense strand (counting from the 5' end of the antisense strand).

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, ΊΤ is 2'-F, q ² is 1, B2' is is 2'-OMe or 2'-F, q ³ is 4, q ⁴ is 0, B3' is 2'-OMe or 2'-F, q ⁵ is 7, T3' is 2'-F, q ⁶ is 1, B4' is 2'-OMe and q ⁷ is 1.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, ΊΤ is 2'-F, q ² is 1, B2' is is 2'-OMe or 2'-F, q ³ is 4, q ⁴ is 0, B3' is 2'-OMe or 2'-F, q ⁵ is 7, T3' is 2'-F, q ⁶ is 1, B4' is 2'-OMe and q ⁷ is 1; there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18-23 of the antisense strands (counting from the 5' end).

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, ΊΤ is 2'-F, q ² is 1, B2' is is 2'-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 2, B3' is 2'-OMe or 2'-F, q ⁵ is 5, T3 ' is 2'-F, q ⁶ is 1, B4' is 2'-F and q ⁷ is 1.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 2, B3' is 2'-OMe or 2'-F, q ⁵ is 5, T3' is 2'-F, q ⁶ is 1, B4' is 2'-F and q ⁷ is 1; in this case, there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18- 23 of the antisense strand (counting from the 5' end of the antisense strand).

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, q ⁴ is 0, B3' is 2'-OMe or 2'-F, q ⁵ is 7, T3' is 2'-F, q ⁶ is 1, B4' is 2'-F and q ⁷ is 1.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, q ⁴ is 0, B3' is 2'-OMe or 2'-F, q ⁵ is 7, T3' is 2'-F, q ⁶ is 1, B4' is 2'-F and q ⁷ is 1; while there are two phosphorothioate modifications

- 5 043057 internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2 and two phosphorothioate modifications of internucleotide bonds within positions 18-23 of the antisense strand (counting from the 5' end of the antisense strand).

The dsRNA agent may have a phosphorus-containing group at the 5' end of the sense strand or antisense strand. The 5' phosphorus containing group may be 5' phosphate (5'-P), 5' phosphorothioate (5'-PS), 5' phosphorodithioate (5'-PS ₂ ),

5' vinyl phosphonate (5'-VP), 5' methylphosphonate (MePhos) or 5'-deoxy-5'-C-malonyl

). If the 5'-terminal phosphorus-containing group is a 5'-vinyl phosphonate (5'-VP), then the 5'-VP can be either the 5'-E-VP isomer (i.e., trans-vinyl phosphate, J) or the 5'-Z-VP isomer (i.e., cis-vinyl phosphate, +), or mixtures thereof.

In one embodiment, the dsRNA agent has a phosphorus-containing group at the 5' end of the sense strand. In one embodiment, the dsRNA agent has a phosphorus-containing group at the 5' end of the antisense strand.

In one embodiment, the dsRNA agent contains 5'-P. In one embodiment, the dsRNA agent contains the 5'-P in the antisense strand.

In one embodiment, the dsRNA agent contains 5'PS. In one embodiment, the dsRNA agent contains 5'-PS in the antisense strand.

In one embodiment, the dsRNA agent contains a 5'VP. In one embodiment, the dsRNA agent contains a 5'-VP in the antisense strand. In one embodiment, the dsRNA agent contains the 5'-E-VP in the antisense strand. In one embodiment, the dsRNA agent contains the 5'-Z-VP in the antisense strand.

In one embodiment, the dsRNA agent contains 5'PS ₂ . In one embodiment, the dsRNA agent contains 5'-PS ₂ in the antisense strand.

In one embodiment, the dsRNA agent contains 5'PS ₂ . In one embodiment, the dsRNA agent contains 5'deoxy-5'-C-malonyl in the antisense strand.

In one embodiment, B1 is 2'-OMe or 2'-F, η ¹ is 8, T1 is 2' F, equals 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, ΊΤ is 2'-F, q ² is 1, B2' is 2 '-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 2, 3' is 2'-OMe or 2'-F, q ⁵ is 5, T3' is is 2'-F, q ⁶ is 1, B4' is 2'-OMe, and q ⁷ is 1. The dsRNA agent also contains 5'-PS.

In one embodiment, B1 is 2'-OMe or 2'-F, η ¹ is 8, T1 is 2' F, equals 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, ΊΤ is 2'-F, q ² is 1, B2' is 2 '-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 2, 3' is 2'-OMe or 2'-F, q ⁵ is 5, T3' is is 2'-F, q ⁶ is 1, B4' is 2'-OMe, and q ⁷ is 1. The dsRNA agent also contains 5'-P.

In one embodiment, B1 is 2'-OMe or 2'-F, η ¹ is 8, T1 is 2' F, equals 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, ΊΤ is 2'-F, q ² is 1, B2' is 2 '-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 2, 3' is 2'-OMe or 2'-F, q ⁵ is 5, T3' is is 2'-F, q ⁶ is 1, B4' is 2'-OMe, and q ⁷ is 1. The dsRNA agent also contains 5'-VP. The 5'-VP may be 5'-E-VP, 5'-Z-VP, or a combination thereof.

In one embodiment, B1 is 2'-OMe or 2'-F, η ¹ is 8, T1 is 2' F, equals 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, ΊΤ is 2'-F, q ² is 1, B2' is 2 '-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 2, 3' is 2'-OMe or 2'-F, q ⁵ is 5, T3' is is 2'-F, q ⁶ is 1, B4' is 2'-OMe, and q ⁷ is 1. The dsRNA agent also contains 5'-PS ₂ .

-6043057

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 2, B3' is 2'-OMe or 2'-F, q ⁵ is 5, T3' is 2'-F, q ⁶ is 1, B4' is 2'-OMe and q ⁷ is 1. The dsRNA agent also contains 5'-deoxy-5'-C-malonyl.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 2, B3' is 2'-OMe or 2'-F, q ⁵ is 5, T3' is 2'-F, q ⁶ is 1, B4' is 2'-OMe and q ⁷ is 1; there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18- 23 of the antisense strand (counting from the 5' end of the antisense strand). The dsRNA tool also contains 5'-P.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 2, B3' is 2'-OMe or 2'-F, q ⁵ is 5, T3' is 2'-F, q ⁶ is 1, B4' is 2'-OMe and q ⁷ is 1; there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18- 23 of the antisense strand (counting from the 5' end of the antisense strand). The dsRNA tool also contains 5'-PS.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 2, B3' is 2'-OMe or 2'-F, q ⁵ is 5, T3' is 2'-F, q ⁶ is 1, B4' is 2'-OMe and q ⁷ is 1; in this case, there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18- 23 antisense strand (counting from the 5' end of the antisense strand). The dsRNA tool also contains 5'-VP. The 5'-VP may be 5'-E-VP, 5'-Z-VP, or a combination thereof.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 2, B3' is 2'-OMe or 2'-F, q ⁵ is 5, T3' is 2'-F, q ⁶ is 1, B4' is 2'-OMe and q ⁷ is 1; there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18- 23 of the antisense strand (counting from the 5' end of the antisense strand). The dsRNA tool also contains 5'-PS ₂ .

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, ΊΤ is 2'-F, q ² is 1, B2' is is 2'-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 2, B3' is 2'-OMe or 2'-F, q ⁵ is 5, T3 ' is 2'-F, q ⁶ is 1, B4' is 2'-OMe and q ⁷ is 1; there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18- 23 antisense strand (counting from the 5' end of the antisense strand). The dsRNA agent also contains 5'-deoxy-5'-C-malonyl.

- 7 043057

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, q ⁴ is 0, B3' is 2'-OMe or 2'-F, q ⁵ is 7, T3' is 2'-F, q ⁶ is 1, B4' is 2'-OMe and q ⁷ is 1. The dsRNA agent also contains 5'-P.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, q ⁴ is 0, B3' is 2'-OMe or 2'-F, q ⁵ is 7, T3' is 2'-F, q ⁶ is 1, B4' is 2'-OMe and q ⁷ is 1. The dsRNA agent also contains 5'-PS.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, q ⁴ is 0, B3' is 2'-OMe or 2'-F, q ⁵ is 7, T3' is 2'-F, q ⁶ is 1, B4' is 2'-OMe and q ⁷ is 1. The dsRNA agent also contains 5'-VP. The 5'-VP may be 5'-E-VP, 5'-Z-VP, or a combination thereof.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, q ⁴ is 0, B3' is 2'-OMe or 2'-F, q ⁵ is 7, T3' is 2'-F, q ⁶ is 1, B4' is 2'-OMe and q ⁷ is 1. The dsRNA agent also contains 5'-PS ₂ .

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, q ⁴ is 0, B3' is 2'-OMe or 2'-F, q ⁵ is 7, T3' is 2'-F, q ⁶ is 1, B4' is 2'-OMe and q ⁷ is 1. The dsRNA agent also contains 5'-deoxy-5'-C-malonyl.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, q ⁴ is 0, B3' is 2'-OMe or 2'-F, q ⁵ is 7, T3' is 2'-F, q ⁶ is 1, B4' is 2'-OMe and q ⁷ is 1; there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18-23 of the antisense strands (counting from the 5' end). The dsRNA tool also contains 5'-P.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, q ⁴ is 0, B3' is 2'-OMe or 2'-F, q ⁵ is 7, T3' is 2'-F, q ⁶ is 1, B4' is 2'-OMe and q ⁷ is 1; there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18-23 of the antisense strands (counting from the 5' end). The dsRNA tool also contains 5'-PS.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, q ⁴ is 0, B3' is 2'-OMe or 2'-F, q ⁵ is 7, T3' is 2'-F, q ⁶ is 1, B4' is 2'-OMe and q ⁷ is 1; while there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the semantic strand (counting from the 5'-end), as well as

- 8 043057 two phosphorothioate modifications of internucleotide bonds at positions 1 and 2 and two phosphorothioate modifications of internucleotide bonds within positions 18-23 of the antisense strand (counting from

5' end). The dsRNA tool also contains 5'-VP. The 5'-VP may be 5'-E-VP, 5'-Z-VP, or a combination thereof.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, q ⁴ is 0, B3' is 2'-OMe or 2'-F, q ⁵ is 7, T3' is 2'-F, q ⁶ is 1, B4' is 2'-OMe and q ⁷ is 1; there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18-23 of the antisense strands (counting from the 5' end). The dsRNA tool also contains 5'-PS ₂ .

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, q ⁴ is 0, B3' is 2'-OMe or 2'-F, q ⁵ is 7, T3' is 2'-F, q ⁶ is 1, B4' is 2'-OMe and q ⁷ is 1; there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18-23 of the antisense strands (counting from the 5' end). The dsRNA agent also contains 5'-deoxy-5'-C-malonyl.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 2, B3' is 2'-OMe or 2'-F, q ⁵ is 5, T3' is 2'-F, q ⁶ is 1, B4' is 2'-F and q ⁷ is 1. The dsRNA agent also contains 5'-P.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 2, B3' is 2'-OMe or 2'-F, q ⁵ is 5, T3' is 2'-F, q ⁶ is 1, B4' is 2'-F and q ⁷ is 1. The dsRNA agent also contains 5'-PS.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 2, B3' is 2'-OMe or 2'-F, q ⁵ is 5, T3' is 2'-F, q ⁶ is 1, B4' is 2'-F and q ⁷ is 1. The dsRNA agent also contains 5'-VP. The 5'-VP may be 5'-E-VP, 5'-U-VP, or a combination thereof.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 2, B3' is 2'-OMe or 2'-F, q ⁵ is 5, T3' is 2'-F, q ⁶ is 1, B4' is 2'-F and q ⁷ is 1. The dsRNA agent also contains 5'-PS ₂ .

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 2, B3' is 2'-OMe or 2'-F, q ⁵ is 5, T3' is 2'-F, q ⁶ is 1, B4' is 2'-F and q ⁷ is 1. The dsRNA agent also contains 5'-deoxy-5'-C-malonyl.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0,

- 9 043057

B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 2, B3' is 2'-OMe or 2'-F, q ⁵ is 5, T3' is 2'-F, q ⁶ is 1, B4' is 2'-F and q ⁷ is 1; in this case, there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18- 23 antisense strand (counting from the 5' end of the antisense strand). The dsRNA tool also contains 5'-P.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 2, B3' is 2'-OMe or 2'-F, q ⁵ is 5, T3' is 2'-F, q ⁶ is 1, B4' is 2'-F and q ⁷ is 1; in this case, there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18- 23 antisense strand (counting from the 5' end of the antisense strand). The dsRNA tool also contains 5'-PS.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 2, B3' is 2'-OMe or 2'-F, q ⁵ is 5, T3' is 2'-F, q ⁶ is 1, B4' is 2'-F and q ⁷ is 1; there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18- 23 of the antisense strand (counting from the 5' end of the antisense strand). The dsRNA tool also contains 5'-VP. The 5'-VP may be 5'-E-VP, 5'-Z-VP, or a combination thereof.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 2, B3' is 2'-OMe or 2'-F, q ⁵ is 5, T3' is 2'-F, q ⁶ is 1, B4' is 2'-F and q ⁷ is 1; there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18- 23 of the antisense strand (counting from the 5' end of the antisense strand). The dsRNA tool also contains 5'-PS ₂ .

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 2, B3' is 2'-OMe or 2'-F, q ⁵ is 5, T3' is 2'-F, q ⁶ is 1, B4' is 2'-F and q ⁷ is 1; there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18- 23 of the antisense strand (counting from the 5' end of the antisense strand). The dsRNA agent also contains 5'-deoxy-5'-C-malonyl.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, q ⁴ is 0, B3' is 2'-OMe or 2'-F, q ⁵ is 7, T3' is 2'-F, q ⁶ is 1, B4' is 2'-F, and q ⁷ is 1. The dsRNA agent also contains 5'-P.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0,

- 10 043057

B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, ΊΤ is 2'-F, q ² is 1, B2' is 2 '-OMe or 2'-F, q ³ is 4, q ⁴ is 0, B3' is 2'-OMe or 2'-F, q ⁵ is 7, T3' is 2'-F, q ⁶ is

1, B4' is 2'-F and q ⁷ is 1. The dsRNA agent also contains

5'-PS.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, q ⁴ is 0, B3' is 2'-OMe or 2'-F, q ⁵ is 7, T3' is 2'-F, q ⁶ is 1, B4' is 2'-F and q ⁷ is 1. The dsRNA agent also contains 5'-VP. The 5'-VP may be 5'-E-VP, 5'-Z-VP, or a combination thereof.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, q ⁴ is 0, B3' is 2'-OMe or 2'-F, q ⁵ is 7, T3' is 2'-F, q ⁶ is 1, B4' is 2'-F and q ⁷ is 1. The dsRNA agent also contains 5'-PS2.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, q ⁴ is 0, B3' is 2'-OMe or 2'-F, q ⁵ is 7, T3' is 2'-F, q ⁶ is 1, B4' is 2'-F and q ⁷ is 1. The dsRNA agent also contains 5'-deoxy-5'-C-malonyl.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, q ⁴ is 0, B3' is 2'-OMe or 2'-F, q ⁵ is 7, T3' is 2'-F, q ⁶ is 1, B4' is 2'-F and q ⁷ is 1; there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18- 23 of the antisense strand (counting from the 5' end of the antisense strand). The dsRNA tool also contains 5'-P.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, ΊΤ is 2'-F, q ² is 1, B2' is is 2'-OMe or 2'-F, q ³ is 4, q ⁴ is 0, B3' is 2'-OMe or 2'-F, q ⁵ is 7, T3' is 2'-F, q ⁶ is 1, B4' is 2'-F and q ⁷ is 1; there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18- 23 of the antisense strand (counting from the 5' end of the antisense strand). The dsRNA tool also contains 5'-PS.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, q ⁴ is 0, B3' is 2'-OMe or 2'-F, q ⁵ is 7, T3' is 2'-F, q ⁶ is 1, B4' is 2'-F and q ⁷ is 1; there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18- 23 of the antisense strand (counting from the 5' end of the antisense strand). The dsRNA tool also contains 5'-VP. The 5'-VP may be 5'-E-VP, 5'-Z-VP, or a combination thereof.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, ΊΤ

- 11 043057 is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, q ⁴ is 0, B3' is 2'-OMe or 2' -F, q ⁵ is 7, T3' is 2'-F, q ⁶ is 1, B4' is 2'-F and q ⁷ is 1; there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18- 23 of the antisense strand (counting from the 5' end of the antisense strand). The dsRNA tool also contains 5'-PS2.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, q ⁴ is 0, B3' is 2'OMe or 2'-F, q ⁵ is 7, T3' is 2'-F, q ⁶ is 1, B4' is 2'-F and q ⁷ is 1; there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18- 23 of the antisense strand (counting from the 5' end of the antisense strand). The dsRNA agent also contains 5'-deoxy-5'-C-malonyl.

In one embodiment, 100%, 95%, 90%, 85%, 80%, 75%, 70%, 65%, 60%, 55%, 50%, 45%, 40%, 35%, or 30% of the agent are modified , which is a dsRNA according to the present invention. For example, if 50% of the dsRNA agent is modified, then 50% of all nucleotides present in the dsRNA agent have the modification described herein.

In one embodiment, the sense and antisense strands of the dsRNA agent are each independently modified with acyclic nucleotides, LNA, HNA, CeNA, 2'-methoxyethyl, 2'-O-methyl, 2'-O-allyl, 2' -C-allyl, 2'-deoxy, 2'-fluorine, 2'-O-Nmethylacetamido (2'-O-NMA), 2'-O-dimethylaminoethoxyethyl (2'-O-DMAEOE), 2'-O- aminopropyl (2'-OAP) or 2'-apa-F.

In one embodiment, each of the sense and antisense strands of the dsRNA agent has at least two different modifications.

In one embodiment, the dsRNA agent of formula (I) further comprises a 3' and/or 5' overhang(s) that is 1-10 nucleotides in length. In one example, a dsRNA agent of formula (I) has a 3' overhang at the 3' end of the antisense strand and a blunt end at the 5' end of the antisense strand. In another example, the dsRNA agent has a 5' overhang at the 5' end of the sense strand.

In one embodiment, the dsRNA agent of the present invention does not have any 2'-E modification.

In one embodiment, the sense strand and/or antisense strand of the dsRNA agent contains one or more of the blocks of phosphorothioate or methylphosphonate internucleotide bonds. According to one example, the sense strand contains one block of two phosphorothioate or methylphosphonate internucleotide bonds. In one example, the antisense strand contains two blocks of two phosphorothioate or methylphosphonate internucleotide bonds. For example, two blocks of phosphorothioate or methylphosphonate internucleotide bonds are separated by 16-18 phosphate internucleotide bonds.

In one embodiment, the sense and antisense strands of the dsRNA agent each contain 15-30 nucleotides. In one example, the sense strand contains 19-22 nucleotides and the antisense strand contains 19-25 nucleotides. According to another example, the sense strand contains 21 nucleotides and the antisense strand contains 23 nucleotides.

In one embodiment, the nucleotide at position 1 from the 5' end of the antisense strand in the duplex is selected from the group consisting of A, dA, dU, U, and dT. In one embodiment, at least one of the first, second, and third base pairs from the 5' end of the antisense strand is an AU base pair.

In one embodiment, the antisense strand of the dsRNA agent of the present invention is 100% complementary to the target RNA to which it hybridizes and suppresses its expression via RNA interference. In another embodiment, the antisense strand of the dsRNA agent of the present invention is at least 95%, at least 90%, at least 85%, at least 80%, at least 75%, at least 70%, at least 65%, at least 60%, at least 55%, or at least 50% complementary to the target RNA.

In one aspect, the present invention relates to a dsRNA agent, defined herein as being capable of repressing the expression of a target gene. Medium

- 12 043057 vo, which is a dsRNA, contains a sense strand and an antisense strand, with each strand containing from 14 to 40 nucleotides. The sense strand contains at least one thermally disturbing nucleotide, wherein at least one said thermally disturbing nucleotide occurs at or near the antisense strand seed site (i.e., at position 2-8 from the 5' end of the antisense threads). Each of the embodiments and aspects described herein with respect to a dsRNA represented by formula (I) can also be applied to a dsRNA containing a thermally disturbing nucleotide.

The thermostability-disrupting nucleotide may occur, for example, between positions 1417 from the 5' end of the sense strand if the sense strand is 21 nucleotides long. The antisense strand contains at least two modified nucleic acids that are less than the sterically demanding 2'-OMe modification. Preferably, the two modified nucleic acids that are smaller than the sterically 2'-OMe are separated by an 11 nucleotide long region. For example, the two modified nucleic acids are at positions 2 and 14 from the 5' end of the antisense strand.

In one embodiment, the dsRNA agent further comprises at least one ASGPR ligand. For example, an ASGPR ligand is one or more GalNAc derivatives attached via a divalent or trivalent branched linker, such as:

In one embodiment, the ASGPR ligand is attached to the 3-terminus of the sense strand.

For example, a dsRNA agent as defined herein may contain:

i) a phosphorus-containing group at the 5' end of the sense strand or antisense strand;

ii) two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18-23 antisense strand (counting from the 5' end of the antisense strand); and iii) a ligand, such as an ASGPR ligand (eg, one or more GalNAc derivatives) at the 5' or 3' end of the sense strand or antisense strand. For example, the ligand may be at the 3'end of the sense strand.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F equals 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 2, B3' is 2'-OMe or 2'-F, q ⁵ is 5, T3' is 2'-F, q ⁶ is 1, B4' is 2'-OMe and q ⁷ is 1; there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18- 23 of the antisense strand (counting from the 5' end of the antisense strand). The dsRNA agent also contains 5'-P and a targeting ligand. In one embodiment, the 5'-P is at the 5' end of the antisense strand and the targeting ligand is at the 3' end of the sense strand.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 2, B3' is 2'-OMe or 2'-F, q ⁵ is 5, T3' is 2'-F, q ⁶ is 1, B4' is 2'-OMe and q ⁷ is 1; there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18- 23 antisense strand (counting from the 5' end of the antisense strand). The dsRNA tool also contains 5'-PS and a targeting ligand. In one embodiment, the 5'-PS is at the 5' end of the antisense strand and the targeting ligand is at the 3' end of the sense strand.

- 13 043057

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 2, B3' is 2'-OMe or 2'-F, q ⁵ is 5, T3' is 2'-F, q ⁶ is 1, B4' is 2'-OMe and q ⁷ is 1; there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18- 23 of the antisense strand (counting from the 5' end of the antisense strand). The dsRNA agent also contains a 5'-VP (eg, 5'-E-VP, 5'-U-VP, or a combination thereof) and a targeting ligand. In one embodiment, the 5'-VP is at the 5' end of the antisense strand and the targeting ligand is at the 3' end of the sense strand.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 2, B3' is 2'-OMe or 2'-F, q ⁵ is 5, T3' is 2'-F, q ⁶ is 1, B4' is 2'-OMe and q ⁷ is 1; there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18- 23 of the antisense strand (counting from the 5' end of the antisense strand). The dsRNA tool also contains 5'-PS2 and a targeting ligand. In one embodiment, 5'-PS2 is at the 5' end of the antisense strand and the targeting ligand is at the 3' end of the sense strand.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 2, B3' is 2'-OMe or 2'-F, q ⁵ is 5, T3' is 2'-F, q ⁶ is 1, B4' is 2'-OMe and q ⁷ is 1; in this case, there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18- 23 of the antisense strand (counting from the 5' end of the antisense strand). The dsRNA agent also contains 5'-deoxy-5'-C-malonyl and a targeting ligand. In one embodiment, 5'-deoxy-5'-C-malonyl is at the 5' end of the antisense strand and the targeting ligand is at the 3' end of the sense strand.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, q ⁴ is 0, B3' is 2'OMe or 2'-F, q ⁵ is 7, T3' is 2'-F, q ⁶ is 1, B4' is 2'-OMe and q ⁷ is 1; there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18-23 of the antisense strands (counting from the 5' end). The dsRNA agent also contains 5'-P and a targeting ligand. In one embodiment, the 5'-P is at the 5' end of the antisense strand and the targeting ligand is at the 3' end of the sense strand.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, q ⁴ is 0, B3' is 2'-OMe or 2'-F, q ⁵ is 7, T3' is 2'-F, q ⁶ is 1, B4' is 2'-OMe and q ⁷ is 1; there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18-23 of the antisense strands (counting from the 5' end). The dsRNA tool also contains 5'-PS and a targeting ligand. In one embodiment, the 5'-PS is at the 5' end of the antisense strand and the targeting ligand is at the 3' end of the sense strand.

- 14 043057

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, q ⁴ is 0, B3' is 2'-OMe or 2'-F, q ⁵ is 7, T3' is 2'-F, q ⁶ is 1, B4' is 2'-OMe and q ⁷ is 1; there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18-23 of the antisense strands (counting from the 5' end). The dsRNA agent also contains a 5'-VP (eg, 5'-E-VP, 5'-U-VP, or a combination thereof) and a targeting ligand. In one embodiment, the 5'-VP is at the 5' end of the antisense strand and the targeting ligand is at the 3' end of the sense strand.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, q ⁴ is 0, B3' is 2'-OMe or 2'-F, q ⁵ is 7, T3' is 2'-F, q ⁶ is 1, B4' is 2'-OMe and q ⁷ is 1; there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18-23 of the antisense strands (counting from the 5' end). The dsRNA tool also contains 5'-PS2 and a targeting ligand. In one embodiment, 5'-PS2 is at the 5' end of the antisense strand and the targeting ligand is at the 3' end of the sense strand.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, q ⁴ is 0, B3' is 2'-OMe or 2'-F, q ⁵ is 7, T3' is 2'-F, q ⁶ is 1, B4' is 2'-OMe and q ⁷ is 1; there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18-23 of the antisense strands (counting from the 5' end). The dsRNA agent also contains 5'-deoxy-5'-C-malonyl and a targeting ligand. In one embodiment, 5'-deoxy-5'-C-malonyl is at the 5' end of the antisense strand and the targeting ligand is at the 3' end of the sense strand.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 2, B3' is 2'-OMe or 2'-F, q ⁵ is 5, T3' is 2'-F, q ⁶ is 1, B4' is 2'-F and q ⁷ is 1; in this case, there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18- 23 of the antisense strand (counting from the 5' end of the antisense strand). The dsRNA agent also contains 5'-P and a targeting ligand. In one embodiment, the 5'-P is at the 5' end of the antisense strand and the targeting ligand is at the 3' end of the sense strand.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 2, B3' is 2'-OMe or 2'-F, q ⁵ is 5, T3' is 2'-F, q ⁶ is 1, B4' is 2'-F and q ⁷ is 1; there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18- 23 of the antisense strand (counting from the 5' end of the antisense strand). The dsRNA tool also contains 5'-PS and a targeting ligand. In one embodiment, the 5'-PS is at the 5' end of the antisense strand and the targeting ligand is at the 3' end of the sense strand.

- 15 043057

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 2, B3' is 2'-OMe or 2'-F, q ⁵ is 5, T3' is 2'-F, q ⁶ is 1, B4' is 2'-F and q ⁷ is 1; in this case, there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18- 23 of the antisense strand (counting from the 5' end of the antisense strand). The dsRNA agent also contains a 5'-VP (eg, 5'-E-VP, 5'-U-VP, or a combination thereof) and a targeting ligand. In one embodiment, the 5'-VP is at the 5' end of the antisense strand and the targeting ligand is at the 3' end of the sense strand.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 2, B3' is 2'-OMe or 2'-F, q ⁵ is 5, T3' is 2'-F, q ⁶ is 1, B4' is 2'-F and q ⁷ is 1; in this case, there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18- 23 of the antisense strand (counting from the 5' end of the antisense strand). The dsRNA tool also contains 5'-PS2 and a targeting ligand. In one embodiment, 5'-PS2 is at the 5' end of the antisense strand and the targeting ligand is at the 3' end of the sense strand.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, T2' is 2'-F, q ⁴ is 2, B3' is 2'-OMe or 2'-F, q ⁵ is 5, T3' is 2'-F, q ⁶ is 1, B4' is 2'-F and q ⁷ is 1; in this case, there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18- 23 of the antisense strand (counting from the 5' end of the antisense strand). The dsRNA agent also contains 5'-deoxy-5'-C-malonyl and a targeting ligand. In one embodiment, 5'-deoxy-5'-C-malonyl is at the 5' end of the antisense strand and the targeting ligand is at the 3' end of the sense strand.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, q ⁴ is 0, B3' is 2'-OMe or 2'-F, q ⁵ is 7, T3' is 2'-F, q ⁶ is 1, B4' is 2'-F and q ⁷ is 1; in this case, there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18- 23 antisense strand (counting from the 5' end of the antisense strand). The dsRNA tool also contains 5'-P and a targeting ligand. In one embodiment, the 5'-P is at the 5' end of the antisense strand and the targeting ligand is at the 3' end of the sense strand.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, q ⁴ is 0, B3' is 2'-OMe or 2'-F, q ⁵ is 7, T3' is 2'-F, q ⁶ is 1, B4' is 2'-F and q ⁷ is 1; there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18- 23 of the antisense strand (counting from the 5' end of the antisense strand). The dsRNA tool also contains 5'-PS and a targeting ligand. In one embodiment, the 5'-PS is at the 5' end of the antisense strand and the targeting ligand is at the 3' end of the sense strand.

- 16 043057

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, q ⁴ is 0, B3' is 2'-OMe or 2'-F, q ⁵ is 7, T3' is 2'-F, q ⁶ is 1, B4' is 2'-F and q ⁷ is 1; in this case, there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18- 23 antisense strand (counting from the 5' end of the antisense strand). The dsRNA agent also contains a 5'-VP (eg, 5'-E-VP, 5'-Z-VP, or a combination thereof) and a targeting ligand. In one embodiment, the 5'-VP is at the 5' end of the antisense strand and the targeting ligand is at the 3' end of the sense strand.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, q ⁴ is 0, B3' is 2'-OMe or 2'-F, q ⁵ is 7, T3' is 2'-F, q ⁶ is 1, B4' is 2'-F and q ⁷ is 1; in this case, there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18- 23 of the antisense strand (counting from the 5' end of the antisense strand). The dsRNA tool also contains 5'-PS ₂ and a targeting ligand. In one embodiment, the 5'-PS ₂ is at the 5' end of the antisense strand and the targeting ligand is at the 3' end of the sense strand.

In one embodiment, B1 is 2'-OMe or 2'-F, n ¹ is 8, T1 is 2' F, n ² is 3, B2 is 2'-OMe, n ³ is 7, n ⁴ is 0, B3 is 2'-OMe, n ⁵ is 3, B1' is 2'-OMe or 2'-F, q ¹ is 9, T1' is 2'-F, q ² is 1, B2' is 2'-OMe or 2'-F, q ³ is 4, q ⁴ is 0, B3' is 2'-OMe or 2'-F, q ⁵ is 7, T3' is 2'-F, q ⁶ is 1, B4' is 2'-F and q ⁷ is 1; in this case, there are two phosphorothioate modifications of internucleotide bonds within positions 1-5 of the sense strand (counting from the 5'-end of the sense strand), as well as two phosphorothioate modifications of internucleotide bonds at positions 1 and 2, and two phosphorothioate modifications of internucleotide bonds within positions 18- 23 of the antisense strand (counting from the 5' end of the antisense strand). The dsRNA agent also contains 5'-deoxy-5'-C-malonyl and a targeting ligand. In one embodiment, 5'-deoxy5'-C-malonyl is at the 5' end of the antisense strand and the targeting ligand is at the 3' end of the sense strand.

In a specific embodiment, the dsRNA agents of the present invention comprise:

(a) a semantic thread characterized by:

(i) a length of 21 nucleotides;

(ii) an ASGPR ligand attached to the 3' end, wherein said ASGPR ligand contains three GalNAc derivatives attached via a trivalent branched linker; and (iii) 2'-T modifications at positions 1, 3, 5, 7, 9-11, 13, 17, 19 and 21, as well as 2'-OMe modifications at positions 2, 4, 6, 8, 12, 14-16, 18 and 20 (counting from the 5' end); and (b) an antisense strand characterized by:

(i) a length of 23 nucleotides;

(ii) 2'-OMe modifications at positions 1, 3, 5, 9, 11-13, 15, 17, 19, 21 and 23, and 2' F modifications at positions 2, 4, 6-8, 10, 14, 16, 18, 20 and 22 (counting from the 5' end); and (iii) phosphorothioate internucleotide bonds between nucleotide positions 21 and 22 and between nucleotide positions 22 and 23 (counting from the 5' end);

wherein the dsRNA agents have a two nucleotide overhang at the 3' end of the antisense strand and a blunt end at the 5' end of the antisense strand.

In another specific embodiment, the dsRNA agents of the present invention comprise:

(a) a thread of meaning characterized by:

(i) a length of 21 nucleotides;

(ii) an ASGPR ligand attached to the 3' end, wherein said ASGPR ligand contains three GalNAc derivatives attached via a trivalent branched linker;

(iii) 2'-B modifications at positions 1, 3, 5, 7, 9-11, 13, 15, 17, 19 and 21, as well as 2'-OMe modifications at positions 2, 4, 6, 8, 12 , 14, 16, 18 and 20 (counting from the 5' end); And

- 17 043057 (iv) phosphorothioate internucleotide bonds between nucleotide positions 1 and 2 and between nucleotide positions 2 and 3 (counting from the 5' end); and (b) an antisense strand characterized by:

(i) a length of 23 nucleotides;

(ii) 2'-OMe modifications at positions 1, 3, 5, 7, 9, 11-13, 15, 17, 19 and 21-23, and 2'-F modifications at positions 2, 4, 6, 8 , 10, 14, 16, 18 and 20 (counting from the 5' end); and (iii) phosphorothioate internucleotide bonds between nucleotide positions 1 and 2, between nucleotide positions 2 and 3, between nucleotide positions 21 and 22, and between nucleotide positions 22 and 23 (counting from the 5' end);

wherein the dsRNA agents have a two nucleotide overhang at the 3' end of the antisense strand and a blunt end at the 5' end of the antisense strand.

In another specific embodiment, the dsRNA agents of the present invention comprise:

(a) a thread of meaning characterized by:

(i) a length of 21 nucleotides;

(ii) an ASGPR ligand attached to the 3' end, wherein said ASGPR ligand contains three GalNAc derivatives attached via a trivalent branched linker;

(iii) 2'-OMe modifications at positions 1-6, 8, 10, and 12-21, 2'-F modifications at positions 7 and 9, and a deoxynucleotide (e.g., dT) at position 11 (counting from 5'- end); and (iv) phosphorothioate internucleotide bonds between nucleotide positions 1 and 2 and between nucleotide positions 2 and 3 (counting from the 5' end); and (b) an antisense strand characterized by:

(i) a length of 23 nucleotides;

(ii) 2'-OMe modifications at positions 1, 3, 7, 9, 11, 13, 15, 17 and 19-23, and 2'-F modifications at positions 2, 4-6, 8, 10, 12 , 14, 16 and 18 (counting from the 5' end); and (iii) phosphorothioate internucleotide bonds between nucleotide positions 1 and 2, between nucleotide positions 2 and 3, between nucleotide positions 21 and 22, and between nucleotide positions 22 and 23 (counting from the 5' end);

wherein the dsRNA agents have a two nucleotide overhang at the 3' end of the antisense strand and a blunt end at the 5' end of the antisense strand.

In another specific embodiment, the dsRNA agents of the present invention comprise:

(a) a semantic thread characterized by:

(i) a length of 21 nucleotides;

(ii) an ASGPR ligand attached to the 3' end, wherein said ASGPR ligand contains three GalNAc derivatives attached via a trivalent branched linker;

(iii) 2'-OMe modifications at positions 1-6, 8, 10, 12, 14 and 16-21, as well as 2'-F modifications at positions 7, 9, 11, 13 and 15; and (iv) phosphorothioate internucleotide bonds between nucleotide positions 1 and 2 and between nucleotide positions 2 and 3 (counting from the 5' end); and (b) an antisense strand characterized by:

(i) a length of 23 nucleotides;

(ii) 2'-OMe modifications at positions 1, 5, 7, 9, 11, 13, 15, 17, 19 and 21-23, and 2'-F modifications at positions 2-4, 6, 8, 10 , 12, 14, 16, 18 and 20 (counting from the 5' end); and (iii) phosphorothioate internucleotide bonds between nucleotide positions 1 and 2, between nucleotide positions 2 and 3, between nucleotide positions 21 and 22, and between nucleotide positions 22 and 23 (counting from the 5' end);

wherein the dsRNA agents have a two nucleotide overhang at the 3' end of the antisense strand and a blunt end at the 5' end of the antisense strand.

In another specific embodiment, the dsRNA agents of the present invention comprise:

(a) a semantic thread characterized by:

(i) a length of 21 nucleotides;

(ii) an ASGPR ligand attached to the 3' end, wherein said ASGPR ligand contains three GalNAc derivatives attached via a trivalent branched linker;

(iii) 2'-OMe modifications at positions 1-9 and 12-21, as well as 2'-B modifications at positions 10 and 11; and (iv) phosphorothioate internucleotide bonds between nucleotide positions 1 and 2 and between nucleotide positions 2 and 3 (counting from the 5' end); and (b) an antisense strand characterized by:

(i) a length of 23 nucleotides;

(ii) 2'-OMe modifications at positions 1, 3, 5, 7, 9, 11-13, 15, 17, 19 and 21-23, as well as 2'-F-

- 18 043057 modifications at positions 2, 4, 6, 8, 10, 14, 16, 18 and 20 (counting from the 5' end); and (iii) phosphorothioate internucleotide bonds between nucleotide positions 1 and 2, between nucleotide positions 2 and 3, between nucleotide positions 21 and 22, and between nucleotide positions 22 and 23 (counting from the 5' end);

wherein the dsRNA agents have a two nucleotide overhang at the 3' end of the antisense strand and a blunt end at the 5' end of the antisense strand.

In another specific embodiment, the dsRNA agents of the present invention comprise:

(a) a thread of meaning characterized by:

(i) a length of 21 nucleotides;

(ii) an ASGPR ligand attached to the 3' end, wherein said ASGPR ligand contains three GalNAc derivatives attached via a trivalent branched linker;

(iii) 2'-F modifications at positions 1, 3, 5, 7, 9-11 and 13, as well as 2'-OMe modifications at positions 2, 4, 6, 8, 12 and 14-21; and (iv) phosphorothioate internucleotide bonds between nucleotide positions 1 and 2 and between nucleotide positions 2 and 3 (counting from the 5' end); and (b) an antisense strand characterized by:

(i) a length of 23 nucleotides;

(ii) 2'-OMe modifications at positions 1, 3, 5-7, 9, 11-13, 15, 17-19 and 21-23, and 2'-F modifications at positions 2, 4, 8, 10 , 14, 16 and 20 (counting from the 5' end); and (iii) phosphorothioate internucleotide bonds between nucleotide positions 1 and 2, between nucleotide positions 2 and 3, between nucleotide positions 21 and 22, and between nucleotide positions 22 and 23 (counting from the 5' end);

wherein the dsRNA agents have a two nucleotide overhang at the 3' end of the antisense strand and a blunt end at the 5' end of the antisense strand.

In another specific embodiment, the dsRNA agents of the present invention comprise:

(a) a thread of meaning characterized by:

(i) a length of 21 nucleotides;

(ii) an ASGPR ligand attached to the 3' end, wherein said ASGPR ligand contains three GalNAc derivatives attached via a trivalent branched linker;

(iii) 2'-OMe modifications at positions 1, 2, 4, 6, 8, 12, 14, 15, 17 and 19-21, and 2'-F modifications at positions 3, 5, 7, 9-11 , 13, 16 and 18; and (iv) phosphorothioate internucleotide bonds between nucleotide positions 1 and 2 and between nucleotide positions 2 and 3 (counting from the 5' end); and (b) an antisense strand characterized by:

(i) a length of 25 nucleotides;

(ii) 2'-OMe modifications at positions 1, 4, 6, 7, 9, 11-13,

15, 17 and 19-23, 2'-F modifications at positions 2, 3, 5, 8, 10, 14, 16 and 18, as well as deoxynucleotides (for example, dT) at positions 24 and 25 (counting from 5' - end); and (iii) phosphorothioate internucleotide bonds between nucleotide positions 1 and 2, between nucleotide positions 2 and 3, between nucleotide positions 21 and 22, and between nucleotide positions 22 and 23 (counting from the 5' end);

wherein the dsRNA agents have a four nucleotide overhang at the 3' end of the antisense strand and a blunt end at the 5' end of the antisense strand.

In another specific embodiment, the dsRNA agents of the present invention comprise:

(a) a thread of meaning characterized by:

(i) a length of 21 nucleotides;

(ii) an ASGPR ligand attached to the 3' end, wherein said ASGPR ligand contains three GalNAc derivatives attached via a trivalent branched linker;

(iii) 2'-OMe modifications at positions 1-6, 8 and 12-21, as well as 2'-P modifications at positions 7 and 9-11; and (iv) phosphorothioate internucleotide bonds between nucleotide positions 1 and 2 and between nucleotide positions 2 and 3 (counting from the 5' end); and (b) an antisense strand characterized by:

(i) a length of 23 nucleotides;

(ii) 2'-OMe modifications at positions 1, 3-5, 7, 8, 10-13, 15 and 17-23, and 2'-P modifications at positions 2, 6, 9, 14 and 16 (counting from the 5' end); and (iii) phosphorothioate internucleotide bonds between nucleotide positions 1 and 2, between nucleotide positions 2 and 3, between nucleotide positions 21 and 22, and between nucleotide positions 22 and 23 (counting from the 5' end);

- 19 043057 wherein the dsRNA agents have a two nucleotide overhang at the 3' end of the antisense strand and a blunt end at the 5' end of the antisense strand.

In another specific embodiment, the dsRNA agents of the present invention comprise:

(a) a semantic thread characterized by:

(i) a length of 21 nucleotides;

(ii) an ASGPR ligand attached to the 3' end, wherein said ASGPR ligand contains three GalNAc derivatives attached via a trivalent branched linker;

(iii) 2'-OMe modifications at positions 1-6, 8 and 12-21, as well as 2'-P modifications at positions 7 and 9-11; and (iv) phosphorothioate internucleotide bonds between nucleotide positions 1 and 2 and between nucleotide positions 2 and 3 (counting from the 5' end); and (b) an antisense strand characterized by:

(i) a length of 23 nucleotides;

(ii) 2'-OMe modifications at positions 1, 3-5, 7, 10-13, 15 and 17-23, and 2'-F modifications at positions 2, 6, 8, 9, 14 and 16 (counting from the 5' end); and (iii) phosphorothioate internucleotide bonds between nucleotide positions 1 and 2, between nucleotide positions 2 and 3, between nucleotide positions 21 and 22, and between nucleotide positions 22 and 23 (counting from the 5' end);

wherein the dsRNA agents have a two nucleotide overhang at the 3' end of the antisense strand and a blunt end at the 5' end of the antisense strand.

In another specific embodiment, the dsRNA agents of the present invention comprise:

(a) a thread of meaning characterized by:

(i) a length of 19 nucleotides;

(ii) an ASGPR ligand attached to the 3' end, wherein said ASGPR ligand contains three GalNAc derivatives attached via a trivalent branched linker;

(iii) 2'-OMe modifications at positions 1-4, 6 and 10-19, as well as 2'-P modifications at positions 5 and 7-9; and (iv) phosphorothioate internucleotide bonds between nucleotide positions 1 and 2 and between nucleotide positions 2 and 3 (counting from the 5' end); and (b) an antisense strand having: (i) a length of 21 nucleotides;

(ii) 2'-OMe modifications at positions 1, 3-5, 7, 10-13, 15 and 17-21, and 2'-F modifications at positions 2, 6, 8, 9, 14 and 16 (counting from the 5' end); and (iii) phosphorothioate internucleotide bonds between nucleotide positions 1 and 2, between nucleotide positions 2 and 3, between nucleotide positions 19 and 20, and between nucleotide positions 20 and 21 (counting from the 5' end);

wherein the dsRNA agents have a two nucleotide overhang at the 3' end of the antisense strand and a blunt end at the 5' end of the antisense strand.

In one embodiment, the dsRNA agents described herein further comprise a thermally disturbing modification at position 7, counting from the 5' end of the antisense strand, at position 15, counting from the 5' end of the sense strand, at position 21, counting from the 5' end of the sense strand, or a combination of both.

In one aspect, the present invention relates to a dsRNA agent capable of repressing the expression of a target gene. The dsRNA tool contains a sense strand and an antisense strand, with each strand containing 14 to 40 nucleotides. The sense strand contains at least one thermally disturbing nucleotide, wherein at least one said thermally disturbing nucleotide occurs at or near the antisense strand seed site (i.e., at position 2-8 from the 5' end of the antisense threads). For example, a thermostability-disrupting nucleotide occurs between positions 14-17 from the 5-end of the sense strand if the sense strand is 21 nucleotides long. The antisense strand contains two modified nucleic acids that are smaller than the sterically demanding 2'-OMe modification, separated by an 11 nucleotide long region. For example, the two modified nucleic acids are at positions 2 and 14 from the 5' end of the antisense strand.

In one embodiment, the sense strand of the dsRNA agent further comprises a nucleotide with an endonuclease sensitive modification at the cleavage site of the sense strand. In one example, the nucleotide with the endonuclease sensitive modification is located at position 11 from the 5' end of the sense strand.

In one embodiment, the antisense strand further comprises a third nucleotide with a modification that gives the nucleotide a steric volume that is less than or equal to the steric volume of the 2'-OMe modification, wherein the third nucleotide with the modification is in

- 20 043057 positions 6-10 from the 5' end of the antisense strand. For example, the third nucleotide with the modification is at position 10 from the 5' end of the antisense strand.

The embodiments for thermostability-disrupting nucleotides are similar to the various embodiments described above for C1 in formula (I). Embodiments for modified nucleic acids that are less than the sterically demanding 2'-OMe modification are similar to the various embodiments described above for T1', T2' and T3' in formula (I). In this case, the embodiments describing the length, protruding ends, additional modifications and conjugations of ligands with dsRNA agents of formula I above are applicable.

The present invention further relates to the use of a dsRNA agent as defined herein to suppress the expression of a target gene. According to one embodiment, the present invention further relates to the use of a dsRNA agent to suppress the expression of a target gene in vitro.

The present invention further relates to the use of a dsRNA agent as defined herein to suppress expression of a target gene in a subject. The subject may be any animal, preferably a mammal, more preferably a mouse, rat, sheep, cattle, dog, cat, or human.

In one aspect, the present invention relates to a dsRNA agent capable of repressing the expression of a target gene. The dsRNA tool contains a sense strand and an antisense strand, with each strand containing 14 to 40 nucleotides. The sense strand contains a nucleotide with a modification conferring endonuclease sensitivity (eg, DNA, RNA, or 2'F) adjacent to the cleavage site of the sense strand. For example, a nucleotide with a modification conferring endonuclease sensitivity is at position 11 from the 5' end of the sense strand. An endonuclease sensitive modification that occurs adjacent to the cleavage site can affect the sensitivity of the cleavage site. For example, thermally disturbing modifications adjacent to the cleavage site may render the cleavage site endonuclease sensitive. The antisense strand contains two modified nucleic acids that are smaller than the sterically demanding 2'OMe modification, separated by an 11 nucleotide long region. For example, the two modified nucleic acids are at positions 2 and 14 from the 5' end of the antisense strand.

According to another aspect, the present invention further provides a method for delivering a dsRNA agent of the present invention to a specific target in a subject by subcutaneous or intravenous administration. The present invention further provides the dsRNA agents of the present invention for use in a method for delivering said agents to a specific target in a subject by subcutaneous or intravenous administration.

Brief description of graphic materials

In FIG. 1A-1C are graphs showing the effect of various modifications at position 17 of the sense strand on in vitro potency as assessed at 10 nM and 0.1 nM concentrations:

(A) mTTR-targeting siRNAs in which the non-F-modified sense strand is paired with the original AS strand;

(B) mTTR-targeting siRNAs in which the non-F-modified sense strand is paired with the non-F-modified AS strand;

(C) siRNAs targeting ANG, ApoC3, and TTRSC, in which the sense strand with the non-F modification is paired with the original AS strand.

In FIG. 2 is a graph showing the effect of position on in vitro potency for the thermostability-disrupting GNA modification at positions 16-18 of the sense strand, which was evaluated at concentrations of 10 nM and 0.1 nM.

In FIG. 3 is a graph showing the effect of modifications at position 2 of the antisense strand on in vitro potency of siRNAs targeting mTTR, ApoC3, TTRSC and TMP, which were evaluated at concentrations of 10 nM and 0.1 nM.

In FIG. 4 is a graph showing the effect of modifications at position 14 of the antisense strand on in vitro potency of siRNAs targeting mTTR, ApoC3, TTRSC, as assessed at concentrations of 10 nM and 0.1 nM.

In FIG. 5 is a graph showing mTTR silencing in mice after a single dose of 2.5 mg/kg SC.

In FIG. 6 is a graph showing the dose effect when using siRNA with modifications containing F, AD-61398 and AD-64273, compared with the original sequences with 2 PS (AD-43527) and 6 PS (AD-57727); a single dose SC administration was performed, and protein levels were measured 96 hours post-dose.

In FIG. 7 is a graph showing the decrease in plasma mTTR protein levels following QW SC administration of siRNA at 1 mg/kg in mice, comparing AD-61398 with the non-F modification with the original motif: AD-57727.

- 21 043057

In FIG. 8 is a graph showing TMPRSS6 mRNA silencing after single SC dose of 3 mg/kg in mice (n=3/group): comparison of non-F constructs with original motif: AD-60490.

In FIG. 9 is a graph showing TMPRSS6 mRNA silencing 7 days after SC administration of a single dose of 3 mg/kg in mice (n=3/group): comparison of non-F constructs with original motif: AD-60490.

In FIG. 10 shows in vitro activity results for two motifs, motif 1 and motif 2, with activity compared to parent compound AD-57727.

In FIG. 11 shows an in vivo evaluation of the silencing activity of siRNAs targeting mTTR.

In FIG. 12 shows enhanced activity with Stability Enhancer Conjugates (SEC-C) with liver samples assessed for activity (mRNA) on day 7 post-dose.

In FIG. 13 is a graph showing an approximately 4-fold increase in activity with the new motifs (motif 1 and 2) compared to the parent compound.

In FIG. 14 is a diagram showing the significantly increased duration of action due to motive 1 and motive 2 for all three sequences.

In FIG. 15 is a graph showing the results for ApoC3-GalNAc3 SAR with SC administration of a single dose of 3 mg/kg hAAV 1x1011 GC/mouse.

In FIG. 16 is a schematic representation of Ago2 with siRNA included, as well as 5'vinylphosphonate (5'-VP), a modified phosphate that mimics stable phosphate. The 5'-phosphonate is attached by cytosolic Clp1 kinase and acts as a key anchor for incorporation into Ago2.

In FIG. 17 is a diagram showing how the presence of 5'-VP generally increases in vivo activity. Evaluation was performed for four different sequences targeting ApoB. LDL levels 7 days after SC administration of a single dose of 3 mg/kg were analyzed for four conjugates (with or without 5'-VP modification).

In FIG. 18 depicts various chemical modifications that can replace the PS bond and provide more stable chemical structures, including phosphorodithioate (PS2) and methylphosphonate (MePhos), which activate endogenous phosphorylation.

In FIG. 19 shows an in vitro evaluation chart of end modifications, including 2'-OMe-MePhos, 2'-OMe-PS, dN(PS2), and 2'-F-PS. Transfection of primary mouse hepatocytes was performed using two ApoB-targeting conjugates at a concentration of 10 nM and 0.1 nM (n=4).

In FIG. 20 shows two graphs showing how a slight change at the 5' end of the antisense strand can significantly increase in vivo efficacy. The diagram on the left shows that 2'-F-PS at position 1 of the antisense strand can increase the activity of 5'-P-dependent sequences (with SC administered as a single dose of 3 mg/kg and F9 activity was measured on day 3). The chart on the right shows a ~3-fold increase in potency due to dN(PS)2 compared to the parent construct, similar to VP (with SC administered as a single dose of 10 mg/kg and LDL was measured on day 3 to determine ApoB) .

In FIG. 21A,B shows a SAR analysis comparing in vitro and in vivo activity of siRNAs targeting ApoB containing the 5'-OH- and 5'-.E-VP modification (at the 5' end of the antisense strand). In FIG. 21A shows the results of in vitro transfection of mouse 1° hepatocytes. In FIG. 21B shows LDL levels 3 days post single dose (SC dosing).

In FIG. 22 shows the results of an in vitro comparison of the efficacy of 5'-.E-VP modification and 5'-Z-VP modification in siRNA-GalNAc conjugates targeting mTTR and F9. The results were obtained by in vitro transfection of mouse primary hepatocytes.

In FIG. 23 shows the results of an in vivo comparison of the activity of the 5'-E-VP modification and the 5'-Z-VP modification of the F9-targeted siRNA-GalNAc conjugate (Single Dose SC).

In FIG. 24A-C are graphs showing the dose effect of PTEN-targeted siRNAs with (A) 5'-OH, (B) 5'-C-malonyl, and (C) 5'-phosphate on primary murine hepatocytes in an in vitro silencing assay. PTEN. All values were obtained as a result of experiments in triplicate.

In FIG. 25 shows enzyme resistance results for siRNAs with 5'-OH, 5'-C-malonyl, and 5'phosphate that were incubated with rat liver tritosomes. Target sequences for siRNA are shown in table. 10. Data normalized to untreated controls.

In FIG. 26 shows the results of incorporation into RISC siRNA with 5'-OH, 5'-C-malonyl and 5'-phosphate (5'-modification in antisense strands) as determined by Ago2 immunoprecipitation from primary mouse hepatocytes and by amplification with RT- Single strand PCR included in Ago2. Endogenous miR122 levels were determined as a control. Target sequences for siRNA are shown in table. 10.

In FIG. 27 is a graph showing in vitro TTR knockdown using single nucleotide (S)-GNA modified siRNA. TTR mRNA levels were measured in primary mouse hepatocytes after incubation with 10 nM siRNA for 24 h. TTR mRNA levels were assessed using RT-qPCR and normalized to PBS-treated cells. All results were the mean of four measurements.

In FIG. 28A is a graph showing in vitro TTR knockdown using siRNA modified with one (S)-GNA base pair. TTR mRNA levels were measured in primary mouse hepatocytes after incubation with 10 nM siRNA for 24 h. TTR mRNA levels were assessed using RT-qPCR and normalized to PBS-treated cells. All results were the mean of four measurements. In FIG. 28B shows combination duplexes in which sense and antisense strands containing (S)-GNA single nucleotides were paired as GNA:RNA heterologous base pairs.

In FIG. 29 is a graph showing in vivo serum TTR levels in mice. Animals received a single dose of siRNA of 2.5 mg/kg. At the indicated time points before and after dosing, animals were bled and serum samples were analyzed using a sandwich ELISA assay using an HRP-conjugated antibody and 3,3',5,5'tetramethylbenzidine for reading at 450 nm. All samples were analyzed in duplicate and each result was the mean of the mice in each cohort (n=3).

In FIG. 30 is a graph showing in vivo quantification of TTR mRNA levels. Animals received a single dose of siRNA of 2.5 mg/kg. At the indicated times after dosing, RNA extraction was performed from the liver homogenate. TTR mRNA was assessed as described above by RT-qPCR using the AACt method with GAPDH as control transcript and normalized to PBS-treated animals. The filled bars show the results for day 21, and the open bars show the results for day 7.

Detailed description

The present inventors have found that the presence of 2'-OMe modifications at nucleotide positions 2 and 14 from the 5' end of the antisense strand attenuates the gene silencing activity of the dsRNA agent. By introducing chemical modifications at the 2'-position or equivalent positions in the ribose-free nucleotide, acyclic nucleotide or nucleotide backbone that confer less steric bulk than the 2'-OMe modification at certain positions in the antisense and/or sense strand of the agent , which are dsRNAs, were able to restore gene silencing activity. The present inventors have also determined that the introduction of a thermally disturbing nucleotide in the sense strand at a site opposite the seeding site of the antisense strand (i.e., positions 2-8 from the 5' end of the antisense strand) provides enhanced gene silencing activity.

The sense strand and antisense strand of a dsRNA agent may be fully modified. The dsRNA agent is optionally conjugated to an asialoglycoprotein receptor (ASGPR) ligand, such as a sense strand. The resulting dsRNA tools are characterized by effective in vivo gene silencing activity.

Accordingly, the present invention provides a double-stranded RNAi agent (dsRNA) capable of suppressing the expression of a target gene. The dsRNA tool contains a sense strand and an antisense strand. The length of each strand of the dsRNA agent can vary from 12 to 40 nucleotides. For example, the length of each strand may be 14-40 nucleotides, 17-37 nucleotides, 25-37 nucleotides, 27-30 nucleotides, 17-23 nucleotides, 17-21 nucleotides, 17-19 nucleotides, 19-25 nucleotides, 19-23 nucleotide, 19-21 nucleotide, 21-25 nucleotide, or 21-23 nucleotide.

The sense strand and antisense strand typically form a duplex dsRNA. The length of the duplex region of a dsRNA agent may be 12-40 base pairs. For example, the length of the duplex region may be 14-40 bp, 17-30 bp, 2535 bp, 27-35 bp, 17-23 bp, 17-21 bp, 17-19 bp, 19-25 bp. bp, 19-23 bp, 19-21 bp, 21-25 bp, or 21-23 bp. In another example, the length of the duplex region is selected from 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26 and 27 bp.

In one embodiment, the dsRNA agent may contain one or more overhangs and/or capping groups at the 3' or 5' end or both ends of the strand in the dsRNA agent. The length of the protruding end can be 1-10 nucleotides, 1-6 nucleotides, for example, 2-6 nucleotides, 1-5 nucleotides, 2-5 nucleotides, 1-4 nucleotides, 2-4 nucleotides, 1-3 nucleotides, 2- 3 nucleotides or 1-2 nucleotides. The protruding ends may be due to the fact that one thread is longer than the other, or be due to the fact that two threads of the same length are offset. The protruding end may form a mismatch with the target mRNA, or it may be complementary to the target gene sequences, or it may have a different sequence. The first and second strands may also be connected, for example, with additional bases to form a hairpin, or with other base-free linkers.

- 23 043057

In one embodiment, each of the nucleotides in the overhang of the dsRNA agent can independently be a modified nucleotide, including but not limited to a 2' sugar modification, e.g., 2-F, 2'-O-methyl, thymidine (T), 2-Omethoxyethyl-5-methyluridine (Teo), 2'-O-methoxyethyladenosine (Aeo), 2'-O-methoxyethyl-5-methylcytidine (t5Ceo), and any combination thereof, or unmodified nucleotide. For example, the sequence of the protruding end at both ends of each thread may be TT. The protruding end may form a mismatch with the target mRNA, or it may be complementary to the target gene sequences, or it may have a different sequence.

The 5' or 3' overhang of the sense strand, the antisense strand, or both strands of the dsRNA agent may be phosphorylated. In some embodiments, the overhang contains two nucleotides with a phosphorothioate bond between the two nucleotides, the two nucleotides being the same or different. In one embodiment, the overhang is present at the 3' end of the sense strand, the antisense strand, or both. In one embodiment, this 3' overhang is present in the antisense strand. In one embodiment, this 3' overhang is present in the sense strand.

The dsRNA agent of the present invention may contain only one overhang, which can enhance the interference activity of the dsRNA without affecting its overall stability. For example, the single-stranded overhang may be located at the 3'terminal end of the sense strand, or alternatively at the 3'terminal end of the antisense strand. dsRNA can also have a blunt end located at the 5' end of the antisense strand (or 3' end of the sense strand), or vice versa. Typically, the dsRNA antisense strand has a nucleotide overhang at the 3' end and the 5' end is blunt. Without wishing to be bound by theory, it is believed that the asymmetric blunt end at the 5' end of the antisense strand and the 3' overhanging end of the antisense strand facilitate incorporation of the guide strand into the RISC process. For example, one overhang is at least two, three, four, five, six, seven, eight, nine, or ten nucleotides long.

In one embodiment, the dsRNA agent of the present invention may also have two blunt ends at both ends of the dsRNA duplex.

In one embodiment, the dsRNA agent of the present invention is a blunt-ended oligomer 19 nucleotides in length, wherein the sense strand contains at least one heat-disrupting nucleotide, with at least one heat-disrupting nucleotide occurring in a site opposite or adjacent to the seeding site of the antisense strand (ie, positions 2-8 from the 5' end of the antisense strand). For example, a thermostability-disrupting nucleotide occurs between positions 14-17 from the 5' end of the sense strand. The antisense strand contains at least two modified nucleic acids that are smaller than the sterically demanding 2'-OMe; preferably, the two modified nucleic acids that are smaller than the sterically 2'-OMe are at positions 2 and 14 from the 5' end of the antisense strand.

In one embodiment, the dsRNA agent of the present invention is a blunt-ended oligomer 20 nucleotides in length, wherein the sense strand contains at least one thermostability-disrupting nucleotide, with at least one thermo-disturbing nucleotide occurring in a site opposite or adjacent to the seeding site of the antisense strand (ie, positions 2-8 from the 5' end of the antisense strand). For example, a thermostability-disrupting nucleotide occurs between positions 14-17 from the 5' end of the sense strand. The antisense strand contains at least two modified nucleic acids that are smaller than the sterically demanding 2'-OMe; preferably, the two modified nucleic acids that are smaller than the sterically 2'-OMe are at positions 2 and 14 from the 5' end of the antisense strand.

In one embodiment, the dsRNA agent of the present invention is a blunt-ended oligomer 21 nucleotides in length, wherein the sense strand contains at least one thermostability-disrupting nucleotide, with at least one thermo-disturbing nucleotide occurring in a site opposite or adjacent to the seeding site of the antisense strand (ie, positions 2-8 from the 5' end of the antisense strand). For example, a thermostability-disrupting nucleotide occurs between positions 14-17 from the 5' end of the sense strand. The antisense strand contains at least two modified nucleic acids that are smaller than the sterically demanding 2'-OMe; preferably, the two modified nucleic acids that are smaller than the sterically 2'-OMe are at positions 2 and 14 from the 5' end of the antisense strand.

In one embodiment, the dsRNA agent of the present invention comprises a sense strand that is 21 nucleotides (nt) long and an antisense strand that is 23 nucleotides (nt) long, where the sense strand contains at least one

- 24 043057 thermostability-disrupting nucleotide, wherein at least one thermostability-disrupting nucleotide occurs at or near the site opposite the seed portion of the antisense strand (i.e., positions 2-8 from the 5' end of the antisense strand). For example, a thermostability-disrupting nucleotide occurs between positions 14-17 from the 5' end of the sense strand if the sense strand is 21 nucleotides long. The antisense strand contains at least two modified nucleic acids that are smaller than the sterically demanding 2'-OMe; preferably, the two modified nucleic acids that are smaller than the sterically 2'-OMe are at positions 2 and 14 from the 5' end of the antisense strand, where one end of the dsRNA is a blunt end while the other end contains a 2 nucleotide overhang. Preferably, the 2 nucleotide overhang is at the 3' end of the antisense strand. Optionally, the dsRNA further comprises a ligand (preferably a receptor ligand, ie an ASGPR ligand).

According to one embodiment, the dsRNA agent contains a sense and an antisense strand, where the length of the sense strand is 25-30 nucleotide residues, while counting from the 5'-terminal nucleotide (position 1), at positions 1-23 of the specified sense strand contains at least 8 ribonucleotides; the length of the antisense strand is 36-66 nucleotide residues and, counting from the 3'-terminal nucleotide, contains at least 8 ribonucleotides in positions paired with positions 1-23 of the sense strand to form a duplex; where at least the 3'-nucleotide of the antisense strand is unpaired with the sense strand, and up to 6 consecutive 3'-terminal nucleotides are unpaired with the sense strand, thereby forming a 3'-single-strand overhang of 1-6 nucleotides; where the 5' end of the antisense strand contains 10-30 consecutive nucleotides unpaired with the sense strand, thereby forming a 5' single-stranded overhang of 10-30 nucleotides; where at least the 5'-terminal and 3'-terminal nucleotides of the sense strand form base pairs with nucleotides of the antisense strand, if the sense and antisense strands are aligned to ensure maximum complementarity, thereby forming a substantially duplex region between the sense and antisense strands; and the antisense strand is sufficiently complementary to the target RNA for at least 19 ribonucleotides along the length of the antisense strand to reduce expression of the target gene when said double-stranded nucleic acid is introduced into a mammalian cell; and wherein the sense strand contains at least one heat-disrupting nucleotide, wherein the at least one heat-disrupting nucleotide occurs at a site opposite the seed portion of the antisense strand (i.e., positions 2-8 from the 5' end of the antisense strand). or next to it. For example, a thermostability-disrupting nucleotide occurs between positions 14-17 from the 5' end of the sense strand. The antisense strand contains at least two modified nucleic acids that are smaller than the sterically demanding 2'-OMe; preferably, the two modified nucleic acids that are smaller than the sterically 2'-OMe are at positions 2 and 14 from the 5' end of the antisense strand.

In one embodiment, the dsRNA agent of the present invention comprises a sense and an antisense strand, wherein said dsRNA agent comprises a sense strand that is at least 25 and not more than 29 nucleotides long and an antisense strand that is is not more than 30 nucleotides, while the semantic strand contains a nucleotide with a modification that gives sensitivity to enzyme cleavage at position 11 from the 5'-end. The antisense strand contains two modified nucleic acids that are smaller than the sterically demanding 2'-OMe located at positions 2 and 14 from the 5' end of the antisense strand; where the specified 3'-end of the specified sense strand and the specified 5'-end of the specified antisense strand form a blunt end, while the specified antisense strand at its 3'-end is 1-4 nucleotides longer than the sense strand, where the length of the duplex region is at least 25 nucleotides, and said antisense strand is sufficiently complementary to the target mRNA for at least 19 nucleotides along the length of said antisense strand to reduce the expression of the target gene when the dsRNA agent is introduced into a mammalian cell, and where cleavage of said dsRNA using Dicer preferably leads to siRNA containing the specified 3'-end of the specified antisense, thereby reducing the expression of the target gene in the mammal. The dsRNA agent optionally additionally contains a ligand.

In one embodiment, the sense strand contains a nucleotide with a modification conferring sensitivity to enzyme digestion at position 11 from the 5' end. The antisense strand contains two modified nucleic acids that are smaller than the sterically demanding 2'-OMe located at positions 2 and 14 from the 5' end of the antisense strand.

In one embodiment, the antisense strand contains two modified nucleic acids that are smaller than the sterically demanding 2'-OMe located at positions 2 and 14 from the 5' end of the antisense strand.

In one embodiment, each nucleotide in the sense strand and antisense strand of the dsRNA agent can be modified. Each nucleotide can

- 25 043057 be modified with the same or different modifications, which may include one or more changes to one or both non-bonding phosphate oxygens and/or one or more bonding phosphate oxygens; changing the constituent element of the ribose sugar, for example, 2-hydroxyl on the ribose sugar; complete replacement of the phosphate moiety with dephosphorized linkers; modifying or replacing a naturally occurring base; and replacing or modifying a ribose phosphate backbone.

Because nucleic acids are polymers of subunits, many modifications occur at a position that is repeated within the nucleic acid, such as modification of a base, or a phosphate moiety, or a non-bonding O atom of a phosphate moiety. In some cases, the modification will occur at all of the considered positions in the nucleic acid, but in many cases it will not. By way of example, the modification may occur only at the 3' or 5' end position, may occur only at the end, for example at the terminal nucleotide position, or at the last 2, 3, 4, 5 or 10 nucleotides of the strand. The modification may occur in the double strand section, in the single strand section, or both. The modification may occur only in a double-stranded RNA region or may occur only in a single-stranded RNA region. For example, the phosphorothioate modification at the position of the non-bonding O atom may occur only at one or both ends, may occur only at the terminal site, for example, at the terminal nucleotide position, or at the last 2, 3, 4, 5, or 10 nucleotides of the strand, or may occur in double-strand and single-strand sections, in particular, at the ends. The 5' end or ends may be phosphorylated.

This makes it possible, for example, to improve stability, to include specific bases in overhangs, or to include nucleotides with modifications or nucleotide mimics in single-stranded overhangs, for example, in the 5' or 3' overhang, or both. For example, it may be desirable to include purine nucleotides in the overhangs. In some embodiments, all or some of the bases at the 3' or 5' overhang may be modified, for example, using the modification described herein. Modifications may include, for example, the use of modifications at the 2'-position of the ribose sugar using modifications known in the art, for example, the use of deoxyribonucleotides, 2'deoxy-2'-fluoro-(2'-F)- or 2'-O -methyl modifications instead of the ribose sugar of the nucleic base, as well as modifications of the phosphate group, for example, phosphorothioate modifications. The overhangs need not be homologous to the target sequence.

In one embodiment, each residue of the sense strand and antisense strand is independently modified with LNA, HNA, CeNA, 2'-methoxyethyl, 2'-O-methyl, 2'-O-α-lyl, 2'C-allyl, 2'-deoxy or 2'-fluorine. Threads can contain more than one modification. In one embodiment, each residue of the sense strand and antisense strand is independently modified with 2'-O-methyl or 2'-fluoro.

At least two different modifications are typically present in the sense strand and the antisense strand. These two modifications may be 2'-deoxy, 2'-O-methyl or 2'-fluoro modifications, acyclic nucleotides, or others.

In one embodiment, each sense strand and antisense strand contains two nucleotides with various modifications selected from 2'-O-methyl or 2'-deoxy.

In one embodiment, each residue in the sense strand and antisense strand is independently modified with a 2'-O-methylnucleotide, 2'-deoxynucleotide, 2'deoxyfluoronucleotide, 2'-O-N-methylacetamido (2'-O-NMA) nucleotide, 2 '-O-dimethylaminoethoxyethyl (2'-O-DMAEOE) nucleotide, 2'-O-aminopropyl (2'-O-AP) nucleotide or 2'-apa-F-nucleotide.

In one embodiment, the dsRNA agent of the present invention contains modifications in an alternating pattern, particularly in the regions B1, B2, B3, B1', B2', B3', B4' shown in Formula I. The term alternating a motif or intermittent pattern as used herein refers to a motif with one or more modifications, each modification occurring in alternating nucleotides on the same strand. An alternating nucleotide may refer to every second nucleotide, or every third nucleotide, or a similar pattern. For example, if each of A, B, and C represents the same type of nucleotide modification, then the alternating motif could be

AAAAAAAAAAAV..., AAAVVVAAAAVVV... or ABSABCSAVS... etc.

The type of modifications contained in the alternating motif may be the same or different. For example, if each of A, B, C, D represents one type of nucleotide modification, then the alternating pattern, ie. modifications of every second nucleotide may be the same, but each sense strand or antisense strand may be selected from several possible modifications within an alternating motif,

ABABAB..., ACASAS..., BDBDBD... or CDCDCD... etc.

such as,

- 26 043057

In one embodiment, the dsRNA agent of the present invention is characterized by a pattern of alternating motif modifications on the sense strand that is shifted relative to the alternating motif modification pattern on the antisense strand. The shift may be such that the modified nucleotide group of the sense strand corresponds to the otherwise modified nucleotide group of the antisense strand, and vice versa. For example, when a sense strand is paired with an antisense strand in a dsRNA duplex, an alternating motif in the sense strand may begin with ABABAB in the 5' to 3' direction of the strand, and an alternating motif in the antisense strand may begin with BABABA in the 3' direction. - to the 5' end of the strand within the duplex region. As another example, an alternating motif in a sense strand may begin with aavbaavb in the 5' to 3' direction of the strand, and an alternating motif in an antisense strand may begin with vbaavbaa in the 3' to 5' direction of the strand. within the duplex region, so that between the sense strand and the antisense strand there is a complete or partial shift in modification patterns.

The dsRNA agent of the present invention may further comprise at least one phosphorothioate or methylphosphonate internucleotide bond. The phosphorothioate or methylphosphonate modification of the internucleotide bond may occur at any nucleotide of the sense strand or the antisense strand, or both strands, at any position on the strand. For example, an internucleotide bond modification may occur at every nucleotide in the sense strand and/or antisense strand; wherein each modification of the internucleotide bond may occur as an alternating pattern in the sense strand or antisense strand; or the sense strand or antisense strand may have both internucleotide bond modifications in an alternating pattern. The alternating pattern of internucleotide bond modifications in the sense strand may be the same as or different from the antisense strand, and the alternating pattern of internucleotide bond modifications in the sense strand may be shifted relative to the alternating pattern of internucleotide bond modifications in the antisense strand.

In one embodiment, the dsRNA agent contains a phosphorothioate or methylphosphonate modification of the internucleotide bond in the overhang. For example, the overhang contains two nucleotides with a phosphorothioate or methylphosphonate internucleotide bond between the two nucleotides. Internucleotide bond modifications can also be made to connect overhangs to terminal paired nucleotides within a duplex region. For example, at least 2, 3, 4, or all of the overhang nucleotides may be linked via a phosphorothioate or methylphosphonate internucleotide bond, and optionally, additional phosphorothioate or methylphosphonate internucleotide bonds may be present connecting the overhang nucleotide to the paired nucleotide that follows the overhang. For example, there may be at least two phosphorothioate internucleotide bonds between the three terminal nucleotides, whereby two of the three nucleotides are overhang nucleotides and the third is a paired nucleotide following the overhang. Preferably, these three terminal nucleotides may be at the 3' end of the antisense strand.

In one embodiment, the sense strand of the dsRNA agent contains 1-10 blocks of two to ten phosphorothioate or methylphosphonate internucleotide bonds separated by 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 or 16 phosphate internucleotide bonds, where one of the phosphorothioate or methylphosphonate internucleotide bonds is located at any position in the oligonucleotide sequence, and the specified sense strand is paired with an antisense strand containing any combination of phosphorothioate, methylphosphonate and phosphate internucleotide bonds, or with an antisense strand containing either a phosphorothioate, or a methylphosphonate, or a phosphate bond.

In one embodiment, the antisense strand of the dsRNA agent contains two blocks of two phosphorothioate or methylphosphonate internucleotide bonds separated by 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17 or 18 phosphate internucleotide bonds, where one of the phosphorothioate or methylphosphonate internucleotide bonds is located at any position in the oligonucleotide sequence, and the specified antisense strand is paired with a sense strand containing any combination of phosphorothioate, methylphosphonate and phosphate internucleotide bonds, or with an antisense strand containing either a phosphorothioate, or a methylphosphonate, or a phosphate bond.

In one embodiment, the antisense strand of the dsRNA agent contains two blocks of three phosphorothioate or methylphosphonate internucleotide bonds separated by 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 or 16 phosphate internucleotide bonds, where one of the phosphorothioate or methylphosphonate internucleotide bonds is located at any position in the oligonucleotide sequence, and the specified antisense strand is paired with a sense strand containing any combination of phosphorothioate, methylphosphonate and phosphate internucleotide bonds, or with an antisense a thread containing either a phosphorothioate, or a methylphosphonate, or a phosphate bond.

- 27 043057

In one embodiment, the antisense strand of the dsRNA agent contains two blocks of four phosphorothioate or methylphosphonate internucleotide bonds separated by 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, or 14 phosphate internucleotide bonds, where one of the phosphorothioate or methylphosphonate internucleotide bonds is located at any position in the oligonucleotide sequence, and the specified antisense strand is paired with a sense strand containing any combination of phosphorothioate, methylphosphonate and phosphate internucleotide bonds, or with an antisense strand, containing either a phosphorothioate, or a methylphosphonate, or a phosphate bond.

In one embodiment, the antisense strand of the dsRNA agent contains two blocks of five phosphorothioate or methylphosphonate internucleotide bonds separated by 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, or 12 phosphate internucleotide bonds , where one of the phosphorothioate or methylphosphonate internucleotide bonds is located at any position in the oligonucleotide sequence, and the specified antisense strand is paired with a sense strand containing any combination of phosphorothioate, methylphosphonate and phosphate internucleotide bonds, or with an antisense strand containing or phosphorothioate, or a methylphosphonate or phosphate bond.

In one embodiment, the antisense strand of the dsRNA agent contains two blocks of six phosphorothioate or methylphosphonate internucleotide bonds separated by 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 phosphate internucleotide bonds, where one of phosphorothioate or methylphosphonate internucleotide bonds is placed at any position in the oligonucleotide sequence, and said antisense strand is paired with a sense strand containing any combination of phosphorothioate, methylphosphonate and phosphate internucleotide bonds, or with an antisense strand containing either phosphorothioate or methylphosphonate, or phosphate bond.

In one embodiment, the antisense strand of the dsRNA agent contains two blocks of seven phosphorothioate or methylphosphonate internucleotide bonds separated by 1, 2, 3, 4, 5, 6, 7, or 8 phosphate internucleotide bonds, where one of the phosphorothioate or methylphosphonate internucleotide bonds bonds is placed at a position in the oligonucleotide sequence, and said antisense strand is paired with a sense strand containing any combination of phosphorothioate, methylphosphonate, and phosphate internucleotide bonds, or with an antisense strand containing either a phosphorothioate, or methylphosphonate, or phosphate bond.

In one embodiment, the antisense strand of the dsRNA agent contains two blocks of eight phosphorothioate or methylphosphonate internucleotide bonds separated by 1, 2, 3, 4, 5, or 6 phosphate internucleotide bonds, where one of the phosphorothioate or methylphosphonate internucleotide bonds is located in which - any position in the oligonucleotide sequence, and the specified antisense strand is paired with a sense strand containing any combination of phosphorothioate, methylphosphonate and phosphate internucleotide bonds, or with an antisense strand containing either a phosphorothioate, or methylphosphonate or phosphate bond.

In one embodiment, the antisense strand of the dsRNA agent contains two blocks of nine phosphorothioate or methylphosphonate internucleotide bonds separated by 1, 2, 3, or 4 phosphate internucleotide bonds, where one of the phosphorothioate or methylphosphonate internucleotide bonds is placed at any position in oligonucleotide sequence, and said antisense strand is paired with a sense strand containing a combination of phosphorothioate, methylphosphonate, and phosphate internucleotide bonds, or with an antisense strand containing either a phosphorothioate, or methylphosphonate, or phosphate bond.

In one embodiment, the dsRNA agent of the present invention further comprises one or more phosphorothioate or methylphosphonate modifications of the internucleotide bond within the 1-10 terminal positions of the sense and/or antisense strand. For example, at least 2, 3, 4, 5, 6, 7, 8, 9, or 10 nucleotides may be linked via a phosphorothioate or methylphosphonate internucleotide bond at one or both ends of the sense and/or antisense strand.

In one embodiment, the dsRNA agent of the present invention further comprises one or more phosphorothioate or methylphosphonate modifications of the internucleotide bond within 1-10 residues of the internal duplex portion of each of the sense and/or antisense strands. For example, at least 2, 3, 4, 5, 6, 7, 8, 9, or 10 nucleotides may be linked via a phosphorothioate or methylphosphonate internucleotide bond at position 8-16 of the duplex region, counting from the 5' end of the sense strand; wherein the dsRNA agent may optionally additionally contain one or more phosphorothioate or methylphosphonate modifications of the internucleotide bond within 1-10 terminal positions.

- 28 043057

In one embodiment, the dsRNA agent of the present invention further comprises one to five internucleotide bond phosphorothioate or methylphosphonate modifications within positions 1-5 and one to five internucleotide bond phosphorothioate or methylphosphonate modifications within positions 18-23 of the sense strand (counting from the 5'-end), as well as from one to five phosphorothioate or methylphosphonate modifications of the internucleotide bond at positions 1 and 2 and from one to five within positions 18-23 of the antisense strand (counting from the 5'-end).

In one embodiment, the dsRNA agent of the present invention further comprises one internucleotide bond phosphorothioate modification within positions 1-5 and one internucleotide bond phosphorothioate or methylphosphonate modification within positions 18-23 of the sense strand (counting from the 5' end) , as well as one phosphorothioate modification of the internucleotide bond at positions 1 and 2 and two phosphorothioate or methylphosphonate modifications of the internucleotide bond within positions 18-23 of the antisense strand (counting from the 5' end).

In one embodiment, the dsRNA agent of the present invention further comprises two internucleotide bond phosphorothioate modifications within positions 1-5 and one internucleotide bond phosphorothioate modification within positions 18-23 of the sense strand (counting from the 5' end), and also one phosphorothioate modification of the internucleotide bond at positions 1 and 2 and two phosphorothioate modifications of the internucleotide bond within positions 18-23 of the antisense strand (counting from the 5' end).

In one embodiment, the dsRNA agent of the present invention further comprises two internucleotide bond phosphorothioate modifications within positions 1-5 and two internucleotide bond phosphorothioate modifications within positions 18-23 of the sense strand (counting from the 5' end), and also one phosphorothioate modification of the internucleotide bond at positions 1 and 2 and two phosphorothioate modifications of the internucleotide bond within positions 18-23 of the antisense strand (counting from the 5' end).

In one embodiment, the dsRNA agent of the present invention further comprises two internucleotide bond phosphorothioate modifications within positions 1-5 and two internucleotide bond phosphorothioate modifications within positions 18-23 of the sense strand (counting from the 5' end), and also one phosphorothioate modification of the internucleotide bond at positions 1 and 2 and one phosphorothioate modification of the internucleotide bond within positions 18-23 of the antisense strand (counting from the 5' end).

In one embodiment, the dsRNA agent of the present invention further comprises one internucleotide bond phosphorothioate modification within positions 1-5 and one internucleotide bond phosphorothioate modification within positions 18-23 of the sense strand (counting from the 5' end), and also two phosphorothioate modifications of the internucleotide bond at positions 1 and 2 and two phosphorothioate modifications of the internucleotide bond within positions 18-23 of the antisense strand (counting from the 5' end).

In one embodiment, the dsRNA agent of the present invention further comprises one phosphorothioate modification of the internucleotide bond within positions 1-5 and one within positions 18-23 of the sense strand (counting from the 5' end), as well as two phosphorothioate modifications internucleotide bond at positions 1 and 2; and one phosphorothioate modification of the internucleotide bond within positions 18-23 of the antisense strand (counting from the 5' end).

In one embodiment, the dsRNA agent of the present invention further comprises one phosphorothioate modification of the internucleotide bond within positions 1-5 (counting from the 5' end) of the sense strand, as well as two phosphorothioate modifications of the internucleotide bond at positions 1 and 2, and one phosphorothioate modification of the internucleotide bond within positions 18-23 of the antisense strand (counting from the 5' end).

In one embodiment, the dsRNA agent of the present invention further comprises two phosphorothioate modifications of the internucleotide bond within positions 1-5 (counting from the 5' end) of the sense strand, as well as one phosphorothioate modification of the internucleotide bond at positions 1 and 2, and two phosphorothioate modifications of the internucleotide bond within positions 18-23 of the antisense strand (counting from the 5' end).

According to one embodiment, the dsRNA agent of the present invention additionally contains two phosphorothioate modifications of the internucleotide bond within positions 1-5 and one within positions 18-23 of the sense strand (counting from the 5' end), as well as two phosphorothioate modifications internucleotide bond at positions 1 and 2; and one phosphorothioate modification of the internucleotide bond within positions 18-23 of the antisense strand (counting from the 5' end).

In one embodiment, the dsRNA agent of the present invention further comprises two phosphorothioate modifications of the internucleotide bond in

- 29 043057 within positions 1-5 and one phosphorothioate modification of the internucleotide bond within positions 18-23 of the sense strand (counting from the 5'-end), as well as two phosphorothioate modifications of the internucleotide bond at positions 1 and 2 and two phosphorothioate modifications of the internucleotide bond at within positions 18-23 of the antisense strand (counting from the 5' end).

In one embodiment, the dsRNA agent of the present invention further comprises two internucleotide bond phosphorothioate modifications within positions 1-5 and one internucleotide bond phosphorothioate modification within positions 18-23 of the sense strand (counting from the 5' end), and also one phosphorothioate modification of the internucleotide bond at positions 1 and 2 and two phosphorothioate modifications of the internucleotide bond within positions 18-23 of the antisense strand (counting from the 5' end).

According to one embodiment, the dsRNA agent of the present invention further comprises two phosphorothioate modifications of the internucleotide bond at positions 1 and 2 and two phosphorothioate modifications of the internucleotide bond at positions 20 and 21 of the sense strand (counting from the 5' end), as well as one a phosphorothioate modification of the internucleotide bond at position 1 and one at position 21 of the antisense strand (counting from the 5' end).

In one embodiment, the dsRNA agent of the present invention further comprises one phosphorothioate modification of the internucleotide bond at position 1 and one phosphorothioate modification of the internucleotide bond at position 21 of the sense strand (counting from the 5' end), as well as two phosphorothioate modifications of the internucleotide bond at positions 1 and 2 and two phosphorothioate modifications of the internucleotide bond at positions 20 and 21 of the antisense strand (counting from the 5' end).

According to one embodiment, the dsRNA agent of the present invention additionally contains two phosphorothioate modifications of the internucleotide bond at positions 1 and 2 and two phosphorothioate modifications of the internucleotide bond at positions 21 and 22 of the sense strand (counting from the 5' end), as well as one a phosphorothioate modification of the internucleotide bond at position 1; and one phosphorothioate modification of the internucleotide bond at position 21 of the antisense strand (counting from the 5' end).

In one embodiment, the dsRNA agent of the present invention further comprises one phosphorothioate modification of the internucleotide bond at position 1 and one phosphorothioate modification of the internucleotide bond at position 21 of the sense strand (counting from the 5' end), as well as two phosphorothioate modifications of the internucleotide bond at positions 1 and 2 and two phosphorothioate modifications of the internucleotide bond at positions 21 and 22 of the antisense strand (counting from the 5' end).

In one embodiment, the dsRNA agent of the present invention further comprises two phosphorothioate modifications of the internucleotide bond at positions 1 and 2 and two phosphorothioate modifications of the internucleotide bond at positions 22 and 23 of the sense strand (counting from the 5' end), as well as one a phosphorothioate modification of the internucleotide bond at position 1; and one phosphorothioate modification of the internucleotide bond at position 21 of the antisense strand (counting from the 5' end).

In one embodiment, the dsRNA agent of the present invention further comprises one phosphorothioate modification of the internucleotide bond at position 1 and one phosphorothioate modification of the internucleotide bond at position 21 of the sense strand (counting from the 5' end), as well as two phosphorothioate modifications of the internucleotide bond at positions 1 and 2 and two phosphorothioate modifications of the internucleotide bond at positions 23 and 23 of the antisense strand (counting from the 5' end).

In one embodiment, the dsRNA agent of the present invention comprises mismatch(es) with the target, within duplexes, or combinations thereof. Mismatching may occur in the overhang or duplex region. Base pairs can be ranked based on their ability to promote dissociation or melting (e.g., based on the free energy of association or dissociation for a particular pairing, with the simplest approach being to examine the pairs individually for each pair, although nearest neighbor analysis or the like can also be used) . In terms of promoting dissociation: A:U is more preferable than G:C; G:U is preferred over G:C; a T:C is more preferred than G:C (1=inosine). Mismatches, such as non-canonical or non-canonical match types (described elsewhere in this document), are preferred over canonical match types (A:T, A:U, G:C); and pairing types that include a universal base are preferred over canonical pairing types.

In one embodiment, the dsRNA agent of the present invention comprises at least one of the first 1, 2, 3, 4, or 5 base pairs within the duplex regions, in the direction from the 5' end of the antisense strand, which can independently be selected from the group consisting of A:U, G:U, I:C, and mismatches, e.g., non-canonical or non-canonical match types, or match types that include a universal base, to promote dissociation of the antisense strand at the 5' end duplex.

- 30 043057

In one embodiment, the nucleotide at position 1 within the duplex region, in the direction from the 5' end of the antisense strand, is selected from the group consisting of A, dA, dU, U, and dT. Alternatively, at least one of the first 1, 2, or 3 base pairs within the duplex region, in the direction from the 5' end of the antisense strand, is an AU base pair. For example, the first base pair within the duplex region, away from the 5' end of the antisense strand, is the AU base pair.

The authors of the present invention have found that the introduction of a nucleotide with a 4'-modification and/or 5'modification at the 3'-end of the phosphodiester (PO), phosphorothioate (PS) and/or phosphorothioate (PS ₂ ) bond of a dinucleotide at any position of a single-stranded or double-stranded oligonucleotide can have a steric effect on the internucleotide bond and, therefore, protect it from the action of nucleases or make it resistant to them.

In one embodiment, the 5'-modified nucleoside is inserted at the 3' end of the dinucleotide at any position in the single-stranded or double-stranded siRNA. For example, a 5'-alkylated nucleoside is inserted at the 3' end of a dinucleotide at any position in a single-stranded or double-stranded siRNA. The alkyl group at the 5' position of the ribose sugar may be racemic or chirally pure R or S isomer. An exemplary 5'-alkylated nucleoside is a 5'-methyl nucleoside. The 5'-methyl may be either racemic or chirally pure R- or S-isomer.

In one embodiment, the 4'-modified nucleoside is inserted at the 3' end of the dinucleotide at any position in the single-stranded or double-stranded siRNA. For example, a 4'-alkylated nucleoside is introduced at the 3' end of a dinucleotide at any position in a single-stranded or double-stranded siRNA. The alkyl group at the 4' position of the ribose sugar may be racemic or chirally pure R or S isomer. An exemplary 4'-alkylated nucleoside is a 4'-methyl nucleoside. The 4'-methyl may be either racemic or chirally pure R- or S-isomer. Alternatively, the 4'-O-alkylated nucleoside can be inserted at the 3' end of the dinucleotide at any position in the single-stranded or double-stranded siRNA. The 4'-O-alkyl ribose sugar may be racemic or chirally pure R- or S-isomer. An exemplary 4'-O-alkylated nucleoside is a 4'-O-methyl nucleoside. 4'-O-methyl may be either racemic or chirally pure R- or S-isomer.

In one embodiment, a 5'-alkylated nucleoside is introduced at a position on the sense strand or antisense strand of the dsRNA, and such modification maintains or increases the potency of the dsRNA. The 5'-alkyl may be either racemic or chirally pure R- or S-isomer. An exemplary 5'-alkylated nucleoside is a 5'-methyl nucleoside. The 5'methyl may be either racemic or chirally pure R- or S-isomer.

In one embodiment, a 4'-alkylated nucleoside is introduced at a position on the sense strand or antisense strand of the dsRNA, and such modification maintains or increases the potency of the dsRNA. The 4'-alkyl may be either racemic or chirally pure R- or S-isomer. An exemplary 4'-alkylated nucleoside is a 4'-methyl nucleoside. 4'methyl can be either racemic or chirally pure R- or S-isomer.

In one embodiment, a 4'-O-alkylated nucleoside is introduced at a position on the sense strand or antisense strand of the dsRNA, and such modification maintains or increases the potency of the dsRNA. The 5'-alkyl may be either racemic or chirally pure R- or S-isomer. An exemplary 4'-O-alkylated nucleoside is a 4'-O-methyl nucleoside. 4'-O-methyl may be either racemic or chirally pure R- or S-isomer.

In one embodiment, the dsRNA sense strand sequence is represented by formula (Is):

where each of B1, B2, B3 is independently a nucleotide having a modification selected from the group consisting of 2'-O-alkyl, 2'-substituted alkoxy, 2'-substituted alkyl, 2'-halogen, ENA and BNA/ LNA;

C1 is a thermostability-disrupting nucleotide (e.g., an acyclic nucleotide such as UNA or GNA, a mismatch, a base-deleted nucleotide, or DNA) located opposite the primer portion of the antisense strand (i.e., at positions 2-8 from 5' the end of the antisense strand);

T1 is a nucleotide having a chemical modification at the 2' position or a similar position in a nucleotide lacking ribose, an acyclic nucleotide, or in the backbone of the nucleotide that confers less steric bulk to the nucleotide than a 2'-OMe modification; for example, T1 is selected from the group consisting of DNA, RNA, LNA, 2'-F and 2'-F-5'-methyl;

the length of n ¹ or n ³ is independently 4 to 15 nucleotides;

length n ⁵ is 1-6 nucleotides;

length n ⁴ is 1-3 nucleotides; alternatively n ⁴ is 0, and

- 31 043057 length n ² is 0-3 nucleotides.

In one embodiment, the sequence of the sense strand of the dsRNA agent, which is 19, 20, 21, or 22 nucleotides in length, is represented by formula (Is):

(Is), where each of B1, B2, B3 is independently a nucleotide having a modification selected from the group consisting of 2'-O-alkyl, 2'-substituted alkoxy, 2'-substituted alkyl, 2'halogen, ENA and BNA/LNA;

C1 is a thermostability-disrupting nucleotide (e.g., an acyclic nucleotide such as UNA or GNA, a mismatch, a base-deleted nucleotide, or DNA) located opposite the primer portion of the antisense strand (i.e., at positions 2-8 from 5' the end of the antisense strand);

T1 is a nucleotide containing a chemical modification selected from the group consisting of DNA, RNA, LNA, 2'-F and 2'-F-5'-methyl;

the length of n ¹ or n ³ is independently 4 to 15 nucleotides;

length n ⁵ is 1-6 nucleotides;

length n ⁴ is 1-3 nucleotides; alternatively, n ⁴ is 0 and n ² is 0-3 nucleotides long.

In one embodiment, the dsRNA agent of formula (Is) further comprises a 3' and/or 5' overhang(s) that is 1-10 nucleotides in length. In one example, a dsRNA agent of formula (Is) has a 5' overhang.

In one embodiment, C1 provides one thermostability-disrupting nucleotide at position 14, 15, 16, or 17 from the 5' end of the sense strand. For example, C1 is an acyclic nucleotide (eg, UNA or GNA), a mismatch, a nucleotide with a base removed, or DNA. According to one specific example, C1 is a GNA.

In one embodiment, T1 provides for DNA, RNA, LNA, 2'-F or 2'-F-5'methyl at position 11 from the 5' end of the sense strand.

In one embodiment, the dsRNA agent of the present invention comprises a sense strand (Is) wherein C1 is an acyclic nucleotide (eg, UNA or GNA), a mismatch, a base-deleted nucleotide, or DNA; and T1 includes DNA, RNA, LNA, 2'-F or 2'-F-5'-methyl at position 11 from the 5' end of the sense strand.

In one embodiment, the sequence of the antisense strand of the dsRNA agent is represented by formula (Ia):

(la), where each of B1', B2', B3' and B4' is independently a nucleotide having a modification selected from the group consisting of 2'-O-alkyl, 2'-substituted alkoxy, 2'-substituted alkyl , 2'-halogen, ENA and BNA/LNA;

each T1', T2' and T3' is independently a nucleotide having a chemical modification at the 2' position or a similar position in a nucleotide lacking ribose, an acyclic nucleotide, or in a nucleotide backbone that confers less steric bulk to the nucleotide than a 2'-OMe modification; for example, each T1', T2' and T3' is selected from the group consisting of DNA, RNA, LNA, 2'-F and 2'-F-5'-methyl;

length q ¹ is independently 4 to 15 nucleotides;

the length of q ³ or q ⁷ is independently 1-6 nucleotides;

the length of q ² or q ⁶ is independently 1-3 nucleotides;

q ⁴ length is independently 0-3 nucleotides; and q ⁵ length is independently 0-10 nucleotides.

In one embodiment, the sequence of the antisense strand of the dsRNA agent, which is 19, 20, 21, 22, 23, 24, or 25 nucleotides in length, is represented by formula (Ia):

(la), where each of B1', B2', B3' and B4' is independently a nucleotide having a modification selected from the group consisting of 2'-O-alkyl, 2'-substituted alkoxy, 2'-substituted alkyl , 2'-halogen, ENA and BNA/LNA;

- 32 043057 each of T1', T2' and T3' independently represents a nucleotide containing a chemical modification selected from the group consisting of DNA, RNA, LNA, 2'-F and 2'-F-5'-methyl;

length q ¹ is independently 4 to 15 nucleotides;

the length of q ³ or q ⁷ is independently 1-6 nucleotides;

the length of q ² or q ⁶ is independently 1-3 nucleotides;

the q ⁴ length is independently 0-3 nucleotides; and the q ⁵ length is independently 0-10 nucleotides.

In one embodiment, the dsRNA of formula (Ia) further comprises a 3' and/or 5' overhang(s) that is 1-10 nucleotides in length. In one embodiment, the dsRNA of formula (Ia) has a 3' overhang.

According to one embodiment, the present invention relates to a double-stranded RNA (dsRNA) agent for suppressing the expression of a target gene. The tool, which is a double-stranded dsRNA, contains a sense strand and an antisense strand, with each strand containing from 14 to 40 nucleotides:

where each of B1, B2, B3, B1', B2', B3' and B4' is independently a nucleotide having a modification selected from the group consisting of 2'-O-alkyl, 2'-substituted alkoxy, 2'-substituted alkyl, 2'-halogen, ENA and BNA/LNA;

C1 is an acyclic nucleotide (eg UNA or GNA);

each of T1, T1', T2' and T3' independently represents a nucleotide containing a chemical modification selected from the group consisting of DNA, RNA, LNA, 2'-F and 2'-F-5'-methyl;

the length of n ¹ , n ³ or q ¹ is independently 4 to 15 nucleotides;

the length of n ⁵ , q ³ or q ⁷ is independently 1-6 nucleotides;

the length of n ⁴ , q ² or q ⁶ is independently 1-3 nucleotides; alternatively, n ⁴ is 0, the length of n ² or q ⁴ is independently 0-3 nucleotides;

the q ⁵ length is independently 0-10 nucleotides and wherein the dsRNA agent has a 3' and/or 5' overhang(s) of 1-10 nucleotides in the antisense and/or sense strand(s).

According to one embodiment, the present invention relates to a double-stranded RNA (dsRNA) agent for suppressing the expression of a target gene. The tool, which is a double-stranded dsRNA, contains a sense strand and an antisense strand, with each strand containing from 14 to 40 nucleotides:

where each of B1, B2, B3, B1', B2', B3' and B4' is independently a nucleotide having a modification selected from the group consisting of 2'-O-alkyl, 2'-substituted alkoxy, 2'-substituted alkyl, 2'-halogen, ENA and BNA/LNA;

C1 is an acyclic nucleotide (eg UNA or GNA);

each of T1, T1', T2' and T3' independently represents a nucleotide containing a chemical modification selected from the group consisting of DNA, RNA, LNA, 2'-F and 2'-F-5'-methyl;

the length of n ¹ , n ³ or q ¹ is independently 4 to 15 nucleotides;

the length of n ⁵ , q ³ or q ⁷ is independently 1-6 nucleotides;

the length of n ⁴ , q ² or q ⁶ is independently 1-3 nucleotides; alternatively, n ⁴ is 0, the length of n ² or q ⁴ is independently 0-3 nucleotides;

length q ⁵ is independently 0-10 nucleotides; and wherein the dsRNA agent has a 2 nucleotide long 3' overhang at the 3' end of the antisense strand.

According to one embodiment, the present invention relates to a double-stranded RNA (dsRNA) agent for suppressing the expression of a target gene. The tool, which is a double-stranded dsRNA, contains a sense strand and an antisense strand, with each strand containing 15-30 nucleotides:

- 33 043057

where each of B1, B2, B3, B1', B2', B3' and B4' is independently a nucleotide having the 2'-OMe modification;

C1 is an acyclic GNA nucleotide;

each of T1, T1', T2' and T3' is independently DNA or RNA;

the length of n ¹ , n ³ or q ¹ is independently 4 to 15 nucleotides;

the length of n ⁵ , q ³ or q ⁷ is independently 1-6 nucleotides;

the length of n ⁴ , q ² or q ⁶ is independently 1-3 nucleotides; alternatively, n ⁴ is 0, the length of n ² or q ⁴ is independently 0-3 nucleotides;

the q ⁵ length is independently 0-10 nucleotides and wherein the dsRNA agent has a 1-6 nucleotide long 3' overhang at the 3' end of the antisense strand.

According to one embodiment, the present invention relates to a double-stranded RNA (dsRNA) agent for suppressing the expression of a target gene. The tool, which is a double-stranded dsRNA, contains a sense strand and an antisense strand, with each strand containing 19-23 nucleotides:

where each of B1, B2, B3, B1', B2', B3' and B4' is independently a nucleotide containing a 2'-OMe modification;

C1 is an acyclic GNA nucleotide;

T1, T1', T2' and T3' are independently DNA or RNA;

the length of n ¹ , n ³ , q ¹ or q ³ is independently 4 to 15 nucleotides;

the length of n ⁵ , q ³ or q ⁷ is independently 1-6 nucleotides;

the length of n ⁴ , q ² or q ⁶ is independently 1-3 nucleotides; alternatively, n ⁴ is 0, the length of n ² , q ⁴ or q ⁵ is independently 0-3 nucleotides;

length q ⁵ is independently 0-10 nucleotides; and wherein the dsRNA agent has a 2 nucleotide long 3' overhang at the 3' end of the antisense strand.

According to one embodiment, the present invention relates to a double-stranded RNA (dsRNA) agent for suppressing the expression of a target gene. The tool, which is a double-stranded dsRNA, contains a sense strand and an antisense strand, with each strand containing from 14 to 40 nucleotides:

where each of B1, B2, B3, B1', B2', B3' and B4' is independently a nucleotide having a modification selected from the group consisting of 2'-O-alkyl, 2'-substituted alkoxy, 2'-substituted alkyl, 2'-halogen, ENA and BNA/LNA;

C1 is an acyclic nucleotide (eg UNA or GNA);

each of T1, T1', T2' and T3' independently represents a nucleotide containing a chemical modification selected from the group consisting of DNA, RNA, LNA, 2'-F and 2'-F-5'-methyl;

the length of n ¹ , n ³ or q ¹ is independently 4 to 15 nucleotides;

the length of n ⁵ , q ³ or q ⁷ is independently 1-6 nucleotides;

the length of n ⁴ , q ² or q ⁶ is independently 1-3 nucleotides; alternatively, n ⁴ is 0, the length of n ² or q ⁴ is independently 0-3 nucleotides;

the q ⁵ length is independently 0-10 nucleotides, and wherein the dsRNA agent has a 1-10 nucleotide long 5' overhang at the 5' end of the sense strand.

According to one embodiment, the present invention relates to a double-stranded RNA (dsRNA) agent for suppressing the expression of a target gene. The tool, which is a double-stranded dsRNA, contains a sense strand and an antisense strand, with each strand containing from 14 to 40 nucleotides:

- 34 043057

where each of B1, B2, B3, B1', B2', B3' and B4' is independently a nucleotide having a modification selected from the group consisting of 2'-O-alkyl, 2'-substituted alkoxy, 2'-substituted alkyl, 2'-halogen, ENA and BNA/LNA;

C1 is an acyclic nucleotide (eg UNA or GNA);

each of T1, T1', T2' and T3' is independently a nucleotide containing a chemical modification selected from the group consisting of DNA, RNA, LNA, 2'-F and 2'-P-5'-methyl;

the length of n ¹ , n ³ or q ¹ is independently 4 to 15 nucleotides;

the length of n ⁵ , q ³ or q ⁷ is independently 1-6 nucleotides;

the length of n ⁴ , q ² or q ⁶ is independently 1-3 nucleotides; alternatively, n ⁴ is 0, the length of n ² or q ⁴ is independently 0-3 nucleotides;

length q ⁵ is independently 0-10 nucleotides; and wherein the dsRNA agent has a 5' overhang of 1-6 nucleotides in length at the 5' end of the sense strand.

According to one embodiment, the present invention relates to a double-stranded RNA (dsRNA) agent for suppressing the expression of a target gene. The tool, which is a double-stranded dsRNA, contains a sense strand and an antisense strand, with each strand containing from 14 to 40 nucleotides:

where each of B1, B2, B3, B1', B2', B3' and B4' is independently a nucleotide having a modification selected from the group consisting of 2'-O-alkyl, 2'-substituted alkoxy, 2'-substituted alkyl, 2'-halogen, ENA and BNA/LNA;

C1 is an acyclic nucleotide (eg UNA or GNA);

each of T1, T1', T2' and T3' is independently a nucleotide containing a chemical modification selected from the group consisting of DNA, RNA, LNA, 2'-F and 2'-P-5'-methyl;

the length of n ¹ , n ³ or q ¹ is independently 4 to 15 nucleotides;

the length of n ⁵ , q ³ or q ⁷ is independently 1-6 nucleotides;

the length of n ⁴ , q ² or q ⁶ is independently 1-3 nucleotides; alternatively, n4 is 0, the length of n ² or q ⁴ is independently 0-3 nucleotides;

q ⁵ is independently 0-10 nucleotides in length, and wherein the dsRNA agent has a 1-10 nucleotide 5' overhang at the 5' end of the sense strand and a 1-10 nucleotide 3' overhang at the 5'- end of the antisense strand.

Modifications violating thermal stability.

The dsRNA agent can be optimized for RNA interference by increasing the susceptibility of the dsRNA duplex to dissociation or melt (reducing the association free energy of the duplex) by introducing a thermostability-disrupting modification into the sense strand at the site opposite the seeding site of the antisense strand (i.e., to positions 2-8 from the 5' end of the antisense strand). This modification may increase the susceptibility of the duplex to dissociate or melt in the seed portion of the antisense strand.

Thermostability-impairing modifications may include modification to remove the nitrogenous base; mismatch with the opposite nucleotide of the opposite strand; and a sugar modification such as a 2'-deoxy modification or an acyclic nucleotide, such as non-locked nucleic acids (UNA) or glyceric nucleic acid (GNA).

Illustrative modifications to remove the nitrogenous base are:

- 35 043057

Illustrative sugar modifications are:

2'-deoxy non-locked nucleic acid glycolic nucleic acid

R=H, OH, O-alkyl ^R=H > ^0Η O-alkyl

The term acyclic nucleotide refers to any nucleotide with an acyclic ribose sugar, for example, in which any of the bonds between the carbon atoms of the ribose (for example, C1'-C2', C2'C3', C3'-C4', C4'-O4' or d '-O4') is absent and/or at least one of the carbon atoms or an oxygen atom of the ribose (eg C1', C2', C3', C4' or O4') is absent from the nucleotide, independently or in combination. In some embodiments, the acyclic nucleotide is:

where B is a modified or unmodified nucleobase, R ¹ and R ² are independently H, halogen, OR ₃ or alkyl, and

R ₃ is H, alkyl, cycloalkyl, aryl, aralkyl, heteroaryl or sugar.

The term UNA refers to an unlocked acyclic nucleic acid in which any of the sugar bonds have been removed to form an unlocked sugar residue. According to one example, UNA also encompasses monomers that have had their C1'-C4' bonds removed (ie the covalent carbon-oxygen-carbon bond between C1' and C4' carbons). According to another example, the C2'C3' bond (i.e. the covalent carbon-carbon bond between C2' and C3' carbons) in the sugar is removed (see Mikhailov et. al., Tetrahedron Letters, 26 (17): 2059 (1985 ); and Fluiter et al., Mol. Biosyst., 10: 1039 (2009), which are hereby incorporated by reference in their entirety). The acyclic derivative provides more backbone flexibility without affecting Watson-Crick base pairing. An acyclic nucleotide may be linked via a 2'-5' or 3'-5' bond.

The term 'GNA' refers to a glycolic nucleic acid, which is a polymer similar to DNA or RNA, but differs in its backbone composition because it consists of glycerol repeating units linked by phosphodiester bonds:

The thermostability-disrupting modification may be mismatches (ie, non-complementary base pairs) between the thermostability-disrupting nucleotide and the opposite nucleotide of the opposite strand of the dsRNA duplex. Illustrative mismatches include G:G, G:A, G:U, G:T, A:A, A:C, C:C, C:U, C:T, U:U, T:T, U:T or a combination. Other types of base mismatch known in the art are also applicable to the present invention. Mismatches can occur between nucleotides that are both naturally occurring and modified nucleotides, i. base mismatches can occur at the nucleobases of the corresponding nucleotides, regardless of the modifications present on the ribose sugars of the nucleotides. In certain embodiments, the dsRNA agent contains at least one mismatched nucleobase, i.e., a 2'-deoxyribonucleotide; for example, a 2'-deoxyribonucleotide in the sense strand.

More examples of base-deleted nucleotides, acyclic nucleotide modifications (including UNA and GNA), and mismatch modifications are described in detail in WO 2011/133876, which is incorporated herein by reference in its entirety.

Thermal stability modifications can also include a universal base with reduced or eliminated ability to form hydrogen bonds with opposite strand bases, as well as phosphate modifications.

Modifications of nucleic bases with a reduced or completely eliminated ability to form hydrogen bonds with bases in the opposite strand have been investigated in relation to

- 36 043057 destabilization of the central region of the dsRNA duplex, as described in document WO 2010/0011895, which is incorporated herein by reference in its entirety. Illustrative nucleobase modifications are:

Exemplary phosphate modifications known to reduce the thermal stability of dsRNA duplexes relative to natural phosphodiester bonds are:

ό ό ό ό όό

O=PSH O=P~CH ₃ O=PCH ₂ COOH O=PR OT NH-R O=P OR IIII II

LLC LLC

R=alkyl

In one embodiment, the dsRNA agent of the present invention may contain 2'-5' bonds (to 2'-H, 2'-OH, and 2'-OMe, as well as to P=O or P=S). For example, 2'-5' linkage modifications can be used to promote nuclease resistance or to suppress binding of the sense and antisense strands, or they can be used at the 5' end of the sense strand to prevent activation of the sense strand by RISC.

In another embodiment, the dsRNA agent of the present invention may contain L-caxapa (eg, L-ribose, L-arabinose with 2'-H, 2'-OH and 2'-OMe). For example, L-sugar modifications can be used to promote nuclease resistance or to suppress binding of the sense and antisense strands, or they can be used at the 5' end of the sense strand to prevent activation of the sense strand by RISC.

In one embodiment, the dsRNA agent is a multimer containing at least two duplexes represented by formula (I), wherein said duplexes are connected by a linker. The linker may be cleavable or non-cleavable. Optionally, said multimer further comprises a ligand. Each dsRNA agent can target one gene or two different genes; or each dsRNA agent can target one gene at two different target sites.

In one embodiment, the dsRNA agent is a multimer containing at least three, four, five, six or more duplexes represented by formula (I), wherein said duplexes are connected by a linker. The linker may be cleavable or non-cleavable. Optionally, said multimer further comprises a ligand. Each dsRNA agent can target one gene or two different genes; or each dsRNA agent can target one gene at two different target sites.

In one embodiment, two dsRNA agents represented by formula (I) are linked together at the 5' end, and one or both of the 3' ends are optionally conjugated to a ligand. Each dsRNA can target one gene or two different genes; or each dsRNA can target one gene at two different target sites.

Various publications describe multimeric siRNAs, all of which can be used with the dsRNAs of the present invention. Such publications include WO2007/091269, US Patent No. 7858769, WO2010/141511, WO2007/117686, WO2009/014887 and WO2011/031520, which are hereby incorporated by reference in their entirety.

In the case of a dsRNA agent that contains one or more carbohydrate moieties conjugated to the dsRNA agent, one or more properties of the dsRNA agent may be enhanced. In many cases, the carbohydrate moiety will be attached to a modified subunit of the dsRNA agent. For example, the ribose sugar of one or more ribonucleotide subunits of the dsRNA agent can be replaced with another moiety, such as a non-carbohydrate (preferably cyclic) carrier, to which a carbohydrate ligand is attached. A ribonucleotide subunit in which the ribose sugar of the subunit has been substituted in this way is referred to herein as a ribose-modified subunit (RRMS). The cyclic carrier may be a carbocyclic ring system, i. e. all atoms in the ring are carbon atoms, or a heterocyclic ring system, i.e. one or more atoms in the ring

- 37 043057 be a heteroatom, for example, a nitrogen, oxygen, sulfur atom. The cyclic carrier may be a monocyclic ring system or may contain two or more rings, such as fused rings. The cyclic carrier may be a fully saturated ring system, or it may contain one or more double bonds.

The ligand may be attached to the polynucleotide via a carrier. Carriers include:

(i) at least one core attachment point, preferably two core attachment points, and (ii) at least one bonding attachment point.

The expression point of attachment to the backbone as used herein refers to a functional group, for example a hydroxyl group, or in general to a bond provided for introducing a support into the backbone and which is suitable for this, for example, to a phosphate or a modified phosphate, for example, sulfur-containing backbone bond of ribonucleic acid. The expression linking point of attachment (TAP), in some embodiments, refers to a ring atom of a cyclic carrier, for example, a carbon atom or a heteroatom (other than the atom that provides the point of attachment to the backbone), to which the selected fragment binds. The fragment may be, for example, a carbohydrate, such as a monosaccharide, a disaccharide, a trisaccharide, a tetrasaccharide, an oligosaccharide, and a polysaccharide. Optionally, the selected fragment is attached via an intermediate linking fragment to a cyclic carrier. Thus, the cyclic carrier will often include a functional group, such as an amino group, or generally provide a bond that is suitable for introducing or linking another chemical entity, such as a ligand, to an atom that is part of the ring.

In one embodiment, the dsRNA agent is conjugated to a ligand via a carrier, where the carrier may be a cyclic group or an acyclic group; Preferably, a cyclic group was chosen from pyrrolidinyl, pyrazolinyl, piraisolidine, imidazolinyl, imidazolidinil, piperidinyl, purezinil, [1.3]-deioxolan, oxasolidinil, Izoxazoilidine, thiazolidinyl, isotiazolydinyl, pyrinelinylinylinylinyl, pyroinylinyl. Dazinonil, Tetrahydrofurila and Decalin; preferably, the acyclic group is selected from a serinol backbone or a diethanolamine backbone.

The double-stranded RNA (dsRNA) agent of the present invention may optionally be conjugated to one or more ligands. The ligand may be attached to the sense strand, the antisense strand, or both strands at the 3' end, 5' end, or both ends. For example, the ligand may be conjugated to the sense strand, in particular the 3' end of the sense strand.

In one embodiment, the dsRNA agents of the present invention are 5'-phosphorylated or include a phosphoryl analog at the 5'-(dash)-terminus. 5'Phosphate modifications include modifications that are compatible with RISC-mediated gene silencing. Suitable modifications include 5'-monophosphate ((HO) ₂ (O)P-O-5');5'-diphosphate ((HO) ₂ (O)P-OR(HO)(O)-O-5');5'-triphosphate ((HO) ₂ (O)P-O-(HO)(O)P-0-P(Ho)(0)-O-5');5'-guanosine cap (7methylated or unmethylated) (7m-G-0-5'-(H0)(0)P-0-(H0)(0)P-0-P(H0)(0)-0- 5');5'-adenosine cap (Appp) and any modified or unmodified nucleotide cap structure (NO-5'(HO)(o)P-0-(HO)(O)P-O-P(HO)(O)- O-5');5'-monothiophosphate(phosphorothioate; (HO) ₂ (S)PO-5');5'monodithiophosphate(phosphorodithioate;(HO)(HS)(S)PO-5'),5'-phosphorothiolate ((HO) ₂ (O)PS-5'); any additional combination of oxygen/sulfur substituted monophosphate, diphosphate and triphosphates (e.g. 5'-alphathiotriphosphate, 5'-gamma-thiotriphosphate, etc.), 5'-phosphoramidates ((HO) ₂ (O)P-NH-5 ', (HO)(NH ₂ )(O)PO-5'), 5'alkylphosphonates (R=αkyl=methyl, ethyl, isopropyl, propyl, etc. e.g. RP(OH)(O)-O- 5'-, 5'alkenylphosphonates (i.e. vinyl, substituted vinyl), (OH)2(O)P-5'-CH2-), 5'-alkyl etherphosphonates (R=alkyl ether=methoxymethyl (MeOCH ₂ -) , ethoxymethyl, etc., for example, RP(OH)(O)-O-5'-). In one example, the modification can be placed on the antisense strand of a dsRNA agent.

Ligands.

A wide range of chemical building blocks can be combined with the oligonucleotides of the present invention. Preferred moieties are ligands that link, preferably covalently, either directly or indirectly through an intermediate linking moiety.

In preferred embodiments, the ligand changes the distribution, targeting, or lifetime of the molecule into which it is introduced. In preferred embodiments, the ligand provides increased affinity for a selected target, e.g., molecule, cell, or cell type, compartment, receptor, e.g., cell compartment or compartment of an organ, tissue, organ, or body site, e.g. there is no such ligand. Ligands that provide increased affinity for a selected target are also referred to as targeting ligands.

Some ligands may have endosomolytic properties. Endosomolytic ligands promote lysis of the endosome and/or transport of the composition of the present invention or its

- 38 043057 components from the endosome to the cytoplasm of the cell. The endosomolytic ligand can be a polyanionic peptide or peptidomimetic that exhibits pH-dependent membrane activity and fusogenicity. In one embodiment, the endosomolytic ligand assumes its active conformation at the pH of the endosome. The active conformation is one in which the endosomolytic ligand promotes lysis of the endosome and/or transport of the composition of the present invention or its components from the endosome into the cytoplasm of the cell. Exemplary endosomolytic ligands include the GALA peptide (Subbarao et al., Biochemistry, 1987, 26: 2964-2972, which is incorporated by reference in its entirety), the EALA peptide (Vogel et al., J. Am. Chem. Soc, 1996, 118: 1581-1586, which is incorporated by reference in its entirety) and derivatives thereof (Turk et al., Biochem. Biophys. Acta, 2002, 1559: 56-68, which is incorporated by reference in its entirety). In one embodiment, the endosomolytic component may contain a chemical group (eg, an amino acid) that will change charge or protonation in response to changes in pH. The endosomolytic component may be linear or branched.

Ligands can improve transport, hybridization, and specificity properties, and can also improve nuclease resistance in the resulting natural or modified oligoribonucleotide or polymer molecule containing any combination of the monomers described herein and/or natural or modified ribonucleotides.

Ligands, as a rule, may include therapeutically active modifiers, for example, to enhance absorption; diagnostic compounds or reporter groups, for example, to track distribution; crosslinkers; and nuclease resistance conferring moieties. General examples include lipids, steroids, vitamins, sugars, proteins, peptides, polyamines, and peptide mimetics.

Ligands may include a naturally occurring substance such as a protein (eg, human serum albumin (HSA), low density lipoprotein (LDL), high density lipoprotein (HDL) or globulin); carbohydrate (eg dextran, pullulan, chitin, chitosan, inulin, cyclodextrin or hyaluronic acid) or lipid. The ligand may also be a recombinant or synthetic molecule such as a synthetic polymer, eg a synthetic polyamino acid, an oligonucleotide (eg an aptamer). Examples of polyamino acids include polyamino acid, which is polylysine (PLL), poly-L-aspartic acid, poly-L-glutamic acid, styrene-maleic anhydride copolymer, L-lactide-glycolide copolymer, divinyl ether-maleic anhydride copolymer, K-( 2-hydroxypropyl)methacrylamide copolymer (HMPA), polyethylene glycol (PEG), polyvinyl alcohol (PVA), polyurethane, poly(2-ethylacrylic acid), N-isopropylacrylamide polymers, or polyphosphazine. Examples of polyamines include polyethyleneimine, polylysine (PLL), spermine, spermidine, polyamine, pseudopeptide polyamine, polyamine peptidomimetic, polyamine dendrimer, arginine, amidine, protamine, cationic lipid, cationic porphyrin, polyamine quaternary salt, or alpha helical peptide.

Ligands can also include targeting groups, eg, a cell or tissue targeting agent, eg, a lectin, glycoprotein, lipid, or protein, eg, an antibody that binds to a specific cell type, eg, a kidney cell. The target group can be thyrotropin, melanotropin, lectin, glycoprotein, surfactant protein A, carbohydrate mucin, polyvalent lactose, polyvalent galactose, N-acetylgalactosamine, N-acetylglucosamine, polyvalent mannose, polyvalent fucose, glycosylated polyamino acids, polyvalent galactose, transferrin, bisphosphonate, polyglutamate, polyaspartate, lipid, cholesterol, steroid, bile acid, folate, vitamin B12, biotin, RGD peptide, RGD peptide mimetic or aptamer. In table. 2 shows some examples of targeting ligands and their associated receptors.

Other examples of ligands include dyes, intercalators (eg, acridines), cross-linkers (eg, psoralen, mitomycin C), porphyrins (TPPC4, texaphyrin, sapphirin), polycyclic aromatic hydrocarbons (eg, phenazine, dihydrophenazine), artificial endonucleases, or a chelator. (e.g. EDTA), lipophilic molecules, e.g. cholesterol, cholic acid, adamantanoacetic acid, 1-pyrenebutyric acid, dihydrotestosterone, 1,3-bis-O(hexadecyl)glycerol, geranyloxyhexyl group, hexadecylglycerol, borneol, menthol, 1,3propanediol , heptadecyl group, palmitic acid, myristic acid, O ₃ (oleoyl)lithocholic acid, O ₃ -(oleoyl)cholenic acid, dimethoxytrityl or phenoxazine, as well as peptide conjugates (for example, antennapedia peptide, Tat-peptide), alkylating agents, phosphate , amino, mercapto, PEG (eg, PEG-40K), MPEG, [MPEG] ₂ , polyamino, alkyl, substituted alkyl, radiolabeled markers, enzymes, haptens (eg, biotin), transport/absorption aids (eg, aspirin, vitamin E, folic acid), synthetic ribonucleases (eg imidazole, bisimidazole, histamine, imidazole clusters, acridine-imidazole conjugates, Eu3 ⁺ tetraazamacrocycle complexes), dinitrophenyl, HRP or AP.

Ligands can be proteins, such as glycoproteins, or peptides, such as molecules with a specific affinity for a co-ligand, or antibodies, such as an antibody that binds

- 39 043057 is associated with a specific cell type, such as a cancer cell, an endothelial cell, or a bone cell. Ligands may also include hormones and hormone receptors. They may also include non-peptidic molecules such as lipids, lectins, carbohydrates, vitamins, cofactors, polyvalent lactose, polyvalent galactose, N-acetylgalactosamine, N-acetylglucosamine, polyvalent mannose, polyvalent fucose, or aptamers. For example, the ligand may be a lipopolysaccharide, a p38 MAP kinase activator, or an NF-kB activator.

The ligand may be a substance, such as a drug, that can increase uptake by the cell of the RNA interference agent, such as by disrupting the cell's cytoskeleton, such as disrupting microtubules, microfilaments, and/or intermediate filaments of the cell. The drug may, for example, be a taxon, vincristine, vinblastine, cytochalasin, nocodazole, japlakinolide, latrunculin A, phalloidin, swincholide A, indanocin, or myoservin.

The ligand can increase uptake of the oligonucleotide into the cell, for example by activating an inflammatory response. Exemplary ligands that will have this effect include tumor necrosis factor alpha (TNF-alpha), interleukin-1-beta, or interferon gamma.

In one aspect, the ligand is a lipid or lipid-based molecule. Such a lipid or lipid-based molecule is preferably associated with a serum protein, such as human serum albumin (HSA). The HSA-binding ligand provides distribution of the conjugate to the target tissue, eg, target body tissue other than kidney tissue. For example, the target tissue may be the liver, including parenchymal liver cells. Other molecules that can bind to HSA can also be used as ligands. For example, naproxen or aspirin can be used. A lipid or lipid-based ligand can:

(a) increase the resistance of the conjugate to degradation, (b) enhance targeting or transport to the target cell or cell membrane, and/or (c) can be used to adjust serum protein binding, eg HSA.

A lipid-based ligand can be used to modulate, for example, regulate the binding of the conjugate to the target tissue. For example, a lipid or lipid-based ligand that binds more strongly to HSA is less likely to be targeted to the kidneys and therefore less likely to be excreted from the body. A lipid or lipid-based ligand that binds less strongly to HSA can be used to target the conjugate to the kidneys.

In a preferred embodiment, the lipid-based ligand binds to HSA. Preferably, it binds to HSA with sufficient affinity such that the conjugate preferentially distributes to tissue other than kidney tissue. However, it is preferred that the affinity is not so strong that the binding of the HSA ligand is irreversible.

In another preferred embodiment, the lipid-based ligand binds little or no to HSA, such that the conjugate will preferentially distribute to the kidney. Other moieties that target kidney cells can also be used instead of or in addition to the lipid based ligand.

In another aspect, the ligand is a moiety, such as a vitamin, that is taken up by a target cell, such as a proliferating cell. This is particularly applicable in the treatment of disorders characterized by unwanted cell proliferation, eg malignant or benign type, eg cancer cells. Exemplary vitamins include vitamin A, E, and K. Other exemplary vitamins include B vitamins, such as folic acid, B12, riboflavin, biotin, pyridoxal, or other vitamins or nutrients taken up by cancer cells. Also included are HAS, low density lipoprotein (LDL), and high density lipoprotein (HDL).

In another aspect, the ligand is a cell-penetrator, preferably a helical cell-penetrator. Preferably, the agent is amphipathic.

An exemplary agent is a peptide such as tat or antennapedia. If the agent is a peptide, it may be modified, the agent may include a peptidyl mimetic, invertomers other than peptide or pseudopeptide bonds, and it may also use D-amino acids. The helical agent is preferably an alpha helical agent which is preferably characterized by lipophilic and lipophobic phases.

The ligand may be a peptide or a peptidomimetic.

A peptidomimetic (also referred to herein as an oligopeptidomimetic) is a molecule capable of folding into a specific three-dimensional structure similar to a natural peptide. The length of the peptide or peptidomimetic fragment may be about 5-50 amino acids, for example, about 5, 10, 15, 20, 25, 30, 35, 40, 45 or 50 amino acids. The peptide or peptidomimetic, for example, may be a peptide that provides penetration into

- 40 043057 cell, cationic peptide, amphipathic peptide or hydrophobic peptide (for example, consisting predominantly of Tyr, Trp or Phe). The peptide fragment may be a dendrimer peptide, a conformationally constrained peptide, or a cross-linked peptide. In another alternative, the peptide fragment may include a hydrophobic membrane transfer control sequence (MTS). An exemplary hydrophobic MTS-containing peptide is RFGF with the amino acid sequence aavallpavllallap. An RFGF analog (eg, the amino acid sequence aallpvllaap) containing a hydrophobic MTS can also be a targeting fragment. The peptide moiety can be a delivery peptide that can transport large polar molecules, including peptides, oligonucleotides, and proteins, across cell membranes. For example, the sequences of the ^T at hiv protein (grkkrrqrrrppq) and the Drosophila Antennapedia protein (RQIKIWFQNRRMKWKK) have been found to be able to function as delivery peptides. The peptide or peptidomimetic may be encoded by a random DNA sequence, such as a peptide identified from a phage display library or a one-bead-one-compound (EIA) combinatorial library (Lam et al., Nature, 354:82-84, 1991, which is incorporated by links in their entirety). Preferably, the peptide or peptidomimetic linked to the RNA interference agent via the introduced monomeric unit is a cell-targeting peptide, such as an arginine-glycine-aspartic acid (RGD)-containing peptide or RGD mimetic. The peptide fragment can vary in length from about 5 amino acids to about 40 amino acids. Peptide fragments may have a structural modification, such as to improve stability or control conformational properties. You can use any of the structural modifications described below. An RGD peptide fragment can be used to target a tumor cell such as an endothelial tumor cell or a breast cancer tumor cell (Zitzmann et al., Cancer Res., 62:5139-43, 2002, which is incorporated by reference in its entirety). An RGD peptide can facilitate targeting of an RNAi agent to tumors from a variety of other tissues, including the lung, kidney, spleen, or liver (Aoki et al., Cancer Gene Therapy 8:783-787, 2001, which is incorporated by reference throughout its fullness). Preferably, the RGD peptide will facilitate targeting of the RNA interference agent to the kidney. The RGD peptide may be linear or cyclic and may be modified, such as glycosylated or methylated, to facilitate targeting to specific tissues. For example, a glycosylated RGD peptide can deliver an RNA interference agent to a tumor cell expressing a _v p ₃ (Haubner et al., Jour. Nucl. Med., 42:326-336, 2001, which is incorporated by reference in its entirety) . You can use peptides that target markers that are enriched in proliferating cells. For example, RGD-containing peptides and peptidomimetics can target cancer cells, in particular cells that have an integrin present on their surface. Thus, RGD peptides, cyclic peptides containing RGD, RGD peptides that include D-amino acids, as well as synthetic RGD mimetics can be used. In addition to RGD, other moieties that target the integrin ligand can be used. Typically, such ligands can be used to control cell proliferation and angiogenesis. Preferred are conjugates with this type of ligand that targets PECAM-1, VEGF, or another cancer gene, such as the cancer gene described herein.

The cell entry peptide allows entry into a cell, for example a microbial cell such as a bacterial or fungal cell, or a mammalian cell such as a human cell. The microbial cell-penetrating peptide, for example, can be an α-helical linear peptide (eg, LL-37 or Ceropin P1), a peptide containing a disulfide bond (eg, α-defensin, β-defensin, or bactenecin), or a peptide containing only one or two predominant amino acids (for example, PR-39 or indolicidin). The cell entry peptide may also include a cellular intranuclear localization signal (NLS). For example, the cell entry peptide can be a two-component amphipathic peptide such as MPG, which is derived from the gp41 HIV-1 NLS fusion peptide domain of the SV40 large T antigen (Simeoni et al., Nucl. Acids Res. 31: 2717-2724, 2003, which is incorporated by reference in its entirety).

In one embodiment, the targeting peptide may be an amphipathic α-helical peptide. Exemplary amphipathic α-helical peptides include, without limitation, cecropins, lycotoxins, paradaxins, buforin, CPF, bombinin-like peptide (BLP), cathelicidins, ceratotoxins, S. clava peptides, myxin intestinal antimicrobial peptides (HFIAP), magainins, brevinins- 2, dermaseptins, mellitins, pleurocidins, H2A peptides, Xenopus peptides, esculentins-1 and cerins. Preferably a number of factors will be considered in connection with maintaining the integrity and stability of the helix. For example, will use the maximum number of helix-stabilizing residues (for example, leu, ala or lys) and will use the minimum co

- 41 043057 number of helix-destabilizing residues (eg proline or cyclic monomeric units). A capping residue will be considered (eg, Gly is an exemplary N-capping residue) and/or C-terminal amidation will be used to provide an additional H-bond to stabilize the helix. The stability may provide for the formation of salt bridges between residues with opposite charges separated by i±3 or i±4 positions. For example, cationic residues such as lysine, arginine, homoarginine, ornithine, or histidine can form salt bridges with anionic residues, glutamate, or aspartate.

Peptide ligands and peptidomimetic ligands include ligands having naturally occurring or modified peptides, eg D- or L-peptides; α-, β- or γ-peptides; N-methylpeptides; azapeptides; peptides that have one or more amide, i.e. peptide bonds are substituted by one or more urea, thiourea, carbamate or sulfonylurea bonds; or cyclic peptides.

A targeting ligand can be any ligand that is capable of targeting a specific receptor. Examples are: folate, GalNAc, galactose, mannose, mannose-bP, sugar clusters such as GalNAc cluster, mannose cluster, galactose cluster or aptamer. A cluster is a combination of two or more sugar units. Targeting ligands also include integrin receptor ligands, chemokine receptor ligands, transferrin, biotin, serotonin receptor ligands, PSMA, endothelin, GCPII, somatostatin, LDL and HDL ligands. The backbone of the ligands can also be a nucleic acid, for example in the case of an aptamer. The aptamer may be unmodified or may have any combination of the modifications disclosed herein.

Agents for release from the endosome include imidazoles, poly- or oligoimidazoles, PEI, peptides, fusogenic peptides, polycarboxylates, polycations, hidden oligo- or polycations or anions, acetals, polyacetals, ketals/polyketals, orthoesters, polymers with hidden or non-hidden cationic or anionic charges, dendrimers with hidden or unhidden cationic or anionic charges.

PK modulator means a modulator of pharmacokinetic parameters. The PK modulator includes lipophils, bile acids, steroids, phospholipid analogs, peptides, protein binders, PEGs, vitamins, etc. Exemplary PK modulators include, without limitation, cholesterol, fatty acids, cholic acid, lithocholic acid, dialkylglycerides, diacylglyceride, phospholipids, sphingolipids, naproxen, ibuprofen, vitamin E, biotin, and the like. Oligonucleotides that contain some phosphorothioate linkages are also known to bind to whey protein, so short oligonucleotides, e.g., oligonucleotides of about 5 bases, 10 bases, 15 bases, or 20 bases, containing multiple backbone phosphorothioate linkages are also useful. in the present invention as ligands (eg as PK modulating ligands).

In addition, aptamers that bind to serum components (eg, serum proteins) are also useful as PK modulating ligands in the present invention.

Other conjugates that are ligands useful in the present invention are described in US patent applications USSN: 10/916185, filed Aug. 10, 2004; USSN: 10/946873 filed September 21, 2004; USSN: 10/833,934, filed August 3, 2007; USSN: 11/115989, filed April 27, 2005, and USSN: 11/944227, filed November 21, 2007, which are incorporated by reference in their entirety for all purposes.

Where two or more ligands are present, all ligands may have the same properties, may have different properties, or some ligands have the same properties while others have different properties. For example, the ligand may have targeting properties, have endosomolytic activity, or have PK modulating properties. According to a preferred embodiment, all ligands have different properties.

Ligands can be attached to the oligonucleotides at various locations, such as at the 3'end, 5'end and/or internal position. In preferred embodiments, the ligand is attached to the oligonucleotides via an intermediate binding moiety, such as the carrier described herein. The ligand or associated ligand may be present on the monomer if said monomer is introduced into the growing thread. In some embodiments, the ligand may be introduced by binding to a precursor monomer after said precursor monomer has been introduced into the growing filament. For example, a monomer containing, for example, a linking fragment with an amino group at the end (ie without an associated ligand), such as TAP(CH ₂ ) _n NH ₂ , can be introduced into the growing oligonucleotide strand. At the next stage, i.e. after introducing the precursor monomer into the strand, a ligand with an electrophilic group such as a pentafluorophenyl ether group or an aldehyde group can then be attached to the precursor monomer by linking the electrophilic group of the ligand to the terminal nucleophilic group of the linking moiety of the precursor monomer.

- 42 043057

According to another example, a monomer with a chemical group suitable for participation in click chemistry reactions can be introduced, for example, a linking fragment/linker with an azide or alkyne group at the end. At the next stage, i.e. after introducing the precursor monomer into the strand, a ligand with a complementary chemical group, such as an alkyne or azide group, can be attached to the precursor monomer by linking the alkyne and azide groups together.

With regard to double-stranded oligonucleotides, the ligands may be attached to one or both strands. In some embodiments, the double-stranded RNAi agent comprises a ligand conjugated to the sense strand. In other embodiments, the double-stranded RNA interference agent comprises a ligand conjugated to an antisense strand.

In some embodiments, the ligand may be conjugated to nucleic bases, sugar moieties, or internucleoside bonds in nucleic acid molecules.

Conjugation with purine nucleic bases or their derivatives can occur in any position, including atoms inside the ring and outside the ring. In some embodiments, a conjugate moiety is attached to the 2-, 6-, 7-, or 8-positions of the purine nucleobase. Conjugation to pyrimidine nucleic bases or derivatives thereof can also occur at any position. In some embodiments, the 2-, 5-, and 6-positions of the pyrimidine nucleobase may be substituted with a conjugate moiety. Conjugation to nucleoside sugar moieties can occur at any carbon atom. Examples of carbon atoms in the sugar moiety to which the conjugate moiety can be attached include 2', 3' and 5' carbon atoms. A conjugate moiety can also be attached to position 1', such as in the remainder, with the nitrogenous base removed. Internucleoside bonds can also carry conjugate moieties. With regard to phosphorus-containing bonds (for example, phosphodiester, phosphorothioate, phosphorodithioate, phosphoramidate, etc.), the conjugate moiety can be attached directly to the phosphorus atom or to the O, N or S atom associated with the phosphorus atom. With respect to amine- or amide-containing internucleotide bonds (eg, PNA), the conjugate fragment may be attached to the nitrogen atom of the amine or amide, or to an adjacent carbon atom.

Any suitable ligand in the field of RNA interference may be used, although the ligand is typically a carbohydrate, eg, a monosaccharide (such as GalNAc), a disaccharide, a trisaccharide, a tetrasaccharide, a polysaccharide.

Linkers by which a ligand is conjugated to a nucleic acid include those described above. For example, the ligand may be one or more GalNAc (N-acetylglucosamine) derivatives attached via a monovalent, divalent, or trivalent branched linker.

In one embodiment, the dsRNA of the present invention is conjugated to divalent and trivalent branched linkers comprising the structures shown in any of formulas (IV)-(VII):

Formula (IV)

2B

T^-L*

Formula (V)

Formula (VI) .

Formula (VII) where q ^2A , q ^2B , q ^3A , q ^3B , q ^4A , q ^4B , q ^5A , q ^5B and q ^5c are independently in each case 0-20, and where the repeating units may be the same or different;

each of R ^2A , R ^2B , R ^3A , P ^3B , R ^4A , R ^4B , R ^5A , R ^5B , R ^5C , T ^{2A , T 2B ,} T ^3A , T ^3B , T ^4A , T ^4B , T ^5A , T ^5B , T ^5C independently in each occurrence is CO, NH, O, S, OS(O), NHC(O), CH ₂ , CH2NH or CH ₂ O;

Q ^2A , Q ^2B , Q ^3A , Q ^3B , Q ^4A , Q ^4B , Q ^5A , Q ^5B , Q ^5C independently in each case are absent, are alkylene, substituted alkylene, where one or more methylene groups may be interrupted or terminated by one or more of O, S, S(O), SO ₂ , N(R ⁿ ), C(R')=C(R), CXC or C(O);

each of R ^2A , R ^2B , R ^3A , R 3B , R ^4A , R ^4B , R ^5A , ^{R 5B ,} R ⁵ c independently in each occurrence is NH, O, S, CH2, C(O)O , C(O)NH, NHCH(R ^s )C(O), -C(O)-CH(R ^a )-NH-, CO, CH=NO,

- 43 043057 or heterocyclyl;

L ^2A , L ^2B , L ^3A , L ^3B , L ^4A , L ^4B , L ^5A , L ^5B and L ^5C are the ligand; those. each independently at each occurrence is a monosaccharide (such as GalNAc), a disaccharide, a trisaccharide, a tetrasaccharide, an oligosaccharide, or a polysaccharide; And

R ^a is H or an amino acid side chain.

In the case of RNAi agents serving to inhibit the expression of a target gene, trivalent linkers are particularly useful, by means of which GalNAc derivatives are conjugated, such as the linkers of formula (VII):

where L ^5A , L ^5B and L ^5C represent a monosaccharide, such as a derivative of GalNAc.

Examples of suitable divalent and trivalent branched linker groups through which GalNAc derivatives are conjugated include, without limitation, the following compounds:

NHAc

NHAc

- 44 043057

Definitions.

As used herein, the terms dsRNA, siRNA, and RNA interference agent are used interchangeably to refer to agents that can mediate the silencing of a target RNA, eg mRNA, eg a transcript of a gene encoding a protein. For convenience, such an mRNA is also referred to herein as an mRNA to be silenced. Such a gene is also called a target gene. In a general sense, the RNA to be silenced is an endogenous or pathogenic gene. In addition, RNAs other than mRNAs, such as tRNAs and viral RNAs, can also be targeted.

As used herein, mediate RNAi refers to the ability to silence a target RNA through a sequence-specific mechanism. Without wishing to be bound by any theory, it is believed that silencing uses an RNAi system or process and a guide RNA, eg, a 21-23 nucleotide siRNA agent.

As used herein, specifically hybridizable and complementary are terms that are used to denote a sufficient degree of complementarity such that stable and specific binding occurs between a compound of the present invention and a target RNA molecule. Specific binding requires a sufficient degree of complementarity to avoid non-specific binding of the oligomeric compound to non-target sequences under conditions under which specific binding is expected to be achieved, i.e. under physiological conditions in the case of assays or therapeutic treatment, or, in the case of in vitro assays, under the conditions under which the assays are carried out. Typically, non-target sequences differ by at least 5 nucleotides.

In one embodiment, the dsRNA agent of the present invention is sufficiently complementary to the target RNA, such as the target mRNA, such that the dsRNA agent results in silencing of the production of the protein encoded by the target mRNA. In another embodiment, the dsRNA agent of the present invention is fully complementary to the target RNA, e.g., the target RNA and the dsRNA duplex agent hybridize, e.g., to form a hybrid produced solely by Watson-Crick base pairing. in the region of complete complementarity. A sufficient degree of complementarity with the target RNA may extend to an internal region (eg, at least 10 nucleotides) that is fully complementary with the target RNA. Moreover, in some embodiments, the dsRNA agent of the present invention specifically recognizes a single nucleotide difference. In this case, the dsRNA agent mediates RNAi only when complete complementarity is established in the region (eg, within 7 nucleotides) of a one nucleotide difference.

As used herein, the term oligonucleotide refers to a nucleic acid molecule (RNA or DNA), for example, that is less than 100, 200, 300, or 400 nucleotides in length.

- 45 043057

The term 'BNA' refers to a bridging nucleic acid, and it is often referred to as hindered or inaccessible RNA. The BNA may contain a 5-, 6-, or even 7-membered bridging structure with a fixed C ₃ '-endo sugar overhang. Typically, a bridge is inserted at the 2', 4' position of the ribose to give a nucleotide 2', 4'-BNA (eg, LNA or ENA). Examples of BNA nucleotides include the following nucleosides:

The term 'LNA' refers to a locked nucleic acid and is often referred to as hindered or inaccessible RNA. LNA is a modified RNA nucleotide. The ribose fragment in the LNA nucleotide is modified with an additional bridge (eg methylene bridge or ethylene bridge) connecting the 2'-hydroxyl to the 4'-carbon of the same ribose sugar. For example, the bridge can lock the ribose in the 3'-endo (northern) conformation:

The term ENA refers to ethylene-bridged nucleic acid and is often referred to as hindered or inaccessible RNA.

As used herein, a cleavage site means a link in the backbone of a target gene or sense strand that is cleaved by a RISC system using an RNA interference tool. In this case, the target section of the cleavage site contains at least one or at least two nucleotides on both sides of the cleavage site. In the case of a sense strand, the cleavage site is a bond in the backbone of the sense strand that must be cleaved when the sense strand itself has become a target to be cleaved by the RNAi mechanism. The cleavage site can be determined using methods known in the art, for example, the 5'-RACE assay detailed in Soutschek et al., Nature (2004) 432, 173-178, which is incorporated by reference in its entirety. As is well known in the art, the cleavage site region for a conical double-stranded RNAi agent contains two strands of 21 nucleotides in length (where these strands form a double-stranded region of 19 consecutive base pairs with single-stranded overhangs of 2 nucleotides at the 3' ends), with this section of the cleavage site corresponds to positions 9-12 from the 5' end of the sense strand.

The term halogen refers to any radical based on a fluorine, chlorine, bromine or iodine atom. The term alkyl refers to saturated or unsaturated non-aromatic hydrocarbon chains, which may be straight chain or branched chain containing the designated number of carbon atoms (they include, but are not limited to, propyl, allyl, or propargyl) into which N, O, or S may optionally be inserted. For example , C ₁ -C ₁₀ means that this group may contain from 1 to 10 (inclusive) carbon atoms. The term alkoxy refers to the -O-alkyl radical. The term alkylene refers to divalent alkyl (i.e. -R-). The term alkylenedioxo refers to a divalent group of atoms with the -ORO- structure, in which R is alkylene. The term aminoalkyl refers to alkyl substituted with an amino group. The term mercapto refers to the -SH radical. The term thioalkoxy refers to the -S-alkyl radical.

The term aryl refers to a 6-carbon monocyclic or 10-carbon bicyclic aromatic ring system where 0, 1, 2, 3, or 4 atoms in each ring may be substituted with a substituent. Examples of aryl groups include phenyl, naphthyl, and the like. The term arylalkyl or the term aralkyl refers to alkyl substituted with aryl. The term arylalkoxy refers to alkoxy substituted with aryl.

The term cycloalkyl as used herein includes saturated and partially unsaturated cyclic hydrocarbon groups with 3-12 carbon atoms, such as 3-8 carbon atoms, and, for example, 3-6 carbon atoms, where the cycloalkyl group may additionally be optionally substituted. Cycloalkyl groups include, without limitation, cyclopropyl, cyclobutyl, cyclopentyl, cyclopentenyl, cyclohexyl, cyclohexenyl, cycloheptyl, and cyclooctyl.

The term heteroaryl refers to an aromatic 5-8 membered monocyclic, 8-12 membered bicyclic, or 11-14 membered tricyclic ring system with 1-3 heteroatoms in the case of a monocyclic system, 1-6 heteroatoms in the case of a bicyclic system, or 1-9 heteroatoms in the case of a tricyclic system, while these heteroatoms are selected from O, N or S (for example, the system contains carbon atoms and 1-3, 1-6 or 1-9 heteroatoms from among N, O or S in the case of a monocyclic , bicyclic or tricyclic system, respectively), where 0, 1, 2, 3 or 4 atoms in each ring may be substituted by a substituent. Examples of heteroaryl groups include pyridyl, furyl or furanyl, imidazolyl, benzimidazolyl, pyrimidinyl, thiophenyl or thienyl, quinolinyl, indolyl, thiazolyl, and the like. The term heteroarylalkyl or the term heteroaralkyl refers to alkyl substituted with heteroaryl. The term heteroarylalkoxy refers to alkoxy substituted with heteroaryl.

The term heterocyclyl refers to a non-aromatic 5-8 membered monocyclic, 8-12 membered bicyclic, or 11-14 membered tricyclic ring system with 1-3 heteroatoms in the case of a monocyclic system, 1-6 heteroatoms in the case of a bicyclic system, or 1-9 heteroatoms in the case of a tricyclic system, wherein said heteroatoms are selected from O, N or S (for example, the system contains carbon atoms and 1-3, 1-6 or 1-9 heteroatoms from N, O or S in the case of a monocyclic, bicyclic or tricyclic systems, respectively), where 0, 1, 2, or 3 atoms in each ring may be substituted with a substituent. Examples of heterocyclyl groups include trizolyl, tetrazolyl, piperazinyl, pyrrolidinyl, dioxanyl, morpholinyl, tetrahydrofuranyl, and the like.

The term oxo refers to an oxygen atom that forms a carbonyl when bonded to carbon, an N-oxide when bonded to nitrogen, and a sulfoxide or sulfone when bonded to sulfur.

The term acyl refers to an alkylcarbonyl, cycloalkylcarbonyl, arylcarbonyl, heterocyclylcarbonyl or heteroarylcarbonyl substituent, any of which may be further substituted with substituents.

The term substituted refers to the replacement of one or more hydrogen radicals in a specified structure with a radical of a specified substituent, including, without limitation: halogen, alkyl, alkenyl, alkynyl, aryl, heterocyclyl, thiol, alkylthio, arylthio, alkylthioalkyl, arylthioalkyl, alkylsulfonyl, alkylsulfonylalkyl, arylsulfonylalkyl, alkoxy , aryloxy, aralkoxy, aminocarbonyl, alkylaminocarbonyl, arylaminocarbonyl, alkoxycarbonyl, aryloxycarbonyl, haloalkyl, amino, trifluoromethyl, cyano, nitro, alkylamino, arylamino, alkylaminoalkyl, arylaminoalkyl, aminoalkylamino, hydroxy, alkoxyalkyl, carboxyalkyl, alkoxycarbonylalkyl, aminocarbonylalkyl, acyl, aralkoxycarbonyl, carboxylic acid, sulfonic acid, sulfonyl, phosphonic acid, aryl, heteroaryl, heterocyclic and aliphatic radical. It is understood that a substituent may be further substituted.

Cleavable linker groups.

A cleavable linker group is a group that is reasonably stable outside of the cell, but which, upon entry into the target cell, cleaves to release the two parts that the linker holds together. In a preferred embodiment of the dsRNA agent of the present invention, the cleavable linker group is cleaved about 10-fold or more, preferably at least 100-fold faster in the target cell, or under a first standard condition (which can, for example, be selected for mimic or simulate intracellular conditions) than in the subject's blood or under a second standard condition (which may, for example, be chosen to mimic or simulate conditions found in blood or serum).

Cleavable linker groups are sensitive to cleavage agents, such as pH, redox potential, or the presence of damaging molecules. Generally, cleavage agents are more abundant or occur at higher levels or are more active within cells than in serum or blood. Examples of such disruptive agents include redox agents that are selected for specific substrates or that are not substrate specific, including, for example, oxidizing or reducing enzymes or reducing agents, such as mercaptans, present in cells that can degrade cleavable carbon using the redox potential of the linker group by reduction; esterases; endosomes or agents that can create an acidic environment, such as those that result in a pH value of five or less; enzymes that can hydrolyze or degrade an acid-cleavable linker group, acting as a universal acid, peptidases (which may be substrate-specific), and phosphatases.

A cleavable linker group, such as a disulfide bond, may be pH sensitive. The human serum pH is 7.4, while the average intracellular pH is slightly lower, in the range of approximately 7.1-7.3. Endosomes have a more acidic pH in the range of 5.5-6.0 and lysosomes have an even more acidic pH, around 5.0. Some linkers will contain a cleavable linker group that cleaves at the preferred pH, thereby releasing the cationic lipid from the ligand into the interior of the cell or into the desired compartment of the cell.

The linker may include a cleavable linker group that is cleavable by a particular enzyme. The type of cleavable linker group included in the linker may depend on

- 47 043057 target cell. For example, liver-targeting ligands can be linked to cationic lipids via a linker that contains an ester group. Liver cells are high in esterases and therefore the linker will be cleaved more efficiently in liver cells rather than in cell types that are not high in esterases. Other cell types rich in esterases include lung, renal cortex, and testicular cells.

Linkers that contain peptide bonds can be used when targeting cell types rich in peptidases, such as liver cells and synoviocytes.

Generally, the suitability of a candidate cleavable linker group can be assessed by testing the ability of a damaging agent (or condition) to cleave that candidate linker group. In addition, it is also desirable to test a candidate cleavable linker group for its ability to resist cleavage in the blood or when brought into contact with other non-target tissue. In this way, the sensitivity to cleavage corresponding to the first and second conditions can be determined, the first conditions being chosen to characterize the cleavage in the target cell and the second conditions being chosen to characterize the cleavage in other tissues or body fluids, e.g. blood or serum. Measurements can be made in cell-free systems, in cells, in cell culture, in organ or tissue culture, or in animals in general. It may be useful to perform initial measurements under cell-free or culture conditions and confirm by further measurements on animals as a whole. In preferred embodiments, applicable candidate compounds are cleaved at least 2, 4, 10, or 100 times faster in the cell (or under in vitro conditions selected to mimic intracellular conditions) compared to blood or serum (or under in vitro conditions selected to mimic extracellular conditions).

Redox-cleavable linker groups.

One class of cleavable linker groups are redox cleavable linker groups that can be used in the dsRNA agent of the present invention, these groups are cleaved by reduction or oxidation. An example of a linker group cleavable upon reduction is a disulfide linker group (-S-S-). To determine whether a candidate cleavable linker group is a suitable reductively cleavable linker group, or, for example, suitable for use with a particular RNA interference moiety and a particular targeting agent, one may refer to the methods described herein. For example, a candidate can be assessed by incubation with dithiothreitol (DTT) or other reducing agent using reagents known in the art that mimic the rate of degradation that would be observed in a cell, eg a target cell. Candidates may also be evaluated under conditions that are chosen to mimic conditions in blood or serum. In a preferred embodiment, at most 10% of the candidate compounds are cleaved in the blood. In preferred embodiments, applicable candidate compounds are degraded at least 2, 4, 10, or 100 times faster in the cell (or under in vitro conditions selected to mimic intracellular conditions) compared to blood (or under in vitro conditions selected to mimic extracellular conditions). The degree of cleavage of candidate compounds can be determined using standard enzyme kinetic assays under conditions chosen to mimic the intracellular environment and compared to conditions chosen to mimic the extracellular environment.

Phosphate cleavable linker groups.

Phosphate cleavable linker groups that can be used in the dsRNA agent of the present invention are cleaved by agents that degrade or hydrolyze the phosphate group. An example of an agent that cleaves phosphate groups in cells are enzymes such as cell phosphatases. Examples of phosphate linker groups are

O-P(O)(ORk)-O-, -OP(S)(ORk)-O-, -ο-P(S) (SRk)-O-, -S-P(O)(ORk)-Ο- , -О-P(О)(ORk)-S-,

-S-P(O)(ORk)-S-, -O-P(S)(ORk)-S-, -S-P(S)(ORk)-O-, -O-P(O)(Rk)O- , -O-P(S)(Rk)-O-, -S-P(0)(Rk)-O-, -S-P(S)(Rk)-0-, -S-P(O)(Rk)S-, -O-P(S)(Rk)-S-.

The preferred embodiments are

-ο-P (O) (OH) -0-, -ο-P (S) (OH) -0-, -O-P (S) (SH) 0-, -S-P (O) (OH) -0- , -ο-P(O) (OH)-S-, -s-P(O) (OH)-S-, -O-P(S)(OH)S-, -S-P (S) (OH)-0 -, -ο-P (O) (H) -0-, -ο-P (S) (H) -0-, -s-P (O) (H) -0-, -s-P (S ) (H) -o-, -s-P (O) (H) -S-, -O-p (S) (H) -s-.

A preferred embodiment is -O-P(O)(OH)-O-. These candidates may be evaluated using methods similar to those described above.

- 48 043057

Acid cleavable linker groups.

The acid-cleavable linker groups that can be used in the dsRNA agent of the present invention are linker groups that are cleavable under acidic conditions. In preferred embodiments, acid-cleavable linker groups are cleaved in an acidic environment at a pH of about 6.5 or less (e.g., about 6.0, 5.5, 5.0 or less) or by means such as enzymes that can act as a universal acid. In the cell, certain low pH organelles, such as endosomes and lysosomes, can provide a medium for cleavage of acid-cleavable linker groups. Examples of acid-cleavable linker groups include, without limitation, hydrazones, esters, and amino acid esters. The acid-cleavable groups may be characterized by the general formula -C=NN-, C(O)O or -OC(O). A preferred embodiment is the case where the carbon atom attached to the oxygen of the ester group (in the alkoxy group) is part of an aryl group, a substituted alkyl group, or a quaternary alkyl group such as dimethylpentyl or tert-butyl. These candidates may be evaluated using methods similar to those described above.

Ester linker groups.

The ester cleavable linker groups that can be used in the dsRNA agent of the present invention are cleaved by the action of enzymes such as esterases and amidases of the cell. Examples of ester cleavable linker groups include, without limitation, esters with alkylene, alkenylene, and alkynylene groups. Ester cleavable linker groups are characterized by the general formula C(O)O- or -OC(O)-. These candidates may be evaluated using methods similar to those described above.

Peptide cleavable groups.

Peptide cleavable linker groups that can be used in a dsRNA tool according to the present invention are cleaved by the action of enzymes such as peptidases and proteases of the cell. Peptide cleavable linker groups are peptide bonds formed between amino acids to form oligopeptides (eg, dipeptides, tripeptides, etc.) and polypeptides. Peptide cleavable groups do not include the amide group (-C(O)NH-). The amide group may be formed between any alkylene, alkenylene or alkynylene. A peptide bond is a special type of amide bond formed between amino acids to form peptides and proteins. The peptide cleavable group is generally limited to the peptide bond (ie, amide bond) formed between amino acids to form peptides and proteins, and does not include the entire amide functional group. Peptide cleavable linker groups are characterized by the general formula NHCHR ^a C(O)NHCHR ^b C(O)-, where R ^A and R ^b are the R groups of two adjacent amino acids. These candidates may be evaluated using methods similar to those described above. As used herein, the term carbohydrate refers to a compound that is a carbohydrate per se formed from one or more monosaccharide units having at least 6 carbon atoms (which may form a linear, branched or cyclic structure) with an oxygen, nitrogen or sulfur atom associated with each carbon atom; or to a compound having as part of it a carbohydrate fragment formed from one or more monosaccharide units, each of which has at least six carbon atoms (which may form a linear, branched or cyclic structure) with an oxygen, nitrogen or sulfur atom linked with every carbon atom. Exemplary carbohydrates include sugars (mono-, di-, tri-, and oligosaccharides containing approximately 49 monosaccharide units) and polysaccharides such as starches, glycogen, cellulose, and polysaccharide gums. Specific monosaccharides include C ₅ and above (preferably Q-Cs) sugars; di- and trisaccharides include sugars with two or three monosaccharide units (preferably C ₅ -C ₈ ).

The present invention further relates to the use of a dsRNA agent as defined herein to suppress the expression of a target gene. According to one embodiment, the present invention further relates to the use of a dsRNA agent to suppress the expression of a target gene in vitro.

The present invention further relates to the use of a dsRNA agent as defined herein to suppress expression of a target gene in a subject. The subject may be any animal such as a mammal, such as a mouse, rat, sheep, cattle, dog, cat, or human.

In one embodiment, the dsRNA agent of the present invention is administered in a buffer.

In one embodiment, the siRNA compounds described herein can be formulated for administration to a subject. The siRNA-containing composition can be formulated to be in a variety of states. In some examples, the composition is at least partially crystalline, uniformly crystalline and/or

- 49 043057 anhydrous (eg contains less than 80%, 50%, 30%, 20% or 10% water). In another example, the siRNA is in an aqueous phase, such as in a solution that includes water.

The aqueous phase or crystalline compositions, for example, may be introduced into a delivery vehicle, eg a liposome (particularly in the case of an aqueous phase) or a particle (eg a microparticle, which may be suitable for a crystalline composition). In general, the siRNA-based composition is formulated in a manner that is compatible with the intended route of administration described herein. For example, in specific embodiments, the composition is prepared using at least one of the following methods: spray drying, lyophilization, vacuum drying, evaporation, fluid bed drying, or a combination of these techniques; or sonication with lipids, freeze-drying, condensation, and other self-assembly methods.

The siRNA preparation may be formulated in combination with another agent, eg, another therapeutic agent, or an agent that stabilizes the siRNA, eg, a protein that complexes with the siRNA to form an iRNP. Other agents include chelators such as EDTA (eg to remove divalent cations such as Mg ²⁺ ), salts, RNase inhibitors (eg RNase inhibitors with a broad spectrum of specificity such as RNasin®) and so on.

In one embodiment, the siRNA preparation comprises another siRNA compound, such as a second siRNA, that can mediate RNAi for a second gene or for the same gene. Another preparation may include at least 3, 5, ten, twenty, fifty or one hundred or more different types of siRNA. Such siRNAs can mediate RNAi for a similar number of different genes.

In one embodiment, the siRNA formulation comprises at least a second therapeutic agent (eg, a non-RNA or non-DNA agent). For example, an siRNA-based composition for treating a viral disease, such as HIV, may include a known antiviral agent (eg, a protease inhibitor or a reverse transcriptase inhibitor). In another example, an siRNA-based composition for cancer treatment may further comprise a chemotherapeutic agent.

Discussed below are exemplary formulations that can be used to administer a dsRNA agent in accordance with the present invention.

Liposomes.

For simplicity, the formulations, compositions, and methods in this section are primarily discussed in relation to compounds that are unmodified siRNA. However, it should be understood that these formulations, compositions, and methods may be practiced with other siRNA compounds, such as modified siRNAs, and such practice is within the scope of the present invention. A preparation based on a siRNA compound, e.g., a double-stranded siRNA compound, or a ssiRNA compound (e.g., based on a precursor, e.g., a larger siRNA compound that can be processed into a compound that is a ssiRNA, or DNA that encodes a siRNA compound, e.g., a double-stranded siRNA compound, or a ssiRNA compound, or a precursor thereof) can be formulated to be delivered in a membrane molecular complex, e.g., in a liposome or micelle. As used herein, the term liposome refers to a vesicle composed of amphiphilic lipids arranged in at least one bilayer, eg, one bilayer or multiple bilayers. Liposomes include single-layer and multilayer vesicles that have a membrane formed from a lipophilic material and an aqueous internal environment. The aqueous portion contains the siRNA-based composition. The lipophilic material separates the aqueous internal environment from the aqueous external environment, which generally does not include the siRNA-based composition, although in some examples it may. Liposomes are useful for carrying and delivering active ingredients to their site of action. Due to the fact that the liposome membrane is structurally similar to biological membranes, when liposomes are applied to tissue, the liposome bilayer merges with the bilayer of cell membranes. As the fusion of the liposome and the cell proceeds, the internal water content that contains the siRNA is delivered to the cell, where the siRNA can specifically bind to the target RNA and can mediate RNAi. In some cases, liposomes are also specifically targeted, for example, to direct siRNA to specific cell types.

A liposome containing an siRNA agent can be prepared in a number of ways. In one example, the lipid component of the liposome is dissolved in a detergent to form micelles from the lipid component. For example, the lipid component may be an amphipathic cationic lipid or a lipid conjugate. The detergent may have a high critical micelle concentration and may be non-ionic. Exemplary detergents include cholate, CHAPS, octylglucoside, deoxycholate, and lauroylsarcosine. The siRNA agent formulation is then added to the micelles, which include the lipid

- 50 043057 component. The cationic groups on the lipid interact with the siRNA and condense around the siRNA to form a liposome. After condensation, the detergent is removed, for example by dialysis, to obtain a siRNA-based liposome preparation.

If necessary, a carrier compound that promotes condensation can be added during the condensation reaction, for example by controlled addition. For example, the carrier compound may be a polymer other than a nucleic acid (eg, spermine or spermidine). You can also adjust the pH value to promote condensation.

A further description of methods for preparing stable polynucleotide delivery vehicles that include a polynucleotide/cationic lipid complex as structural components of the delivery vehicle is described, for example, in WO 96/37194. The formation of liposomes may also include one or more aspects of the exemplary methods described in the publication

Feigner, P. L. et al. Proc. Natl. Acad. Sc., USA 8:7413-7417,

1987; US Pat. No. 4,897,355; US patent No. 5171678; publications

Bangham, et al. M. Mol. Biol. 23:238, 1965; Olson, et al.

biochim. Biophys. Acta 557:9, 1979; Szoka, et al. Proc. Natl.

Acad. sci. 75:4194, 1978; Mayhew, et al. biochim. Biophys. Acta 775:169, 1984; Kim, et al. biochim. Biophys. Acta 728:339, 1983;

and Fukunaga, et al. Endocrinol. 115:757, 1984, which are incorporated by reference in their entirety. Commonly used techniques for obtaining lipid aggregates of suitable size for use as delivery vehicles include sonication and freeze-thaw extrusion (see, for example, Mayer, et al., Biochim. Biophys. Acta 858:161, 1986, incorporated by reference in its entirety). Microfluidization can be used where consistently small (50 to 200 nm) and relatively uniform aggregates are required (Mayhew, et al., Biochim. Biophys. Acta 775:169, 1984, incorporated by reference in its entirety). These methods are readily adaptable for packaging siRNA-based drugs into liposomes.

Liposomes that are pH sensitive or negatively charged trap nucleic acid molecules instead of complexing with them. Since both the nucleic acid molecules and the lipid have the same charge, repulsion occurs instead of complex formation. However, some nucleic acid molecules become trapped within the aqueous interior of these liposomes. pH-sensitive liposomes have been used to deliver DNA encoding the thymidine kinase gene to cell monolayers in culture. Exogenous gene expression was detected in target cells (Zhou et al., Journal of Controlled Release, 19, (1992) 269-274, which is incorporated by reference in its entirety).

One main type of liposome compositions includes phospholipids other than naturally occurring phosphatidylcholine. Compositions for neutral liposomes, for example, can be formed from dimyristoylphosphatidylcholine (DMPC) or dipalmitoylphosphatidylcholine (DPPC). Formulations for anionic liposomes are generally formed from dimyristoylphosphatidylglycerol, while anionic fusogenic liposomes are formed predominantly from dioleylphosphatidylethanolamine (DOPE). Another type of liposomal composition is formed from phosphatidylcholine (PC), such as, for example, soybean PC and egg PC. Another type is formed from mixtures of phospholipid and/or phosphatidylcholine and/or cholesterol.

Examples of other methods for introducing liposomes into cells in vitro are included in US patent No. 5283185; US patent No. 5171678; WO 94/00569; W093/24640; WO 91/16024; Feigner, J. Biol. Chem. 269:2550, 1994; Nabel, Proc. Natl. Acad. sci. 90:11307, 1993; Nabel, Human Gene Ther. 3:649, 1992; Gershon, Biochem. 32:7143, 1993; and Strauss EMBO J. 11:417, 1992.

In one embodiment, cationic liposomes are used. The advantage of cationic liposomes is that they are able to fuse with the cell membrane. Non-cationic liposomes, although not able to fuse as efficiently with the plasma membrane, are taken up by macrophages in vivo and can be used to deliver siRNA to macrophages.

Additional advantages of liposomes include the following: liposomes derived from natural phospholipids are biocompatible and biodegradable; a variety of water- and lipid-soluble drugs can be incorporated into liposomes; liposomes can protect siRNAs encapsulated in their internal compartments from metabolic conversion and degradation (Rosoff in Pharmaceutical Dosage Forms, Lieberman, Rieger and Banker (Eds.), 1988, volume 1, p. 245). Important factors to consider when preparing liposome formulations are lipid surface charge, vesicle size, and liposome water volume.

A positively charged synthetic cationic lipid, N-[1-(2,3dioleyloxy)propyl]-N,N,N-trimethylammonium chloride (DOTMA), can be used to form small liposomes that spontaneously react with nucleic acid to form lipid-nucleic acid complexes. acid that are able to fuse with negatively charged lipids in the cell membranes of tissue culture cells, resulting in siRNA delivery (see, for example, Feigner, P.

- 51 043057

L. et al., Proc. Natl. Acad. Sci., USA 8:7413-7417, 1987 and US Patent No. 4,897,355 for describing DOTMA and its use with DNA, which are incorporated by reference in their entirety).

The DOTMA analog, 1,2-bis-(oleoyloxy)-3-(trimethylammonium)propane (DOTAP), can be used in combination with a phospholipid to form vesicles complexed with DNA. Lipofectin™ (Bethesda Research Laboratories, Gaithersburg, Maryland) is an effective vehicle for delivering highly anionic nucleic acids to living tissue culture cells that contains positively charged DOTMA liposomes that spontaneously interact with negatively charged polynucleotides to form complexes. The net charge of the resulting complexes is also positive when liposomes with a sufficient positive charge are used. The positively charged complexes produced in this way spontaneously attach to negatively charged cell surfaces, fuse with the plasma membrane, and efficiently deliver functional nucleic acids, for example, to tissue culture cells. Another commercially available cationic lipid, 1,2-bis-(oleoyloxy)-3,3(trimethylammonium)propane (DOTAP) (Boehringer Mannheim, Indianapolis, IN), differs from DOTMA in that the oleyl moieties are linked by esters rather than ethers. connections.

Other published compounds with cationic lipids include those that have been conjugated to a number of moieties, including, for example, carboxyspermine, which has been conjugated to one of two types of lipids, and includes compounds such as 5-carboxyspermylglycine dioctaoleoilamide (DOGS) (Transfectam™, Promega, Madison, Wisconsin) and dipalmitoylphosphatidylethanolamine 5-carboxyspermylamide (DPPES) (see, for example, US Pat. No. 5,171,678).

Another cationic lipid conjugate involves lipid derivatization with cholesterol (DC-Choi) that has been formulated as liposomes in combination with DOPE (see, Gao, X. and Huang, L., Biochim. Biophys. Res. Commun. 179:280, 1991). Lipopolylysin obtained by conjugation of polylysine with DOPE has been reported to be effective for transfection in the presence of serum (Zhou, X. et al., Biochim. Biophys. Acta 1065:8, 1991, which is incorporated by reference in its entirety). In certain cell lines, such liposomes containing conjugated cationic lipids are reported to exhibit lower toxicity and provide more efficient transfection than DOTMA-containing formulations. Other commercially available cationic lipid products include DMRIE and DMRIE-HP (Vical, La Jolla, CA) and Lipofectamine (DOSPA) (Life Technology, Inc., Gaithersburg, MD). Other cationic lipids suitable for delivery of oligonucleotides are described in WO 98/39359 and WO 96/37194.

Liposomal formulations are particularly suitable for topical application, with liposomes having some advantages over other formulations. Such benefits include reduced side effects associated with extensive systemic absorption of the administered drug, increased accumulation of the administered drug at the desired target, and the ability to deliver siRNA to the skin. In some embodiments, liposomes are used to deliver siRNA to epidermal cells, as well as to improve the penetration of siRNA into dermal tissues, such as the skin. For example, liposomes can be applied topically. Topical delivery of drugs formulated as liposomes into the skin has been described (see, for example,

Weiner et al., Journal of Drug Targeting, 1992, vol. 2.405-410 and du

Plessis et al. Antiviral Research, 18, 1992, 259-265; Mannino, R. J. and Fould-Fogerite, S., Biotechniques 6:682-690, 1988;

Itani, T. et al. Gene 56:267-276. 1987; Nicolau, C. et al. Meth.

Enz. 149:157-176, 1987; Straubinger, R. M. and Papahadjopoulos,

D. Meth. Enz. 101:512-527, 1983; Wang, C. Y. and Huang, L.,

Proc. Natl. Acad. sci. USA 84:7851-7855, 1987, which are incorporated by reference in their entirety).

Non-ionic liposomal systems have also been investigated to determine their suitability for delivering drugs to the skin, in particular systems containing a non-ionic surfactant and cholesterol. Non-ionic liposome formulations containing Novasome I (glyceryl dilaurate/cholesterol/polyoxyethylene-10-stearyl ether) and Novasome II (glyceryl distearate/cholesterol/polyoxyethylene-10-stearyl ether) were used to deliver the drug to the dermal layer of the skin of mice. Such siRNA formulations are useful in the treatment of a dermatological disorder.

Liposomes that include siRNA can be made with high deformability. Such deformability may allow liposomes to penetrate through a pore that is smaller than the average liposome radius. For example, a type of deformable liposomes are transfersomes. Transferosomes can be prepared by adding surface boundary activators, typically surfactants, to a standard liposomal formulation. Transferosomes that include siRNA can be delivered, for example, subcutaneously by injection to deliver siRNA to keratinocytes

- 52 043057 leather. In order to pass through intact mammalian skin, lipid vesicles must pass through a series of small pores, each less than 50 nm in diameter, under the influence of a suitable transdermal gradient. In addition, due to the properties of lipids, such transferosomes can be self-optimizing (adapting to the shape of pores, for example, in the skin), self-healing, and often can reach their targets without fragmenting, and can often be self-loading.

Other formulations useful in the present invention are described in US Provisional Application Serial Nos. 61/018616, filed January 2, 2008; 61/018611, filed January 2, 2008; 61/039748, filed March 26, 2008; 61/047087, filed April 22, 2008, and 61/051528, filed May 8, 2008. PCT Application No. PCT/US2007/080331, filed October 3, 2007, also describes formulations useful in the present invention.

Surfactants.

For simplicity, the formulations, compositions, and methods in this section are primarily discussed in relation to compounds that are unmodified siRNA. However, it should be understood that these formulations, compositions, and methods may be practiced with other siRNA compounds, such as modified siRNA compounds, and such practice is within the scope of the present invention. Surfactants find wide use in formulations such as emulsions (including microemulsions) and liposomes (see above). Compositions based on siRNA (or a precursor, for example, larger dsiRNAs that can be processed into siRNAs, or DNAs that encode siRNAs or a precursor) may include a surfactant. In one embodiment, the siRNA is formulated as an emulsion that includes a surfactant. The most common way to classify and rank the properties of many different types of surfactants, both natural and synthetic, is by using the hydrophilic-lipophilic balance (HLB). The nature of the hydrophilic group provides the most efficient means of categorizing the various surfactants used in formulations (Rieger, in Pharmaceutical Dosage Forms, Marcel Dekker, Inc., New York, NY, 1988, p. 285).

If the surfactant molecule is not ionized, then it is classified as a nonionic surfactant. Nonionic surfactants find wide use in pharmaceutical products and can be used over a wide pH range. Typically, their HLB values range from 2 to about 18, depending on their structure. Nonionic surfactants include nonionic esters such as ethylene glycol esters, propylene glycol esters, glyceryl esters, polyglyceryl esters, sorbitan esters, sucrose esters, and ethoxylated esters. Nonionic alkanoamides and ethers, such as fatty alcohol ethoxylates, propoxylated alcohols and ethoxylated/propoxylated block copolymers, are also included in this class. Polyoxyethylene surfactants are the most common members of the nonionic surfactant class.

If a surfactant molecule carries a negative charge when dissolved or dispersed in water, then the surfactant is classified as anionic.

Anionic surfactants include carboxylates such as saponifiers, acyl lactylates, amino acid acylamides, sulfuric acid esters such as alkyl sulfates and ethoxylated alkyl sulfates, sulfonates such as alkyl benzene sulfonates, acyl isethionates, acyl taurates and sulfosuccinates, and phosphates. The most important members of the class of anionic surfactants are alkyl sulfates and saponifiers.

If the surfactant molecule carries a positive charge when dissolved or dispersed in water, then the surfactant is classified as cationic. Cationic surfactants include quaternary ammonium salts and ethoxylated amines. Quaternary ammonium salts are the most used representatives of this class.

If a surfactant molecule has the ability to carry both a positive and a negative charge, then the surfactant is classified as amphoteric. Amphoteric surfactants include acrylic acid derivatives, substituted alkyl amides, N-alkyl betaines, and phosphatides.

The use of surfactants in formulations, formulations and emulsions has been reviewed (Rieger, in Pharmaceutical Dosage Forms, Marcel Dekker, Inc., New York, NY, 1988, p. 285).

Micelles and other membrane compounds.

For simplicity, micelles and other formulations, compositions and methods in this section are mainly discussed in relation to compounds that are unmodified siRNA. However, it should be understood that micelles and other formulations, compositions, and methods may be practiced with other siRNA compounds, such as compounds

- 53 043057 which are modified siRNAs, and such practice is within the scope of the present invention. A composition based on a siRNA compound, e.g., a double-stranded siRNA compound, or a ssiRNA compound (e.g., based on a precursor, e.g., a larger siRNA compound that can be processed into a compound that is a ssiRNA, or DNA that encodes a siRNA compound, eg, a double-stranded siRNA compound, or a ssiRNA compound, or a precursor thereof) can be presented as a micellar formulation. Micelles are defined herein as a particular type of molecular assembly in which the amphipathic molecules are arranged in a spherical structure such that all the hydrophobic portions of the molecules are directed inwards, leaving the hydrophilic portions in contact with the surrounding aqueous phase. The opposite arrangement occurs if the environment is hydrophobic.

A mixed micellar formulation suitable for transdermal membrane delivery can be prepared by mixing an aqueous solution of the siRNA based formulation, C ₈ C22 alkali metal alkyl sulfate and micelle forming compounds. Exemplary micelle forming compounds include lecithin, hyaluronic acid, pharmaceutically acceptable salts of hyaluronic acid, glycolic acid, lactic acid, chamomile extract, cucumber extract, oleic acid, linoleic acid, linolenic acid, monoolein, monooleates, monolaurates, borage oil, evening primrose oil. , menthol, trihydroxyoxocholanyl-glycine and its pharmaceutically acceptable salts, glycerol, polyglycerol, lysine, polylysine, triolein, polyoxyethylene ethers and analogues, polydocanol-alkyl ethers and analogues, chenodeoxycholate, deoxycholate and mixtures thereof. The micelle-forming compounds can be added simultaneously or after the addition of the alkali metal alkyl sulfate. Mixed micelles will form with almost any type of ingredient mixing, but smaller micelles require vigorous mixing.

According to one method, a first micellar composition is prepared which contains an siRNA-based composition and at least an alkali metal alkyl sulfate. The first micellar composition is then mixed with at least three micelle-forming compounds to form a mixed micellar composition. According to another method, a micellar composition is prepared by mixing the composition based on siRNA, alkali metal alkyl sulfate and at least one of the micelle forming compounds, followed by adding the remaining micelle forming compounds with vigorous stirring.

Phenol and/or m-cresol can be added to the mixed micellar composition to stabilize the composition and protect against bacterial growth. Alternatively, phenol and/or m-cresol can be added with micelle forming ingredients. An isotonic agent such as glycerin may also be added after the mixed micellar composition has been formed.

To deliver the micellar formulation as a sprayable solution, the formulation can be placed in an aerosol dispenser and the sprayer charged with a propellant. The propellant, which is under pressure, is in liquid form in the nebulizer. The ratios of the ingredients are adjusted so that the aqueous phase and the propellant phase become one, i.e. there is one phase. If two phases are present, it is necessary to shake the atomizer before spraying a portion of the contents, for example, through a dosing valve. The nebulized dose of the pharmaceutical agent is forced out of the metering valve in the form of a fine aerosol.

Propellants may include hydrogen-containing chlorofluorocarbons, hydrogen-containing fluorocarbons, dimethyl ether, and diethyl ether. In certain embodiments, HFA 134a (1,1,1,2-tetrafluoroethane) can be used.

Specific concentrations of the main ingredients can be determined by relatively simple experiments. Absorption through the oral cavity often requires an increase, for example, at least doubling or tripling, the dose intended for injection or administration through the gastrointestinal tract.

Particles.

For simplicity, the particles, formulations, compositions, and methods in this section are primarily discussed in relation to compounds that are modified siRNA. However, it should be understood that these particles and other formulations, compositions and methods may be practiced with other siRNA compounds, such as unmodified siRNA compounds, and such practice is within the scope of the present invention. In another embodiment, preparations based on a siRNA compound, e.g., a double-stranded siRNA compound, or an ssiRNA compound (e.g., based on a precursor, e.g., a larger siRNA compound that can be processed into a compound that is ssiRNA, or DNA that encodes a compound that is siRNA, for example, a compound that is a double-stranded siRNA, or a compound that is

- 54 043057 ssiRNA, or their precursor) can be introduced into a particle, for example, into a microparticle. The microparticles can be obtained by spray drying, but can also be obtained by other methods, including lyophilization, evaporation, fluid bed drying, vacuum drying, or a combination of these techniques.

pharmaceutical compositions.

The RNA interference agents of the present invention may be formulated for pharmaceutical use. The present invention further relates to a pharmaceutical composition containing a dsRNA agent as defined herein. Pharmaceutically acceptable compositions contain a therapeutically effective amount of one or more of the dsRNA agents according to any of the preceding embodiments, taken alone or formulated with one or more pharmaceutically acceptable carriers (additives), excipients and/or diluents.

In particular, pharmaceutical compositions may be formulated for administration in solid or liquid form, including compositions that are adapted to:

(1) oral administration, eg liquid dosage forms (aqueous or non-aqueous solutions or suspensions), tablets, eg tablets intended for buccal, sublingual and systemic absorption, boluses, powders, granules, pastes for application to the tongue;

(2) parenteral administration, for example, by subcutaneous, intramuscular, intravenous or epidural injection, such as a sterile solution or suspension, or a sustained release formulation;

(3) topical application, for example, as a controlled release cream, ointment or patch applied to the skin or spray solution;

(4) intravaginal or intrarectal administration, for example, in the form of a pessary, cream or foam;

(5) sublingual administration;

(6) ophthalmic administration;

(7) transdermal or (8) intranasal administration. Particularly preferred may be delivery using subcutaneous or intravenous routes of administration.

As used herein, the phrase "therapeutically effective amount" means that amount of a compound, material, or composition containing a compound of the present invention that is effective to produce some desired therapeutic effect in at least a subset of cells in an animal at an acceptable benefit/risk ratio applicable to any therapeutic treatment.

The phrase pharmaceutically acceptable is used herein to refer to those compounds, materials, compositions and/or dosage forms which, based on careful medical evaluation, are suitable for use in contact with human and animal tissues without undue toxicity, irritation, allergic reaction, or other problems or complications. commensurate with an acceptable benefit/risk ratio.

As used herein, the phrase "pharmaceutically acceptable carrier" means a pharmaceutically acceptable material, composition, or medium such as a liquid or solid filler, diluent, excipient, manufacturing aid (e.g., lubricant, talc, magnesium, calcium or zinc stearate, or stearic acid). acid) or solvent encapsulation material involved in the transfer or transport of the compound in question from one organ or body part to another organ or body part. Each carrier must be acceptable in the sense that it must be compatible with the other ingredients of the formulation and not be harmful to the patient. Some examples of materials that may serve as pharmaceutically acceptable carriers include:

(1) sugars such as lactose, glucose and sucrose;

(2) starches such as corn starch and potato starch;

(3) cellulose and its derivatives such as sodium carboxymethyl cellulose, ethyl cellulose and cellulose acetate;

(4) powdered tragacanth;

(5) malt;

(6) gelatin;

(7) lubricants such as magnesium stearate, sodium lauryl sulfate and talc;

(8) fillers such as cocoa butter and suppository waxes;

(9) oils such as peanut oil, cottonseed oil, safflower oil, sesame oil, olive oil, corn oil and soybean oil;

(10) glycols such as propylene glycol;

(11) polyols such as glycerol, sorbitol, mannitol and polyethylene glycol;

(12) esters such as ethyl oleate and ethyl laurate;

- 55 043057 (13) agar;

(14) buffering agents such as magnesium hydroxide and aluminum hydroxide;

(15) alginic acid;

(16) pyrogen-free water;

(17) isotonic saline solution;

(18) Ringer's solution;

(19) ethyl alcohol;

(20) pH buffer solutions;

(21) polyesters, polycarbonates and/or polyanhydrides;

(22) bulking agents such as polypeptides and amino acids;

(23) a blood serum component such as serum albumin, HDL and LDL; and (22) other non-toxic compatible substances used in pharmaceutical formulations.

For convenience, the formulations may be presented in unit dosage form and may be prepared by any of the methods well known in the art of pharmacy. The amount of active ingredient that can be combined with the carrier material to form a unit dosage form will vary depending on the recipient to be treated, the particular route of administration. The amount of active ingredient that can be combined with the carrier material to form a unit dosage form will usually be the amount of the compound that provides a therapeutic effect. Typically, on a one hundred percent basis, this amount will range from about 0.1% to about ninety-nine percent active ingredient, preferably from about 5% to about 70%, most preferably from about 10% to about 30%.

In certain embodiments, the composition of the present invention comprises an excipient selected from the group consisting of cyclodextrins, celluloses, liposomes, micelle forming agents, eg bile acids, and polymeric carriers, eg polyesters and polyanhydrides; and a compound of the present invention. In certain embodiments, the above formulation imparts oral bioavailability to a compound of the present invention.

An RNAi agent preparation may be formulated in combination with another agent, such as another therapeutic agent, or an agent that confers stability to the RNAi agent, such as a protein that combines with the RNAi agent to form an iRNP. Other agents include chelators such as EDTA (eg to remove divalent cations such as Mg ²⁺ ), salts, RNase inhibitors (eg RNase inhibitors with a broad spectrum of specificity such as RNasin®) and so on.

Methods for preparing these formulations or compositions include the step of bringing into association the compound of the present invention with the carrier and optionally with one or more accessory ingredients. Generally, formulations are prepared by uniformly and intimately combining a compound of the present invention with liquid carriers or finely divided solid carriers or both and then shaping the product if necessary.

In some cases, to prolong the effect of the drug, it is required to slow down the absorption of the drug when administered subcutaneously or intramuscularly. This can be achieved by using a liquid suspension based on a crystalline or amorphous material having poor water solubility. In addition, the rate of absorption of the drug depends on its rate of dissolution, which, in turn, may depend on the size of the crystal and on the crystalline form. As an alternative, delayed absorption of the parenterally administered form of the drug is achieved by dissolving or suspending the drug in an oil vehicle.

The compounds of the present invention may be formulated for administration by any conventional route for human or veterinary use in a manner similar to other pharmaceutical preparations.

The term treatment is also intended to cover prevention, therapy and cure. The patient receiving this type of treatment is any animal requiring it, including primates, in particular humans, as well as other mammals such as horses, cattle, pigs and sheep; as well as poultry and pets in a general sense.

Double-stranded RNAi agents are generated in the cell in vivo, for example, based on exogenous DNA templates that are delivered to the cell. For example, DNA templates can be inserted into vectors and used as gene therapy vectors. Gene therapy vectors can be delivered to a subject, for example, by intravenous injection, topical administration (US Pat. No. 5,328,470, which is incorporated by reference in its entirety), or stereotaxic injection (see, for example, Chen et al. (1994) Proc. Natl Acad. Sci. USA 91:3054-3057, which is incorporated by reference in its entirety). The gene therapy vector pharmaceutical formulation may include the gene therapy vector in an acceptable diluent, or may contain a template for delayed

- 56 043057 release, which contains a gene delivery vehicle. DNA templates may include, for example, two transcription units, one of which generates a transcript that includes the top strand of the dsRNA agent, and the other generates a transcript that includes the bottom strand of the dsRNA agent. When the templates are transcribed, a dsRNA agent is formed and is processed into siRNA agent fragments that mediate gene silencing.

Shipping ways.

A dsRNA agent, as defined herein, or a pharmaceutical composition containing a dsRNA agent, as defined herein, can be administered to a subject using different delivery routes. A composition comprising an RNA interference agent may be delivered to a subject via a variety of routes. Exemplary routes include: intravenous, subcutaneous, topical, rectal, anal, vaginal, intranasal, pulmonary, ocular.

The RNA interference molecules and/or dsRNA agent of the present invention may be included in pharmaceutical compositions suitable for administration. Typically, such compositions include one or more types of RNAi agents and a pharmaceutically acceptable carrier. As used herein, the formulation of a pharmaceutically acceptable carrier is intended to include any solvents, dispersion media, coatings, antibacterial and antifungal agents, isotonic agents and absorption delay agents, and the like, that are compatible with use in pharmaceutical preparations. The use of such media and agents for pharmaceutically active substances is well known in the art. Unless any standard medium or agent is incompatible with the active compound, their use in the compositions is contemplated. Additional active compounds may also be included in the compositions.

The compositions of the present invention can be administered in a variety of ways depending on whether topical or systemic treatment is needed and on the site to be treated. Administration may be topical (including ophthalmic, vaginal, rectal, intranasal, transdermal), oral or parenteral. Parenteral administration includes intravenous drip, subcutaneous, intraperitoneal or intramuscular injection, or intrathecal or intraventricular administration.

The route and site of administration can be chosen to increase targeted exposure. For example, to target muscle cells, the logical choice would be intramuscular injection into the muscles of interest. It is possible to purposefully influence lung cells by introducing an RNA interference agent in the form of an aerosol. It is possible to purposefully influence vascular endothelial cells by applying an agent for RNA interference to a balloon catheter and mechanically introducing DNA.

Dosage.

In one aspect, the present invention provides a method of administering a dsRNA agent, eg, an siRNA agent, to a subject (eg, a human subject). According to another aspect, the present invention further relates to a dsRNA agent as defined herein for use in suppressing the expression of a target gene in a subject. The method or medical use comprises administering a single dose of a dsRNA agent, e.g., an siRNA agent, e.g., a double-stranded siRNA agent, wherein:

(a) the length of the double-stranded part of which is 14-40 nucleotides (nt), for example, 21-23 nt, (b) which is complementary to the target RNA (for example, endogenous or target RNA of the pathogen), and optionally (c) which includes at least one 3' overhang, 1-5 nucleotides long.

In one embodiment, the single dose is less than 10 mg per 1 kg of body weight, or less than 10.5, 2.1, 0.5, 0.1, 0.05, 0.01, 0.005, 0.001, 0.0005, 0 .0001, 0.00005 or 0.00001 mg per 1 kg of body weight, and less than 200 nmol of an RNA agent (for example, approximately 4.4 x 10 ¹⁶ copies) per 1 kg of body weight or less than 1500, 750, 300, 150, 75, 15, 7.5, 1.5, 0.75, 0.15, 0.075, 0.015, 0.0075, 0.0015, 0.00075, 0.00015 nmol of RNA agent per 1 kg of body weight.

The given amount may be an amount that is effective to treat or prevent a disease or disorder, such as a disease or disorder associated with the target RNA. For example, a single dose can be administered by injection (eg, intravenous, subcutaneous, or intramuscular), as an inhaled dose, or by topical application. In some embodiments, doses may be less than 10, 5, 2, 1, or 0.1 mg/kg of body weight.

In some embodiments, the single dose is administered at a frequency of less than once per day, such as less than once every 2, 4, 8, or 30 days. In another embodiment, the single dose is not administered at a specific frequency (eg, irregular frequency). For example, a single dose may be administered once.

In one embodiment, the effective dose is administered along with other conventional

- 57 043057 therapeutic effects. In one embodiment, the subject has a viral infection, wherein the treatment is the use of an antiviral agent other than the dsRNA agent, eg other than the siRNA agent. In another embodiment, the subject has atherosclerosis, wherein an effective dose of a dsRNA agent, eg, an siRNA agent, is administered in combination, eg, after surgery, eg, angioplasty.

In one embodiment, the subject is administered an initial dose and one or more maintenance doses of a dsRNA agent, e.g., an siRNA agent (e.g., a precursor, e.g., a larger dsRNA agent that can be processed into a siRNA, or DNA that encodes a dsRNA agent, such as a siRNA agent, or a precursor thereof). The maintenance dose or doses may be the same or lower than the initial dose, eg, half the initial dose. The maintenance regimen may include treating the subject with a dose or doses ranging from 0.01 μg to 15 mg/kg of body weight per day, for example, 10, 1, 0.1, 0.01, 0.001, or 0.00001 mg per 1 kg of body weight per day. For example, maintenance doses are administered no more than once every 2, 5, 10, or 30 days. In addition, the treatment regimen may be continued for a period of time, which will vary depending on the nature of the particular disease, its severity and the general condition of the patient. In certain embodiments, the dose may be delivered no more than once a day, such as no more than once every 24, 36, 48 or more hours, such as no more than once every 5 or 8 days. After treatment, the patient is monitored for changes in his condition and for relief of the symptoms of the disease state. The dose of the compound may be increased if the patient is sufficiently responsive to the current dose levels, or the dose may be reduced if relief of symptoms of the disease state is observed, if the disease state has been relieved, or if undesirable side effects are observed.

An effective dose may be administered as a single dose or as two or more doses as needed or as deemed appropriate under the circumstances. If repeated or frequent infusions are to be facilitated, implantation of a delivery device, such as a pump, a semi-permanent stent (eg, intravenous, intraperitoneal, intracisternal, or intraarticular) or reservoir, may be appropriate.

In one embodiment, the composition includes a plurality of types of dsRNA agents. In another embodiment, a species of dsRNA agent has sequences that are non-overlapping and non-contiguous with sequences of another species, given the naturally occurring target sequence. In another embodiment, the plurality of dsRNA agents are specific for different naturally occurring target genes. In another embodiment, the dsRNA agent is allele specific.

The dsRNA agents of the present invention described herein can be administered to mammals, in particular large mammals such as non-human primates or humans, via a variety of routes.

In one embodiment, a composition based on a dsRNA agent, e.g., an siRNA agent, is administered by parenteral, e.g., intravenous (e.g., bolus or diffusing infusion), intradermal, intraperitoneal, intramuscular, intrathecal, intraventricular, intracranial, subcutaneous, transmucosal, buccal, sublingual, endoscopic, rectal, oral, vaginal, topical, pulmonary, intranasal, urethral, or ocular routes. The introduction can be carried out by the subject or another person, for example, a medical professional. The drug may be provided as metered doses or in a dispenser that delivers a metered dose. The selected modes of delivery are discussed in more detail below.

The present invention provides methods, compositions, and kits for rectal administration or delivery of the dsRNA agents described herein.

In specific embodiments, the present invention relates to the dsRNA agents of the present invention for use in the methods described above.

Methods for suppressing target gene expression.

Embodiments of the present invention also relate to methods for suppressing the expression of a target gene. The method includes the step of administering the dsRNA agents according to any of the previous embodiments in an amount sufficient to suppress expression of the target gene. The present invention further relates to the use of a dsRNA agent as defined herein to suppress the expression of a target gene in a target cell. According to a preferred embodiment, the present invention

- 58 043057 further relates to the use of a dsRNA agent to suppress the expression of a target gene in a target cell in vitro.

According to another aspect, the present invention relates to a method for modulating the expression of a target gene in a cell, comprising delivering a dsRNA agent of the present invention to said cell. In one embodiment, the target gene is selected from the group consisting of factor VII, Eg5, PCSK9, TPX2, apoB, SAA, TTR, RSV, PDGF-beta gene, Erb-B gene, Src gene, CRK gene, GRB2 gene, RAS gene , MEKK gene, JNK gene, RAF gene, Erk1/2 gene, PCNA(p21) gene, MYB gene, JUN gene, FOS gene, BCL-2 gene, hepcidin, activated protein C, cyclin D gene, VEGF gene, EGFR gene , cyclin A gene, cyclin E gene, WNT-1 gene, beta-catenin gene, cMET gene, PKC gene, NFkB gene, STAT3 gene, survivin gene, Her2/Neu gene, topoisomerase I gene, topoisomerase II alpha gene, mutations in p73 gene, mutations in the p21(WAF1/CIP1) gene, mutations in the p27(K1P1) gene, mutations in the PPM1D gene, mutations in the RAS gene, mutations in the caveolin I gene, mutations in the MIB I gene, mutations in the MTAI gene, mutations in the M68 gene, mutations in tumor suppressor genes, and mutations in the p53 tumor suppressor gene.

In specific embodiments, the present invention relates to the dsRNA agents of the present invention for use in the methods described above.

The invention is further illustrated by the following examples, which should not be construed as further limiting. The contents of all references, pending patent applications and published patents cited throughout this application are hereby expressly incorporated by reference.

Examples

Example 1 In vitro screening for siRNA duplexes.

Cell culture and transfection.

Human Hep3B cells or rat H.II.4.E cells (ATCC, Manassas, VA) were grown to near confluence at 37°C in 5% CO2 in RPMI (ATCC) supplemented with 10% FBS, streptomycin, and glutamine (ATCC) , prior to separation from the culture dish by treatment with trypsin. Transfection was performed by adding 14.8 µl of Opti-MEM plus 0.2 µl of Lipofectamine RNAiMax per well (Invitrogen, Carlsbad, CA, cat # 13778-150) to 5 µl of siRNA duplexes per well in a 96-well plate and incubated at room temperature for 15 min. 80 μl of complete culture medium without antibiotic containing ~2x10 ⁴ Hep3B cells was then added to the siRNA mixture. Cells were incubated for either 24 h or 120 h before RNA purification. Single dose experiments were performed at a final duplex concentration of 10 nM and 0.1 nM, and dose effect experiments were performed using 8 consecutive 4-fold serial dilutions, with a maximum dose of a final duplex concentration of 10 nM.

Isolation of total RNA using the DYNABEADS mRNA Isolation Kit (Invitroqen, Cat # 610-12).

Cells were harvested and lysed in 150 μl of lysis/binding buffer, then mixed for 5 min at 850 rpm using an Eppendorf Thermomixer (mixing speed was the same throughout the process). Ten microliters of magnetic beads and 80 µl of lyse/binding buffer mixture were added to a round bottom plate and mixed for 1 minute. The magnetic beads were held with a magnetic device and the supernatant was removed without displacement of the beads. After removing the supernatant, the lysed cells were added to the remaining beads and mixed for 5 min. After the supernatant was removed, the magnetic beads were washed 2 times with 150 µl wash buffer A and stirred for 1 min. The beads were again held and the supernatant was removed. The beads were then washed with 150 μl of wash buffer B, retained, and the supernatant discarded. In the next step, the beads were washed with 150 μl of elution buffer, retained and the supernatant was discarded. The granules were left to dry for 2 minutes. After drying, 50 µl of elution buffer was added and stirred for 5 min at 70°C. The pellets were held with a magnet for 5 minutes. 40 μl of the supernatant was removed and added to another 96-well plate.

cDNA synthesis using the ABI High capacity cDNA reverse transcription kit (Applied Biosvstems, Forster City, CA, Cat # 4368813).

A master mix of 1 µl 10x buffer, 0.4 µl 25x dNTP, 1 µl random primers, 0.5 µl reverse transcriptase, 0.5 µl RNase inhibitor and 1.6 µl H ₂ O per reaction was added to 5 µl total RNA . cDNA was generated using a Bio-Rad C-1000 or S-1000 thermal cycler (Hercules, CA) through the following steps: 25°C 10 min, 37°C 120 min, 85°C 5 s, storage at 4°C.

PCR in real time.

µl cDNA was added to a master mix containing 0.5 µl TaqMan probe for GAPDH (Applied Biosystems, Cat # 4326317E (human), Cat # 4308313 (rodent)), 0.5 µl TaqMan probe for TTR (Applied Biosystems cat # HS00174914_m1 (human), cat # Rn00562124_m1 (rat)) and 5 µl

- 59 043057 master mix with Lightcycler 480 probe (Roche, p/n 04887301001) per well in a 384-well plate (Roche, p/n 04887301001). Real-time PCR was carried out in a device for

Real-time PCR Roche LC480 (Roche). Each duplex was tested in at least two independent transfections and each transfection was evaluated in duplicate unless otherwise noted.

To calculate relative fold change, real-time data were analyzed using the AACt method and normalized to analyzes performed with cells transfected with 10 nM AD-1955 or mock-transfected cells. IC ₅₀ was calculated using a 4-way fit model using XLFit and normalized to cells transfected with AD-1955 or naive cells at the same dose range, or to self lowest dose. IC ₅₀ values were calculated for each individual transfection as well as for the combination, with the individual IC ₅₀ value consistent with both transfections.

The results of gene silencing by an exemplary siRNA duplex with various motif modifications of the present invention are shown in the table below.

Example 2 Synthesis of RNA and hybridization of duplexes.

1. Synthesis of oligonucleotides.

All oligonucleotides were synthesized on an AKTAoligopilot synthesis unit or on an ABI 394 synthesis unit. Unless otherwise noted, commercially available controlled pore glass solid support (dT-CPG, 500A, Prime Synthesis) and RNA phosphoramidites with standard protecting groups, 5'-O-dimethoxytrityl-N6-benzoyl-2'-tert-butyldimethylsilyl-adenosine-3'-O-N,N'-diisopropyl-2-cyanoethylphosphoramidite, 5'-O-dimethoxytritylN4-acetyl-2'-tert- butyldimethylsilylcytidine-3'-O-N,N'-diisopropyl-2-cyanoethylphosphoramidite, 5'-Odimethyloxytrityl-N2-isobutryl-2'-tert-butyldimethylsilyl-guanosine-3'-O-N,N'-diisopropyl-2cyanoethylphosphoramidite and 5' -O-dimethoxytrityl-2'-tert-butyldimethylsilyl-uridine-3'-O-N,N'diisopropyl-2-cyanoethylphosphoramidite (Pierce Nucleic Acids Technologies). 2'-P-phosphoramidites, 5'-Odimethoxytrityl-N4-acetyl-2'-fluoro-cytidine-3'-O-N,N'-diisopropyl-2-cyanoethyl-phosphoramidite and 5'-Odimethoxytrityl-2'-fluorouridine-3 '-O-N,N'-diisopropyl-2-cyanoethyl-phosphoramidite was purchased from Promega. All phosphoramidites were used at 0.2M in acetonitrile (CH3CN), except for guanosine, which was used at 0.2M in 10% THF/ANC (v/v). The binding/recycle time was 16 minutes. The activator is 5-ethylthiotetrazole (0.75 M, American International Chemicals); iodine/water/pyridine was used for PO oxidation and PADS (2%) in 2,6-lutidine/ACN (1:1 v/v) was used for PS oxidation.

Ligand-conjugated strands were synthesized using a solid support containing the appropriate ligand. For example, the introduction of a carbohydrate moiety/ligand (eg, for GalNAc) at the 3' end of the sequence was performed by starting the synthesis with a solid support with the appropriate carbohydrate. Similarly, the cholesterol fragment at the 3'-end was introduced, starting the synthesis on a support with cholesterol. Typically, the ligand moiety was coupled to trans-4hydroxyprolinol via a chosen bond, as described in the preceding examples, to form a hydroxyprolinol ligand moiety. The hydroxyprolinol ligand fragment was then coupled to a solid support via a succinate linker or converted to phosphoramidite using standard phosphorylation conditions to obtain the desired carbohydrate conjugate building blocks. Fluorophore labeled siRNAs were synthesized on a solid support with the appropriate phosphoroamidite purchased from Biosearch Technologies. An oleyllithocholic acid (GalNAc) ₃ polymer support was prepared in the laboratory at a loading of 38.6 µmol/g. The mannose (Man) ₃ polymer support was made in the laboratory at a loading of 42.0 µmol/g.

Conjugation of the selected ligand at the desired position, eg at the 5' end of the sequence, is accomplished by coupling the appropriate phosphoramidite to the growing chain under standard phosphoramidite coupling conditions, unless otherwise indicated. The binding of a 0.1 M solution of phosphoramidite in anhydrous CH3CN in the presence of the 5-(ethylthio)-1H-tetrazole activator to the oligonucleotide bound on a solid support was performed for 15 min. Oxidation of internucleotide phosphite to phosphate was performed using a standard iodine-water mixture as described (1), or by treatment with a mixture of tert-butyl hydroperoxide/acetonitrile/water (10:87:3), while the duration of oxidation of the conjugated oligonucleotide was 10 min. Phosphorothioate was introduced by oxidizing phosphite to phosphorothioate using a sulfur transfer reagent such as DDTT (purchased from AM Chemicals), PADS and/or Bocage's reagent. Cholesterol-phosphoramidite was synthesized in the laboratory and used at a concentration of 0.1 M in dichloromethane. The binding time for cholesterol-phosphoramidite was 16 minutes.

2. Deprotection-I (deprotection of the nucleic base).

After completion of the synthesis, the substrate was transferred to a 100 ml glass vial (VWR). The oligonucleotide was separated from the support while deprotecting the base and phosphate groups with 80 ml of alcohol ammonia [ammonia:ethanol (3:1)] for 6.5 hours at 55°C. Vial fast

- 60 043057 was cooled on ice, and then the mixture based on alcohol ammonia was filtered into a new 250 ml vial.

The CPG was washed with 2x40 ml portions of ethanol/water (1:1 v/v). The volume of the mixture was then reduced to ~30 ml using a rotary evaporator. The mixture was then frozen on dry ice and dried under vacuum on a high speed vacuum unit.

3. Deprotection II (removal of the 2'-TBDMS group) The dried residue was taken up in 26 ml of a mixture of triethylamine, triethylamine trihydrofluoride (TEA.3HF) or pyridine-HF and DMSO (3:4:6) and heated at 60°C for 90 min to remove tert-butyldimethylsilyl (TBDMS) groups at the 2' position. The reaction mixture was then quenched with 50 ml of 20 mM sodium acetate and the pH was adjusted to 6.5 while stored in a freezer until purified.

4. Analysis.

Before purification, the oligonucleotides were analyzed by high performance liquid chromatography (HPLC) and the choice of buffer and column depended on the nature of the sequence and/or conjugated ligand.

5. Purification with HPLC.

The ligand-conjugated oligonucleotides were purified by reverse phase preparative HPLC. Unconjugated oligonucleotides were purified by anion exchange HPLC on a laboratory packed TSKgel column. Buffers were 20 mM sodium phosphate (pH 8.5) in 10% CH3CN (buffer A) and 20 mM sodium phosphate (pH 8.5) in 10% CH3CN, 1M NaBr (buffer B). Fractions containing full length oligonucleotides were pooled, desalted and lyophilized. Approximately 0.15 OD of desalted oligonucleotides were diluted in water to 150 μl and then pipetted into special vials for CGE and LC/MS analysis. Finally, compounds were analyzed by LC-ESMS and CGE.

6. Obtaining siRNA.

To obtain siRNA, equimolar amounts of sense and antisense strand were heated in 1xPBS at 95°C for 5 min and slowly cooled to room temperature. Duplex integrity was confirmed by HPLC analysis.

Example 3. In vitro silencing activity using siRNA ANGPTL3 with various chemical modifications.

Cell culture and transfection.

Hep3B cells (ATCC, Manassas, VA) were grown to near confluence at 37° C. in 5% CO ₂ in RPMI (ATCC) supplemented with 10% FBS, streptomycin and glutamine (ATCC) until detached from the culture dish by trypsinization. Transfection was performed by adding 14.8 µl of Opti-MEM plus 0.2 µl of Lipofectamine RNAiMax per well (Invitrogen, Carlsbad, CA, cat # 13778-150) to 5 µl of siRNA duplexes per well in a 96-well plate and incubated at room temperature for 15 min, 80 µl of complete culture medium without antibiotic containing ~2x10 ⁴ Hep3B cells was then added to the siRNA mixture. Cells were incubated for either 24 h or 120 h before RNA purification. Unless otherwise noted, single dose experiments were performed at a final duplex concentration of 10 nM and 0.1 nM, and dose effect experiments were performed at a final duplex concentration of 10, 1, 0.5, 0.1, 0.05, 0.01, 0.005, 0.001, 0.0005, 0.0001, 0.00005 and 0.00001 nM.

cDNA synthesis using the ABI High capacity cDNA reverse transcription kit (Applied Biosystems, Foster City, CA, Cat # 4368813).

A master mix of 2 µl 10x buffer, 0.8 µl 25x dNTP, 2 µl random primers, 1 µl reverse transcriptase, 1 µl RNase inhibitor and 3.2 µl H2O per reaction was added to 10 µl total RNA. cDNA was generated using a Bio-Rad C-1000 or S-1000 thermal cycler (Hercules, CA) through the following steps: 25°C 10 min, 37°C 120 min, 85°C 5 s, storage at 4°C.

PCR in real time.

µl cDNA was added to a master mix containing 0.5 µl TaqMan probe for GAPDH (Applied Biosystems, cat. no. 4326317E), 0.5 µl TaqMan probe for ANGPTL (Applied Biosystems, cat. no. Hs00205581_m1) and 5 µl master - mix with Lightcycler 480 probe (Roche, cat. no. 04887301001) per well in 50 384-well plates (Roche, cat. no. 04887301001). Real time PCR was performed on an ABI 7900HT Real Time PCR system (Roche) using ΔΔCt(RQ) analysis. Each duplex was tested in two independent transfections and each transfection was evaluated in duplicate unless otherwise indicated in the summary tables.

To calculate relative fold change, real-time data were analyzed using the ΔΔCt method and normalized to analyzes performed with cells transfected with 10 nM AD-1955 or mock-transfected cells. IC ₅₀ was calculated using a 4-way fit model using XLFit and normalized to cells transfected with AD-1955 or naive cells at the same dose range, or to self lowest dose. The AD-1955 sequence, which was used as a negative control, targets luciferase and has the following sequence:

semantic thread: cuuAcGcuGAGuAcuucGAdTsdT;

antisense strand: UCGAAGuACUcAGCGuAAGdTsdT .

The various embodiments described above may be combined to provide additional embodiments. All US patents, US patent application publications, foreign patents, foreign patent applications and non-patent publications mentioned in this specification are incorporated herein by reference in their entirety. Aspects of the embodiments may be modified if ideas from various patents, applications, and publications are to be used to provide additional embodiments.

These and other changes may be made to the embodiments in light of the detailed description above. In general, in the following claims, the terms used are not to be construed as limiting the claims to the specific embodiments disclosed in this specification and the claims, but are to be construed to include all possible embodiments together with the full scope of equivalents applicable to such formula points. Accordingly, the claims are not limited to this disclosure.

Example 4 Chemical modifications in siRNAs and in vitro silencing using modified siRNAs.

The construction of a semantic thread.

Construction of the ligand and conjugation site.

The sense strand was conjugated with the GalNAc ligand at the 3' position, as in the original compound.

Position 11 of the semantic thread.

Position 11 of the sense strand at the hypothetical cleavage site (opposite position 11 of the antisense strand (AS) if the sense strand is 21 nucleotides long and the antisense strand is 23 nucleotides long) was modified with a nuclease-sensitive (e.g., DNA) modification. Data from statistical analyzes covering many different conjugates suggest the importance of this provision.

Violation of thermal stability of the 3'-section of the semantic thread (positions 16-18).

This region was modified with thermostability-disrupting modifications such as GNA or mismatch with the opposite AS strand. Modifications to position 16 or 17 appear to be the most effective. In FIG. 1 and in table. 1, this position/site and the effect of thermostability impairment on in vitro performance are noted. Efficient knockdown comparable to that of the original template was obtained with GNA or other thermostability-impairing modifications such as deletion of the nitrogenous base (Y34) or mismatches of bases with the antisense strand. On the other hand, in general, a decrease in silencing was observed for 2'-OMe or DNA modifications of the strand complementary to the opposite AS strand.

Table 1

Strand position 17: siRNA sequence and modifications evaluated in vitro (see Fig. 1)

Target ID duplex ID sense n Sense strand (5-3')

IDAS

Antisense strand -3') p-RR AD 57727 21 A-ll ⁷ 799 49 AlsasCfdGfuGtuU CtUti<GfcUrcUfdUtaAFL96 A 117*00 20 usJtsaUaGbG.vAtagaAfiAUUfgUtLsibU

AD 61291 1 A-123259 1

AD 6-297 1 A-123260 1

MU & ¹¹ , MH d d'U U ' !d!*d LJ"

AD 61301 1 A-P32609 yGaiaCheb

AD 6'395 1 A-1233169 asa-ca _c ugud ^ujCL'qTgria jaA96

A-YUT 29 us JtsaLfaGfaG'cAfagaAfcAf^UfgUfLsusu d-117800 30 usJfsabf^GfaGWagaAfcAfcUfgUh susu D 11780033 usUGabidGUG nTR AD 61364 1 A-G32609 ^arugu irrtTugc yaG zia D96 D-133263 5 us4~'3 ia~<crPagar^(?sgi ususu nT'R AD 6'305 1 A-1233169 asascagugua ^UoCuciTgria xaChOb Α-123Σ63 6 us DGai agagcdAagad-CK tguususu nT-R WWHA MGM Bl Afssslfa .fuC-iuU % A-117L1P usl)4al B-VaCWA tagaifiAirtligHf, mTR AD 64426 1 A-1282S? 1 daa^agugu KdTugcuc ^b 3 -a' ?Λ96 mTR AD-64392 1 A-128288 1 abascagu^UJ. o~ugcjcaa «.aaL96 mT'R AD 64397 1 A-1282C9 1 .i'-Gschgs ssrtb3b mVR AD 65403 1 A-РЯ^Н 1 asa-catu-ui m^R AD 64411 I A-128305 1 isab^u ^u to if ssssbash L96 mT"R AD-64421 1 A-1282902 aa:_a _{b u} ^u a=L96

A-117800 '6 J''S3lTaSuGfcA*agaAfcArcUrgUfbsu;u

A-xP&OOL usJt5abfaG ^f 3GfcAfagaAfcAfcUTgUft.sus4 az rum

A-1M0 ⁷ 3 usJtsJL t^GfgG'cAfa^ΊAfcAfcUf _S UfL5UΣU

A-117800 79 irJUabfaGfaGT-AtagaAhAndJ^

A 123268 13 n.. JT.ai a^dicdAagad-c-icbgoucusu

ANG ANG ANG ANG

AD 5 ⁷ 927 7

AD 6314-1 2

A-117426 25 AfscsAfuAfuUruGf A^f-AfgbfcUfuUfuUf l9G A-117^ 23 asAfsaAfaGfaCfuGfaucAfaA^AwG-ususg

ΑΟ6ΠΊ1 A-P95t8 1 ^wuaujui dungi roTuju AD 64761 1 A-129591 1 ncsau^ucdGu. uuL95

AD F4 ⁷ h7 1 Α-ΙΡ^^Ή msiiHiiiiL-lTi p rdt'ujuul^b

A 117^-27 28 asAfsaAfaGfdCuGfaucAfaAfuA^uGdjsusg eNAsl/ 11

A-U7'?7 31 asAfsartfaGfaCfuGfaucAfaAfuAfuG^usg

A-117427 34 asAGaAfaGfaCfuGWAfaAfuA*uG‘iKusg

ApoC3 ADfc^7&7 1 A-1173Y23 Gfb.&UfuAMAUGfGFGIX^GfuAFuUkUrL96 A-129Sa _o 17 dsCfbaAfuAkU^UI^UiuJ)

ApoSZ AD 6482'1 A-129551 2 rs"judd _C a--6 _S srgisi196 A-129546 21 asCFaAtuAWgUfcrrUfuUt jAfaStcsaor

ApoC3 AD-6^806 1 A-129556 7 cfc^u^ar^a^-uNish^LEb A-129M6 36 asCFaAfjaWgUfcxUW*jnfaSfoas?

ApoC3 AD6^794 1 A 129554 4 „scsuu iiis D96 A-129546 24 a.GLaAfuAfcU^Uf .cUtuJfjAfaS-c a>c

4po( 3 AD-h4P$ 1 A-l2a=in02 Rn ^u.ini n1% A-'^Am3O asChaAfuAfi IJfgLJfi 111'uUf jAFh ifc=a>,?

ApoC3 AD-647391 A-1295612 gscsuuaaaajai’Ca^bU-uu; L96 A-129545 31 asCfsaAtuAfcUfgUfc:cUfuU* lAfaG^csase

ApoC3 AD 647951 A-129562 2 <d 4-129546 32 dsCfsd^fuMtUfeUh

AD 6447^ 1 A-128493 1

^T RSC AD 64493

TTRSC AD-64460

TTRSC AD-6 455

KFSC AD-64432

Ub^G^AfjUtuC MtU) _? lTaA ^c cCtdAWL96 A 128-.94 1 uXt ub^SfubfaC aj^dAfuG C^bjsc

A-128505 1 i'gs^gaj' LcadTguaaccaag L96

A-128507 ^1T guaacXa

A-128506 1 .bgSggajutCdrTguaan^riaagdLSo

A-128511 1 j jg > 196

TTRSC AD 64488 1 A-128-42 1 poison

TTRSC AD-64494 1 A-128513 1 gsfgajut anTgua=c^aagaL96

A-128A94 8 usCRul rgG'ubtaC^jgAfaAfuCfcCfa:

A-128^94 10 usC^ubigGiuLWadgAfaAW^^

A-128^-94 9 UbCFuLlgGtuL'f^augAfaAfu^cCfas jsc

A-123^94 14 ibCf ubf^Gfubi iC j ^AfdAfuCdCG^^ M-GFR 15 ti-rMiLifgGfuUtaC a tgAraAfuCto^ isr A-1284M 16 usCFubfgSfulfaC-a t-AfaAfbCWasjsc

- 62 043057

L96: .

In FIG. 2 and in table. 2 shows the effect of the position of the thermostability-disrupting GNA modification for the entire 3′ region (positions 16-18). These results indicate that the GNA modification at positions 16 and 17 showed good efficacy, similar to the original construct, while the GNA at position 18 showed reduced activity.

table 2

Posture effect of GNA-mediated thermal instability:

siRNA sequence and modifications evaluated in vitro (see Fig. 2)

Target ID duplex ID meaning, n. Semantic thread (5' - Z ⁷ )

Modification 10 nM SO 0.1 nM SD dmv

ANG

DM6 ANG AreSZ DrsSZ DroSZ

AroSe

DO-57927 7 D-117226 25 AEcsAfuAfuJfLGfAfUfcAfgUkUf^^ D-117427 23 ^D-64756 1 4

UP TO 531^1 2 A

AD-64777 1 A

AD-6'787 1

AD-6-S8 1 A

DO-64794 1 A

L 0-61900 1 L

P ^Q W 1

126526 4 1205561 .SCS-ll-U 11- = i'csg I ' ₅ ,) ,LU and G % yes u and yes I c j c u, u L96 sc: im 1 cg c 1 > ii L9- 5

117361 23 Gtsc = UtuAfaAtaGtGtG-aCtaG'uAtjUtcUtL96 L

129553 5 yes:- j aaa ₃ --ja a~i ό c L96 A

12955th

12955^ 1

With. and ate -a a and 7 "। ui o L96 :scs_ j: yes.edaa msmspl

117497 97

117427 23

117427 29

129516 .7

129546 23

129546 24

129516 25

Antisense strand (5' - 3') asAfsaAfaGfaCt jGfaucAfaAfuAfcGfi susg asAfscAtaGtaCf iGfaiKAfaAf uAti.Gti,SL=g asAfseAfaGfaCt iGtau-AfaAtuA+cGtc c-g asAfsc-taGtaCt iGfaucA^aAturtuGrtsi sg asGtsaAtuA-cUfgUfcccUtuJtu7raGtc:asa asGfsaAfuA-cUfgUfcccUfuJfuAfaGfcsasa asGfsaAfuA-cUfBUtacUfuUfuAfaGfcjasa asGfsa VuA^UfgUfcccUf uUfuAfaGfcsasa

.1, .l, 1dacha 7.0 - In 11.3 Yes 38.8 11.3 GNA at 17 11-3 37.5 " 1 yes to yes 19.8 65.8 ! 4 L yudada gna i yes V. 1-1 J dachas ....... m 45.0 ..........69D 2401 1l ldag 8.7 1: 50.2 15.9 LI S15 3/l 10 4 98° Yes -

Table 3

Position 2 of the antisense strand: siRNA sequence and modifications evaluated in vitro (see Fig. 3)

Target ID duplex ID meaning, n. Sense strand (5'-3') IDAS Antisense strand (5'-3'J Modification 10 nM SD OD nM SD

n'TF AD-57727 Al А-Ц77С9 13 A4 _cS c'aGfuGfuUfCfU'uGtcUf..UfaUfeAfL96 A-117800 71 usUkaUfaGfaGfcAfagaAfcAWgUfi sus j tn'TP AD-5139S5 A-123316 21 yes adada .Tu,, cdid L95 A-123268 11 us ~sa dadada daa _o a -c - _t ususu ηΓΤΡ AD-64.293- A-12331c 23 daaz adadada 'ada s'. ia yes 195 A-LS294 1 im for adada -aga-.-cac j usu.-u mTTR hD-t4-15 - A-12331c 27 a~u , . u with caa 1% A-1_825S 1 is -aaal -aja -yes g give u >b , days; h 3 1 4-.' With DCHK ,- , 4 yes 6.5 2-M- M1 5.0 79.7 v · AOK . > 1 J.e. 61.7 •' AreSZ AD-6-737.1 A-117361.23 GfscsU ^£ uAfaAfaGfGfG ^£ aC ^£ aGfuAfuUfcUfL95 A-129526 17 asGfsaAWfcU-'gUfcccUf uU ^£ u AfaGfcsasa AroSZ AD-64802.1 A-117361.3 3 GfccsU ^£ uAfaAfaGfGfG ^£ aCfaGfuAfuUfcUfL96 A-129571.5 as-sC-saauaci -gsTcccWu juaagcsasa ApoSZ AD-54790.1 A-117361 31 GfscsU ⁼ uAfaAfaGfGfG'aC ^f aGfuAfuUfcUfL95 A-129569 2 as: ₃ au sdaO'sssGy inaxsasa Initial 3' 4-..1 2D.1 ONH 31.2 11.0 63.9 ID 24Mr Hi 5 4-9 84.5 8.5 TTRSC AD-54474 1 4-128493 1 UfsgsGfgAfubfuCfAfUfgUfaAfcCfaAfgAfL96 D-12L-94 1 usCfsullfgGfuUfaCfa ugAtaAfuCfcCfeusc TTRSC AD-64472 1 A-1284C3 14 UfsgsGfgAfuLfuCf 4fUfgUfaAf:Cfa4fgAfL96 A-12S525 1 ussOsu ia CaugWa shdaazizs TTRSC AD-64484 1 A-12&193 16 UfsgsGfgAfuUfuCfAfUfgUfaAfcCfaAfgAfL96 A-128527 1 us.su iggu-adCaugdAada c.asusc TTRSC AD-64426 1 A-12S493 IS UfsgsG'gAfubfuCfAfUfgUfaAfcCfaAfgAfLOo A-128529 1 usCsuuggu ia~ Cai Eadaada-daazizs >’< HFHF 17 5 10.2 72.2 1 741 24.8 W 34- 27.0 2 iMc 39.9 17 1 126 6 44 iPHK 8? f 31 5 102.3 22 1 TMP AO-bLOAO- _C f- _lj;G fgijf _aU f _L4)fC f^^ A-122746 Ή usUUgUfaCf.Tf ^{&A £} ggaAfaUfaCfcA ^£ ^ TMP AD-64 ^C 57 ' A-IProC ¹ 4 sjsgg a in yes ig-g ^lT _iua a 196 A-129067 1 us sg date u 1 Agga--a a. c-gsasg TMP AD-64586 1 A-1266C2 7 sis _s ia u Cu-g gi ι i >a anL96 A-1290S5 1 us ¹ , sgi acccu. Agga--aia:c=gsasg And, f-dchaya '•2 2 14.2 66.5 P.1 AMI 1.7 4 15.4 87.2 • EVN 48.9 16.0 food with 37.1

Construction of the antisense strand.

Position 2 AS.

This position was identified by statistical analysis of a large array of conjugate data and by position walking throughout the AS strand to be sensitive to sterically demanding 2'-modifications, including 2'-OMe. However, the present inventors have found that certain modifications, including DNA and in some cases RNA, as well as other modifications that do not impart steric bulk at the 2' position, may be perfectly acceptable in the context of constructs with modifications that do not contain F The results of in vitro silencing studies are summarized in FIG. 3 and in table. 3, and they indicate that constructs with DNA, as well as RNA in position 2, generally retain the activity of constructs with modifications that do not contain F, similar to constructs in the form of the original template, while the 2'-OME modification is generally not quite acceptable and leads to a decrease in activity.

Regulation 14 AS.

This position was identified by statistical analysis of a large array of conjugate data and by position walking throughout the AS strand as sensitive to sterically demanding 2'-modifications, including 2'-OMe. However, it has been found that some modifications, including DNA and in some cases RNA, and other modifications that do not impart steric bulk at the 2' position, may be perfectly acceptable in the context of a construct with non-F modifications. Results in vitro silencing studies are summarized in FIG. 4 and in table. 4, and they indicate that constructs with DNA as well as RNA at position 14 generally retain the activity of constructs with modifications that do not contain F, similar to constructs in the form of the original template, while the 2'-OME modification is generally not quite acceptable and leads to a decrease in activity.

- 63 043057

Table 4

Position 14 of the antisense strand: siRNA sequence and chemical structures evaluated in vitro (see Fig. 4)

Target ID duplex ID meaning, n. Semantic thread (S'-3')

IDAS Antisense strand (S'-3') Modification 10nM SD 0.1nM SD mTTR

ApoC3 AroC3 AroC3 TTRSC “TR5C ~TRSC

IRSL zhs

AO-57727 41 A-llTTeg.TS.AfsaiCtiGUGfuU'CMjbfcU^Ufdb^^^^^ A-117800 71 usUfsaUfaGfaGtcAng.YcA YfgU ^; Lb.b J

AD-ol3 ^c S 5 A-123316 21 =sas a_jri. I ^r Tusc io'_r a =-L ^J 6 A-12326S 11 LbTaa .acaY agad-c o gi Lsusu

40-64^73 2 4-1233162=1 -,j; 'ujtja=-LA. A-'ZB'rtU u->3 sa a _; a Y-aga-c.- gu sum.

AD-64132 1 A-123316' ³ isas aoi'i ) - uy id _d A-12824-. 1 usd -aa;a ^d-aga - >- 's rg , _and h1ya

I. lodnach , 1U 2J 1

Ύκ'...............-γ-yo--,......

?nc...............iYY

LNA~ ' + 45.5( ~2,li

AD-o-<871 A-1173nl2b Of^sUfjAtaafaGrC-Qt-CtaGri ArU4UtL-o A-12954ο 1 a-bbaAtiAfoUtgLKo,JluUtuAtaGKia, II <_4ΐ4Y Y

AD-64802 1 A-117361 33 GfaCdUtjAtaAfaGfC ^£ a = CfaGfi AWcUfLOS A-125571 5 oia CsaSe Yh.Y 1121

Υκ 2U 7^4 2 <

1Y . 7 ¹ lY ~ Y tY Y .J YT

5 10li

13'it

AD-644741 iA-128493.1 iUfsgsGfgAfu'UfuCfAfUfgUfaAfcCfaAfgAfL96 |A-128494.1 usGsuUfgGfu WaCfaugAfaAfuCfcb^ G in? ¹ 1Γ

AD-64472 1 A-128403'4 iJt,giG' _£ AtuUtuC-AtU-iUNAT.-Ct3At _S 4*La6 4-12845 1 u5 su _Sb i a. '--ug 1, and , ,,u . DNA 21, ' 1 ^J 1, 1'5,?.P

AD-64458 1 A-1284Ch 19 Uf-g-.GfgAtuUfaL ^£ AfUsUfaAfcCfaAfgA ^f L9o A-128530 1 and G-L g _s i a- aug^a ica=usc Y Γ· 1' _ .>H'O1<· »

AD-645111 A-1285ut.2 bgSoB ^J| n Sh'G goyo-shu; a-gaL96 А-128525 2 us ilsii gfi a-.aHg'-a uorasijsc UHK , 20., 10 8, 12 213^ jAl>o4«HW^42i^4-128530 2 us 5s and g ₅ ia .aug^ai cm:c -ΉΚ -M; 1-2 n 2

Table 5

Sequence and chemical structures of siRNAs that target mTTRs used in an in vivo study in mice

Duplex ID ID Meaning, n.o. Sense strand (5'-3') IDAS Antisense strand (S'- 3') Construct

AD-57727 A-117799 AtsasCfaGtuGfuUtCfUfuGfcUfcUfaUt A-117800

AD-61398 A-123316 3Sascagu^U,.cdTugC',-a~Y'ajaaL96 A-123268

AD-64273 А-123316 bsascagugu .rdTugcbcr Yaаэ1_96 А-128300 usUfsaUfaGfaGtcAtagaA‘cAfcUtgUfususu original usdTsauagagcdAagadAcacuguususu original with revenge without F uscTsai-agagcdAagaAcacjgs usus u AS: RNA at 14

In vivo evaluation of siRNAs targeting mTTR.

Animals (n=3/group) received a single dose of siRNA at 2.5 mg/kg with serum FVII protein levels measured pre-dose and on days 4, 7, 13, 22, 29 and 36. FIG. 5 shows the FVII protein concentration-time profile for 2 siRNA AD-61398 and 64273 with no F modification compared to the parent compound AD-57727. In FIG. 6, a decrease in mTTR protein levels 96 hours post-dose is shown for two non-F-modified siRNAs at three different dose levels compared to parent compounds. In FIG. 7 shows the reduction profiles of serum mTTR protein for a repeated dose regimen (1 mg/kg, QW) up to and including the day (6 doses in total).

Overall, these studies indicate that siRNA AD-61398 and AD-642733 with F-free modifications exhibited in vivo efficacy and activity similar to the parent template construct.

siRNA targeting TMPRSS6.

Table 6

Sequences and chemical structures of siRNAs that target TMPRSS6

10 nM 0.1 nM ID duplex ID semantic semantic sequence IDAS Antisense strand sequence Avg. SD Avg. SD AD- 60940.7 122745.22 CfsusGfgUfaUfuUfCfCfuAfgGfgU laClaAIL96 A-122746.24 usUfsgUfaCfcCfuAfggaAfaUfa CfcAfgsasg 32.2 14.2 66.5 12.1 AD- 64604.1 A-129074.2 csusgguatlTuucdCuaggg(Tgn)acaa L96 A-129085.2 usUsguacccudAggatlAauaccags asg 25.2 10.2 57.5 27.9 AD- 64601.1 A-129073.1 csusggudAuuucdCuaggg(Tgn)acaa L96 A-129067.6 usdTsguacccudAggadAauacca gsasg 23.0 8.6 74.7 17.7 AD- 64567.1 A-126602.4 csusgguauuucdCuaggg(Tgn)acaaL 96 A-129067.1 usdTsguacccudAggadAauacca gsasg 62.4 15.4 87.2 23.8 AD- 64569.1 A-129083.1 csusgguadTuucdCuagggdAacaaL9 6 A-129067.16 usdTsguacccudAggadAauacca gsasg 49.7 9.4 49.4 19.7

The results indicate differing in vivo efficiencies for constructs with non-F modifications, depending on the exact location of the modifications and the combination of sense and AS strands. Although in vitro data suggest that compounds with F-free modifications had similar activity/potency to the parent compound, the F-free compound AD-64604, which appeared to be the most active in vivo, was still significantly less effective than the original AD-60940 (see Fig. 8).

Conducted additional design improvements with modifications that do not contain F, and conducted their evaluation, the data of which are summarized in table. 7. In FIG. 9 shows TMPRSS6 mRNA silencing in the liver 7 days after SC administration of a single dose of 3 mg/kg.

Table 7

Sequence and chemical structures of the 2nd set of siRNAs targeting TMPRSS6

Duplex ID ID Meaning, n.o. Sense strand (5 ^J - 3') IDAS Antisense strand (S'- 3 ^J ) Initial

AD-64601 A-129073.2 csusggudAuuucdCuag^(Tgn)acaal96 A-129067.18 usdTsguacccudAggadAauaccagsasg AD-60940

AD-65105 A-129073.2 csusggudAuuucdCuaggg(Tgn)acaaL96 A-129085.5 usUsguacccudAggadAauaccagsasg AD-60940

AD-65106 A-129073.2 csusggudAuuucdCuaggg(Tgn)acaaL96 A-129086.2 usdTsguacccudAs^adAsauaccagsasg AD-60940

AD-65107 A-129710.1 csusggudAuuucdCuagggdAacaaL96 A-129067.18 usdTsguacccudAggadAauaccagsasg AD-60940

AD-65108 A-129710.1 csusggudAuuucdCuagggdAacaaL96 A-129085.5 usUsguacccudAggadAauaccagsasg AD-60940

AD-65109 A-129710.1 csusggudAuuucdCuagggdAacaaL96 A-129086.2 usdTsguacccudAsggadAsauaccagsasg AD-60940

AD-65110 A-130024.1 csusggudAsuuucdCuaggg(Tgn)acaaL96 A-129067.18 usdTsguacccudAggadAauaccagsasg AD-60940

AD-65111 A-130024.1 csusggudAsuuucdCuaggg(Tgn)acaaL96 A-129085.5 usUsguacccudAggadAauaccagsasg AD-60940

AD-65112 A-130024.1 csusggudAsuuucdCuaggg(Tgn)acaaL96 A-129086.2 usdTsguacccudAsggadAsauaccagsasg AD-60940

AD-65104 A-129875.1 usgsguadTuuccdTagggudTcaaaL96 A-129876.1 usdTsuguacccdTaggdAaauaccasgsa AD-61002

In FIG. 9 shows that the improvement resulted in at least one compound with a non-F modification (AD-65105) that was comparable in vivo to the parent compound (AD-60940). This compound contains a sense strand with DNA at positions 6 and 11 and an antisme

- 64 043057 spruce strand with RNA in position 2 and DNA in positions 10, 14.

Motive construction.

When constructing the motif, the sense strand was conjugated to the GalNAc ligand at the 3' position using the same procedure that was used for the parent compound. Additional motifs were designed in accordance with embodiments of the present invention. Typical sequences are listed in Table. 8.

Table 8

Characteristic sequences

duplex name A thread Sequences AD-57727 S AfsasCfaGfuGfuUfCfUfuGfcUfcUfaUfaAfL96 AS usUfsaUfaGfaGfcAfagaAfcAfcUfgUfususu AD-57553 S GfscsUfuAfaAfaGfGfGfaCfaGfuAfuUfcUfL96 AS asGfsaAfuAfcUfgUfcccUfuUfuAfaGfcsAfsa AD-63042 S GfsusUfgCfuCfuAfUfAfaAfcCfgUfgUfuAfL96 AS usAfsaCfaCfgGfuUfuauAfgAfgCfaAfgsasa AD-63085 S UfscsCfuCfuGfaUfGfGfuCfaAfaGfuCfcUfL96 AS asGfsgAfcUfuUfgAfccaUfcAfgAfgGfascsa AD-65703 S gscsuuaaAfaGfGfGfacaguauucaL96 AS usGfsaauAfcUfGfucccUfuUfuaagcsasa AD-65704 S gscsuuaaAfaGfGfGfacaguauucaL96 AS usGfsaauacugucccUfuuuaagcsasa

Results of m vitro studies.

In FIG. 10 shows that among ten sequences representing three targets, two motifs were found, motif 1 (six phosphorothioate modifications of internucleotide bonds in the sense and antisense strands; four 2'-F modifications at positions 7 and 9-11 of the sense strand from 5 'ends of the sense strand and four 2'-F modifications at positions 2, 6, 14 and 16 of the antisense strand from the 5'end of the antisense strand) and motif 2 (six phosphorothioate modifications of internucleotide bonds in the sense and antisense strands; four 2'-F - modifications at positions 7 and 9-11 of the sense strand from the 5' end of the sense strand and six 2'-P modifications at positions 2, 6, 8-9, 14 and 16 of the antisense strand from the 5' end of the antisense strand) are characterized a statistically significant increase in activity compared to the parent compound AD-57727.

In vivo evaluation.

Silencing of the target under the action of siRNA was assessed using qPCR. The effectiveness of motives was assessed by the targeted effect on mTTR. Animals (n=3/group) were dosed with a single dose of siRNA at 3 mg/kg, with liver protein levels measured first pre-dose and then on days 7 and 22 as shown in FIG. eleven.

In FIG. 12 shows enhanced activity with Stability Enhancer Conjugates (SEC-C) with liver samples assessed for activity (mRNA) on day 7 post-dose. Animals received a single dose of 3 mg/kg (s.c.). These data demonstrate the effect of motifs on in vivo activity.

In FIG. 13 shows the increased activity (approximately 4-fold increase in activity) for the novel motifs (motifs 1 and 2) compared to the parent compound, with data assessed on day 7 post-administration. These data demonstrate the effect of motifs on in vivo activity. The fold increase was observed consistently for all sequences.

In FIG. 14 shows a significantly increased duration of action for all three sequences, demonstrating that the new motifs exhibit an increased duration of action.

In FIG. 15 shows the results for ApoC3-GalNAc3 SAR with SC administration of a single dose of 3 mg/kg hAAV 1x1011 GC/mouse.

Example 5 VP and PS ₂ modifications at the 5' end of the antisense strand.

The following are exemplary protocols for the synthesis of oligonucleotides containing 5'vinylphosphonate (VP) and the synthesis of oligonucleotides containing 2'-deoxythymidine linked via a phosphorodithioate (PS ₂ ) bond at the 5' end of the oligonucleotide. One skilled in the art will appreciate that the same or similar techniques can be used to synthesize analogous oligonucleotides. Other synthetic techniques known to those skilled in the art may also be used to synthesize and generate data and similar oligonucleotides and modifications, including, but not limited to, the synthetic techniques disclosed in

Whittaket et al., Stereoselective synthesis of highly functionalized P-stereogenic nucleosides via palladiumcatalyzed P-C cross-coupling reactions, Tetrahedron Letters 49:

6984-87 (2008); Zhao and Caruthers, Synthesis and Preliminary

Biochemical Studies with 5'-Deoxy-5'-methylidyne Phosphonate

Linked Thymidine Oligonucleotides, Tetrahedron Letters 37(35): 6239-42 (1996);

and U.S. Patent Application Publication No. 2013/0084576, each of which is incorporated herein

- 65 043057 document by reference in its entirety.

Protocols for the synthesis of oligonucleotides containing 5'-vinylphosphonate.

Introduction of pivaloyloxymethyl-(POM)-protected VP

binding and oxidation.

Coupling of amidite was carried out under standard synthesis conditions using 0.25 M 5(ethylthio)-1H-tetrazole in acetonitrile for activation.

Standard thiolation protocols were followed using either 3(dimethylaminomethylene)amino-3H-1,2,4-dithiazole-5-thione (DDTT) or phenylacetyl disulfide (PADS) to convert the phosphite triester to a phosphorothioate bond. Since the vinyl phosphonate building block does not contain a DMT protecting group at the 5' position, the final detritylation step was not performed.

Removal of protective groups and cleavage. After synthesis, vinyl phosphonate-containing oligonucleotides were deprotected in a 3:1 mixture of aqueous NH ₃ and EtOH with the addition of 2.5% v/v 40% methylamine solution for 5 h at 60°C or 16 h at 35°C.

Introduction of ethyl-protected VP

binding and oxidation.

Coupling of amidite was carried out under standard synthesis conditions using 0.25 M 5(ethylthio)-1H-tetrazole in acetonitrile for activation.

Standard thiolation protocols were followed using either 3(dimethylaminomethyl)amino-3H-1,2,4-dithiazole-5-thione (DDTT) or phenylacetyl disulfide (PADS) to oxidize the phosphite triester and introduce a phosphorothioate bond. Since the vinyl phosphonate structural block does not contain a DMT protecting group at the 5' position, the final detritylation step was not performed.

Removal of protective groups and cleavage.

A 50:1 (v/v) solution of acetonitrile (ACN) and pyridine (Pyr) was prepared and 3A pore size molecular sieves were added to keep the mixture as dry as possible. To this mixture was added 3.5 ml (5 g) of trimethylsilyl iodide (TMSI) for every 135 ml of ACN/Pyr solution. This solution must be freshly prepared as it has a maximum shelf life of one day. A 0.5M solution of mercaptoethanol in 1:1 (v/v) acetonitrile-triethylamine was then prepared and 3A pore size molecular sieves were added. When the 5'-VP-containing oligonucleotide was on the resin and in the synthesis column, the TMSI solution was slowly added in an amount of approximately 5-10 vol. column and allowed the reaction to run for 15 min. This step was repeated twice resulting in a total exposure time of approximately 45 minutes. The resin was then thoroughly washed with ACN followed by a flow of approximately 5-10 column volumes of mercaptoethanol solution through the column, allowing the reaction to proceed for 10 minutes. This step was repeated once with a total exposure time of 20 min. After another thorough washing with ACN, the oligonucleotide bound to the support was deprotected and cleaved from the support using standard conditions.

Protocol for the synthesis of oligonucleotides containing 2'-deoxythymidine linked via a phosphorodithioate bond at the 5' end of the oligonucleotide

- 66 043057

. oh oh. .

! _p /=\

5CH _g CH^-C-< y

binding and oxidation.

The phosphoramidite solution was prepared from commercially available dT-thiophosphoramidite (Glen Research) according to the manufacturer's protocol in dry acetonitrile at a concentration of 0.15 M. Coupling was performed under standard conditions using 0.25 M 5-(ethylthio)-1H-tetrazole in acetonitrile during a total binding time of 17 minutes. In this cycle of synthesis, the capping step was not performed. Oxidation (thiolation) was carried out using 3(dimethylaminomethylene)amino-3H-1,2,4-dithiazole-5-thione (DDTT) by extending the reagent addition time and reaction time to 3x10 min. Final detritylation was carried out using standard synthesis conditions.

Removal of protective groups and cleavage.

The solid support (in column) was washed with 0.5 M piperidine in ACN (2x15 min exposure time) before transferring the resin to a suitable container and working under standard conditions (e.g., 3:1 NH:EtOH aqueous solution for 5 hours at 60° C. or 16 hours at 35° C.) for cleavage from the solid support and deprotection of the oligonucleotide.

The remaining procedures of the method for the synthesis of oligonucleotides are similar to those described in example 2.

In FIG. 16 shows a schematic representation of Ago2 with siRNA included. In general, 5'phosphate-functionalized siRNAs (ESC conjugates) show increased in vitro activity. For example, ∼80% of the sequences tested showed an increase in intrinsic efficiency when transfected in vitro, and ∼30% showed an approximately 10-fold increase in IC ₅₀ . However, in vivo there is a rapid loss of 5'-phosphate in the endo/lysosomal compartments. A modified phosphate, 5'-vinylphosphonate (5'-VP), mimicking stable phosphate, also shown in FIG. 16 is attached to the 5' end of the modified oligonucleotide. This phosphonate was originally designed by Merck.

One embodiment of the present invention is directed to 5' end modifications to increase efficiency (inclusion in RISC). End modifications provide stable mimics of phosphate and activate endogenous phosphorylation.

In FIG. 17 is a diagram showing how the presence of 5'-VP generally increases in vivo activity based on the evaluation of four different ApoB targeting sequences. LDL levels 7 days after SC administration of a single dose of 3 mg/kg were analyzed for four conjugates (with or without 5'-VP modification). It can be seen from the diagram that a 3-fold increase in ED ₅₀ is observed with certain sequences targeting ApoB. The in vivo increase was confirmed with additional compounds/targets including ApoC3, Tmpssr6 and TTR. ApoB targeting sequences are listed in Table 1. 9.

_____________________________Table 9

Name duplex Sense thread (5'-3') Antisense strand (5'-3') AD 63750 AfsasAfgAfgGfuGfUfAfuGf gCfuUfcAfaAfL96 usUfsuGfaAfgCfcAfuacAfcCf uCfuUfuscsa AD 64557 AfsasAfgAfgGfuGfUfAfuGf gCfuUfcAfaAfL96 VPusUfsuGfaAfgCfcAfuacAfc CfuCfuUfuscsa AD 63734 CfsusGfgAfcAfuUfCfAfgAf aCfaAfgAfaAfL96 usUfsuCfuUfgUfuCfugaAfuGf uCfcAfgsgsg AD 64560 CfsusGfgAfcAfuUfCfAfgAf aCfaAfgAfaAfL96 VPusUfsuCfuUfgUfuCfugaAfu GfuCfcAfgsgsg AD 63716 UfsgsUfgAfcAfaAfUfAfuGf gGfcAfuCfaAfL96 usUfsgAfuGfcCfcAfuauUfuGf uCfaCfasasa AD 64559 UfsgsUfgAfcAfaAfUfAfuGf gGfcAfuCfaAfL96 VPusUfsgAfuGfcCfcAfuauUfu GfuCfaCfasasa AD 63711 CfscsUfgGfaCfaUfUfCfaGf aAfcAfaGfaAfL96 usUfscUfuGfuUfcUfgaaUfgUf cCfaGfgsgsu AD 64561 CfscsUfgGfaCfaUfUfCfaGf aAfcAfaGfaAfL96 VPusUfscUfuGfuUfcUfgaaUfg UfcCfaGfgsgsu

In FIG. 18 depicts various chemical modifications that can replace the PS bond, including phosphorodithioate (PS ₂ ) and methylphosphonate (MePhos), which activate endogenous phosphorylation. Modified siRNAs are generally not optimal substrates for Clp1 kinase, possibly due to the negative effect of the 2'-OMe modification on the first nucleotide of the AS strand. However, the 2'-OMe modification together with the phosphorothioate linkage is desirable for protection against exonucleases. Substitution of 2'-OMe-modification, for example, 2'-F, as well as modification of the PS-linkage can activate protection against exonucleases while maintaining metabolic stability.

- 67 043057

In FIG. 19 shows an in vitro evaluation chart of end modifications, including 2'-OMe-MePhos, 2'-OMe-PS, dN(PS ₂ ), and 2'-F-PS. This diagram shows that the link dn(PS ₂ ) and 2'-F-PS showed increased in vitro activity compared to the original (2'-OMe-PS). In particular, dN(PS) ₂ was stable in an in vitro assay with tritosomes, while 2'-F-PS showed a propensity for metabolic conversion. Transfections of primary mouse hepatocytes were performed using two ApoB-targeting conjugates at concentrations of 10 nM and 0.1 nM (n=4).

In FIG. 20 shows two graphs showing how a slight change at the 5' end of the antisense strand can significantly increase in vivo efficacy. In FIG. 20A shows that 2'F-PS at position 1 of the antisense strand can increase the activity of 5'-P dependent sequences, and FIG. 20B shows a ~3-fold increase in dN(PS) ₂ potency compared to parent compound, this increase was similar to 5'-VP.

Example 6 5'-VP modifications and evaluation for siRNA activity.

Synthesis of 5'-vinylphosphonate-phosphoramidite with a pivaloxymethyl protecting group

Scheme 1

Reagents and reaction conditions for Scheme 1:

(a) Dess-Martin periodinan, DCM, 0°C;

(b) NaH, tetra(pivaloyloxymethyl)-bisphosphonate, THF, -78°C, followed by stirring at 0°C, 70% (E and Z isomers);

(c) formic acid: water, 1:1, 24 hours, separation of E and Z isomers by silica gel column chromatography or RP-HPLC (reverse phase HPLC);

(d) 2-cyanoethyl N,N,N',N'-tetraisopropylphosphordiamidite, 5-(ethylthio)-1H-tetrazole, ACN, 6 h, room temperature, 65%.

Synthesis of tetra(pivaloyloxymethyl)-bisphosphonate (X)

Tetramethylmethylenebisphosphonate (120 g, 0.51 mol), NaI (308 g, 2 mol), chloromethyl pivalate (387 g, 2.5 mol) and acetonitrile (400 ml) were mixed and refluxed overnight. Product formation was confirmed by TLC (thin layer chromatography) in EtOAc with 5% methanol. The reaction mixture was diluted with ether (1000 ml) and washed with water (2×1000 ml), dried with Na ₂ SO ₃ and evaporated. The solid was washed with cold hexane and dried in vacuo to give 148 g (45%) of X as a pale yellow solid.

1 H NMR (500 MHz, CDCl ₃ ): δ 5.73-5.63 (m, 8H), 2.65 (t, 2H), 1.22 (s, 36H); ³¹ P NMR (500 MHz, CDCl ₃ ): δ 18.61.

Get connection 2

To an ice-cold solution of compound 1 (3.0 g, 8 mmol) in 150 ml anhydrous dichloromethane was added Dess-Martin periodinan (DMP) (1.4 eq.; 4.7 g, 11.2 mmol). The reaction mixture was stirred at 0° C. for 1 hour and then at room temperature for 3 hours. Product formation was confirmed by TLC. Then the reaction mixture was added to 200 ml of a solution of 10% Na ₂ S ₂ O ₃ and saturated NaHCO ₃ (1:1), followed by the addition of 200 ml of ethyl acetate. The crude aldehyde was extracted into dried ethyl acetate and concentrated under reduced pressure. The crude aldehyde was used without any purification in the next step.

- 68 043057

Yield=2.87 g (97%); the purity according to NMR was approximately 70%; LC-MS: mass/charge

371.

Get Compound 3

stirring at room temperature Ratio E Ζ (88Ή) for 1h. - ^from -

A solution of tetrapolyoxometalate (POM)-bisphosphonate sodium salt was prepared by adding bisphosphonate (X) in 14 ml THF (12.6 g, 20 mmol) to a suspension of NaH (0.58 g, 24 mmol) in 20 ml THF at 78°C and stirred for 15 min.

A solution of aldehyde 2 (2.86 g) in 40 ml anhydrous THF was added dropwise to the tetra(ROM)-bisphosphonate sodium salt solution prepared above at -78°C. The reaction mixture was stirred at -78° C. for 1 hour, at 0° C. for another hour, and then at room temperature for 1 hour. Product formation was confirmed by TLC (EtOAc: hexane 7:3). The crude reaction mixture was added to 300 ml of saturated ammonium chloride solution and extracted with 300 ml of ethyl acetate. The organic layer was washed with brine, dried over sodium sulfate. The solution was then concentrated under reduced pressure, and the residue was purified by silica gel column chromatography (EtOAc in hexane=20-100%) to give compound 3 (4.0 g) as a mixture of E/Z isomers (88/12) with yield 72%.

Get Compound 4

A solution of 3 (4 g, 5.7 mmol) in 200 ml HCOOH/H2O (1:1, v:v) was stirred at room temperature for 24 h. Product formation was confirmed by TLC (MeOH:CH2Cl2=5: 95).

This solution was concentrated under reduced pressure, and the residue was purified by silica gel column chromatography (MeOH:CH ₂ Cl ₂ =7:93 v/v). Fractions were checked for RP-HPLC (column C18, buffer A=0.05% TFA in water, buffer B=0.05% TFA in ACN; gradient 5-95% over 25 min) to confirm the purity of the two isomers (E- and Z isomers): The E isomer eluted at 14.1 min and the Z isomer eluted at 14.9 min. The primary fractions obtained by chromatography on silica gel contained a mixture of E and Z isomers, and the remaining fractions contained the E isomer. Fractions containing a mixture of E and Z isomers were purified by RP-HPLC. 4-E-isomer 2.3 g was obtained, yield 71%.

E-isomer:

1H NMR (400 MHz, acetonitrile^): δ 8.98 (s, 1H), 7.30 (d, J=8.1 Hz, 1H), 6.80 (ddd, J=23.7, 17, 2, 5.0 Hz, 1H), 6.02 (ddd, J=21.6, 17.1, 1.7 Hz, 1H), 5.77 (d, J=3.2 Hz, 1H), 5.57 (m, 5H), 4.32 (m, 1H), 4.01 (dd, J=7.0, 5.4 Hz, 1H), 3.82 (dd, J=5.5, 3.2 Hz, 1H), 3.41 (s, 3H), 1.14 (d, J=1.5 Hz, 18H); ³¹ P NMR (162 MHz, acetonitrile-d ₃ ): δ 18.29.

Z-isomer:

¹ H NMR (500 MHz, acetonitrile-d3): δ 9.50 (s, 1H), 7.44 (d, J=8.1 Hz, 1H), 6.69 (ddd, J=54.4, 13.3, 8.7 Hz, 1H), 5.93 (ddd, J=17.8, 13.3, 1.3 Hz, 1H), 5.80 (d, J=2.9 Hz, 1H ), 5.69 - 5.58 (m, 5H), 5.22 (m, 1H), 4.01 (dd, J=7.1, 5.3 Hz, 1H), 3.88 (dd, J=5.3, 2.9 Hz, 1H), 3.49 (s, 3H), 1.19 (d, J=5.8 Hz, 18H); ³¹ P NMR (202 MHz, acetonitrile-d3): δ 18.75.

Get Compound 5

- 69 043057

To a solution of compound 4-E-isomer (2.1 g, 3.62 mmol) and ethylthiotetrazole (0.46 g, 3.62 mmol) in ACN (40 ml) was added 2-cyanoethyl N,N,N',N '-tetraisopropylphosphordiamidite (1.311 g, 4.35 mmol). This mixture was stirred at room temperature for 2 hours. Product formation was confirmed by TLC in hexane: EtOAc (2:8 in 0.15% TEA). The reaction mixture was filtered, concentrated and loaded onto a silica gel column. The sample was eluted with 20-100% EtOAc in hexane with TEA (0.15%) to give compound 5 as a white foam (1.75 g, 62%).

E-isomer:

¹ H NMR (400 MHz, acetonitrile^): δ 9.09 (s, 1H), 7.38 (d, J=8.1 Hz, 1H), 6.89 (m, 1H), 6.10 ( dddd, J=21.4, 17.1, 2.8, 1.7 Hz, 1H), 5.86 (t, J=3.8 Hz, 1H), 5.67 - 5.55 (m, 5H), 4.66 - 4.50 (m, 1H), 4.40 - 4.20 (m, 1H), 3.99 (m, 1H), 3.92 - 3.57 (m, 4H) , 3.44 (s, 3H), 2.73-2.64 (m, 2H), 2.14 (s, 1H), 1.24-1.14 (m, 3OH); ³¹ P NMR (162 MHz, acetonitrile-d3): δ 151.79 (d, J=71.3 Hz), 18.07 (d, J=54.0 Hz).

Z-isomer:

1H NMR (400 MHz, acetonitrile^): δ 9.02 (s, 1H), 7.41 (dd, J=8.1, 1.6 Hz, 1H), 6.62 (dddd, J=53, 7, 13.1, 9.7, 7.0 Hz, 1H), 5.97 (dd, J=17.4, 13.1 Hz, 1H), 5.80 (dd, J=7.0, 3.5 Hz, 1H), 5.70 - 5.52 (m, 5H), 5.41 (m, 1H), 4.40 - 4.10 (m, 1H), 4.06 - 3.98 (m, 1H), 3.93 - 3.56 (m, 4H), 3.47 (s, 3H), 2.68 (m, 2H), 2.14 (s, 1H), 1.33 - 1.11 (m, 30H); ³¹ P NMR (202 MHz, acetonitrile^): δ 150.81 (d, J=141.4 Hz), 15.17.

Protocols for the synthesis of oligonucleotides containing 5'-vinylphosphonate.

The synthesis of vinylphosphonate monomers and 5'-VP-modified oligonucleotides was carried out similarly to the procedures described in the literature (WO 2008/100447, Chen et al., Lima et al., SingleStranded siRNAs Activate RNAi in Animals, Cell 150: 883-894 (2012) ; Prakash et al., Identification of metabolically stable 5-phosphate analogs that support single-stranded siRNA activity, Nucleic Acids Research 43: 2993-3011 (2015), which are hereby incorporated by reference in their entirety). Briefly, the 5'phosphate was protected with ethyl ether, and then the ethyl ether-protected phosphate was subjected to a two-step deprotection:

1) TMS-I on a solid support under anhydrous conditions and

2) standard deprotection of the oligonucleotide to obtain a 5'-VP-modified oligonucleotide.

This method is also discussed in Example 5.

Effect of metabolically stable (E-) and (Z-) 5'-vinylphosphone on siRNA activity.

Double-stranded small interfering RNA (siRNA) with a 5'-phosphorylated antisense strand facilitates efficient incorporation into the RNA-inducible silencing complex (RISC) to ensure robust RNAi-mediated gene silencing. Thus, 5'-phosphorylation of synthetic siRNAs by Clp1 kinase is key for RISC incorporation and strand selection (Weitzer et al., The human RNA kinase hClp1 is active on 3' transfer RNA exons and short interfering RNAs, Nature 447: 222-226 (2007)). Phosphate mimics providing a metabolically stable bond have been used to modify nucleosides used as antiviral agents (WO 2008/100447, Chen et al.), to 5'-end modify siRNA to increase gene silencing activity compared to the corresponding non-phosphorylated siRNA, in particular, single-stranded siRNAs (Lima et al. Single-Stranded siRNAs Activate RNAi in Animals, Cell 150: 883-894 (2012);

Prakash et al., Identification of metabolically stable 5phosphate analogs that support single-stranded siRNA activity,

Nucleic Acids Research 43: 2993-3011 (2015)).

In this example, the effect of phosphate mimics in double-stranded siRNAs was evaluated both in vitro and in vivo.

The siRNA sequences used in this example are shown in the tables below.

Duplex ID Sequence of semantic thread Antisense strand sequence AD-66572 usgsgaagCfaGfUfAfuguugaugg aL96 usCfscauCfaAfCfauacUfgCf uuccasasa AD- 68365.3 usgsgaagCfaGfUfAfuguugaugg aL96 VPuCfcauCfaAfCfauacUfgCf uuccasasa AD- 68431.1 usgsgaagCfaGfUfAfuguugaugg aL96 VPUfCfcauCfaAfCfauacUfgC fuuccasasa AD- 68433.1 usgsgaagCfaGfUfAfuguugaugg aL96 VP (Tam)CfcauCfaAfCfauacU fgCfuuccasasa

u=2' OMe, 5' OH U;

Vpu=2' OMe, 5' VP U;

VPUf=2' F, 5' VP U;

VP(Tam)=2'N-methylacetamide, 5' VP T.

- 70 043057

Modification in N1 antisense strand % factor IX knockdown at 1 mg/kg (Day 14) 2'0Me, 5'OH U 46 2'0Me, 5'VP U 80 2'F, 5'VP U 83 2'N-methylacetamide, 5'VP T 77

The effect of 5'-vinylphosphonate (VP) with E and Z geometry on siRNA activity was compared. The results show that in vivo efficiency of chemically modified siRNAs can be increased with 5'-trans-(E-)VP, which mimics natural phosphate perfectly, while 5'-cis-(Z-)VP showed no increase in efficiency, indicating that the Z-isomer does not adequately mimic natural phosphate.

In FIG. 21A-B shows a SAR analysis comparing in vitro and in vivo activity of siRNAs targeting ApoB that contain the 5'-OH- and 5'-E-VP modification (at the 5' end of the antisense strand). In FIG. 21A shows the results of in vitro transfection of mouse 1° hepatocytes. In FIG. 21B shows LDL levels 3 days post single dose (SC dosing). The results in FIG. 21B demonstrate that ApoB-targeting siRNAs that were modified with 5'-E-VP showed increased activity.

In FIG. 22 shows the results of an in vitro comparison of the efficacy of 5'-E-VP modification and 5'-Z-VP modification in siRNA-GalNAc conjugates targeting mTTR and F9. The results were obtained by in vitro transfection of mouse primary hepatocytes. As shown in the figure, the siRNA conjugate that was modified with 5'-E-VP showed a maintenance or increase in potency, while the siRNA conjugate that was modified with 5-Z-VP showed a decrease in potency.

In FIG. 23 shows the results of an in vivo comparison of the activity of the 5'-E-VP modification and the 5'-Z-VP modification of the F9-targeted siRNA-GalNAc conjugate (Single Dose SC). The results demonstrate that the siRNA conjugate that was modified with 5-E-VP showed increased gene silencing activity compared to the 5'-OH control, while the siRNA conjugate that was modified with 5-U-VP showed activity similar to that of the 5'-OH control.

The results in these figures show that 5'-phosphorylation of the antisense strand is desirable for efficient RNAi-mediated gene silencing.

The efficiency of chemically modified siRNAs can be increased by using 5'-transvinylphosphonate (5'-E-VP), which perfectly mimics natural phosphate.

Example 7 5'-C-malonyl modifications and evaluation for siRNA activity.

Synthesis and introduction of 5'-C-malonyl-modified nucleotides into the 5'-end of siRNA.

General conditions of the experiment. All moisture-sensitive reactions were carried out under anhydrous conditions under an argon atmosphere. Flash chromatography was performed on a Combi Flash system from Teledyne ISCO (Lincoln, Neb.) using pre-packed ReadySep silica gel columns from Teledyne ISCO. High resolution electrospray ionization mass spectrometry (ESI-HRMS) spectra were recorded on a Q-Tof API-US spectrometer from Waters (Milford, MA) using direct sample injection in positive mode (capillary voltage=3000 kV, cone=35, temperature source=120°C and desolvation temperature=350°C). 1H and ¹³ C NMR spectra were recorded at room temperature on Varian spectrometers (Palo Alto, CA) at 400 MHz (1H) and 126 MHz ( ¹³ C) and chemical shifts in ppm are given relative to residual solvent peaks. Interaction constants are given in hertz. Signal splitting patterns are described as singlet (s), doublet (d), triplet (t), quartet (q), broadband signal (br), or multiplet (m). ³¹ P NMR spectra were recorded at 162 MHz in the mode of suppression of spi-spin interaction with protons, and chemical shifts are given relative to external H ₃ PO ₄ (80%). LC/ESI-MS was performed on a single quadrupole LC/MS 6130 system from Agilent (Santa Clara, CA) using an XBridge C8 column (2.1x50 mm, 2.5 µm) at 60°C. Buffer A consisted of 100 mM 1,1,1,3,3,3-hexafluoro-2-propanol (HFIP) and 16.3 mM triethylamine (TEA) in H ₂ O, and buffer B was 100% methanol.

- 71 043057

Scheme 2

Reagents and conditions for scheme 2:

(a) benzyloxymethyl acetal (BOM) chloride, DBU, DMF, 30 min, 0°C, quantitative yield (Kurosu et al., Synthetic studies towards the identification of novel capuramycin analogs with mycobactericidal activity, Heterocycles 77: 217-225 (2009) ; Kurosu et al., Concise Synthesis of Capuramycin, Org. Lett. 11:2393-2396 (2009), each of which is incorporated by reference in its entirety);

(b) methyltriphenoxyphosphonium iodide, DMF, 15 min, room temperature, 92%;

(c) sodium methoxide, dimethylmalonate, 1,2-DME, 24 h, reflux, 92%;

(d) 10% Pd/C, H ₂ atmosphere, i-PrOH/H ₂ O (10:1, v/v), 0.05 eq. formic acid, 12 h, room temperature, 98% (Aleiwi et al., A reliable Pd-mediated hydrogenolytic deprotection of BOM group of uridine ureido nitrogen, Tetrahedron Lett. 53: 3758-3762 (2012), which is incorporated by reference throughout its completeness);

(f) NEt ₃ -3HF, THF, 48 h, room temperature, 88%;

(d) 2-cyanoethyl-N,N-diisopropylchlorophosphoramidite, DIEA, DCM, 18 hours, room temperature, 56%;

(g) 1 M piperidine aqueous solution, 24 h, room temperature; then a mixture of 30% aqueous ammonia/ethanol (3:1, v/v), 36 h, room temperature, quantitative yield, 2 ⁺ =piperidinium.

Synthesis of N ³ -benzyloxymethyl-2'-O-methyl-3'-O-tert-butyldimethylsilyluridine (2).

2'-O-methyl-3'-O-tert-butyldimethylsilyl-uridine (1.20 g, 53.7 mmol) was converted to 2 (26.5 g, quantitative yield) according to a variant of the previously reported procedure.

Synthesis of N ³ -benzyloxymethyl-5'-deoxy-5'-iodo-2'-O-methyl-3'-O-tert-butyldimethylsilyluridine (3)

Compound 2 (10 g, 20.3 mmol) was dissolved in 100 ml anhydrous DMF and 20 g (40.6 mmol) methyltriphenoxyphosphonium iodide was added. The mixture was stirred at room temperature for 15 minutes. Methanol (200 ml) was added to the reaction mixture and the mixture was stirred for an additional 15 minutes. The solvents were evaporated to dryness; the residue was dissolved in dichloromethane (DCM) and washed with 5% Na ₂ S ₂ O ₃ solution, followed by washing with water. The organic layers were collected, dried over Na ₂ SO ₄ , filtered and evaporated to dryness. The crude residue was purified by silica gel chromatography using 0-50% ethyl acetate (EtOAc) in hexanes as eluent to give 3 as a white foam (11.2 g, 92%).

1H NMR (400 MHz, DMSO-d6): δ 7.77 (d, J=8.2 Hz, 1H), 7.30 (m, 5H), 5.90 (d, J=5.2 Hz, 1H), 5.85 (d, J=8.2 Hz, 1H), 5.33 (d, J=13.0 Hz, 1H), 5.30 (d, J=13.0 Hz, 1H) , 4.58 (s, 2H), 4.23 (t, J=4.5 Hz, 1H), 4.07 (t, J=5.1 Hz, 1H), 3.87 (q, J= 6.1 Hz, 1H), 3.55 (dd, J=10.6, 6.3 Hz, 1H), 3.39 (dd, J=10.6, 6.3 Hz, 1H), 3, 32 (s, 3H), 0.89 (s, 9H), 0.14 (s, 3H), 0.12 (s, 3H). ¹³ C NMR (126 MHz, DMSO-d ₆ ): δ 161.7, 150.7, 140.2, 138.0, 128.2, 127.4, 127.3, 101.6, 87.9, 83.3, 80.8, 72.7, 71.0, 70.1, 57.6, 25.6, 17.7, 6.2, -4.7, -4.8.

HRMS-ESI: calc. for C ₂₄ H ₃₅ IN ₂ NaO ₆ Si (M+Na) ⁺ is 625.1207; found value: 625.1205.

Synthesis of N ³ -benzyloxymethyl-5'-deoxy-5'-C-(dimethylmalonyl)-2'-O-methyl-3'-O-tert-butyldimethylsilyluridine (4)

Sodium methoxide (2 g, 33 mmol) was placed in a dry round bottom flask, dimethyl malonate (12 ml, 100 mmol) and anhydrous 1,2-dimethoxyethane (DME, 100 ml) were added and the mixture heated to reflux. Compound 3 (10 g, 16.5 mmol), after co-evaporating twice with anhydrous acetonitrile, was dissolved in 70 ml of anhydrous DME and added to a solution of dimethylmalonate and sodium methoxide heated under reflux. Reflux was carried out for 24 hours. The reaction mixture was cooled to room temperature and methanol (50 ml) was added to quench the reaction. Solvents and volatile components were evaporated in vacuo. The crude residue was purified by silica gel chromatography using 0-100% EtOAc in hexanes as eluent to give compound 4 as a colorless oil (9.2 g, 92%).

- 72 043057 ¹ H NMR (400 MHz, DMSO-d ₆ ): δ 7.66 (d, J=8.2 Hz, 1H), 7.30 (m, 5H), 5.80 (d, J= 8.2 Hz, 1H), 5.76 (d,

J=4.0Hz, 1H), 5.33(d, J=13.4Hz, 1H), 5.30(d, J=13.4Hz, 1H), 4.58(s, 2H) , 4.14 (t, J=5.4 Hz, 1H), 3.91 (m,

1H), 3.76 (m, 1H), 3.64 (m, 4H), 3.60 (s, 3H), 3.33 (s, 3H), 2.37 - 2.09 (m, 2H ), 0.87 (s, 9H), 0.09 (s, 3H), 0.08 (s, 3H). ¹³ C NMR (126 MHz, DMSO-d6): δ 169.1, 168.8, 161.9, 150.6, 140.4, 138.0, 128.1, 127.4, 127.3, 101 ,3,

88.6, 88.5, 81.3, 80.9, 73.1, 71.0, 70.0, 59.7, 57.5, 52.5, 48.0, 31.5, 25, 6, 17.7, -4.76, -5.06.

HRMS-ESI: calc. for C ₂₉ H ₄₂ N ₂ NaO ₁₀ Si (M+Na)+ is 629.2506; found value: 629.2508.

Synthesis of 5'-deoxy-5'-C-(dimethylmalonyl)-2'-O-methyl-3'-O-tert-butyldimethylsilyluridine (5).

Compound 4 (8.7 g, 14.3 mmol) was dissolved in 660 ml of iso-propanol/water (10:1, v/v) and 0.9 g of 10% Pd/C was added, followed by the addition of 27 ml (0.7 mmol) formic acid. Air was removed from the flask under vacuum; the reaction flask was purged with hydrogen and stirred under normal pressure hydrogen at room temperature for 12 hours. The reaction mixture was filtered through celite and washed with ethanol. The filtrates were collected and evaporated to dryness. The crude residue was purified by silica gel chromatography using 0-5% MeOH in DCM as eluent. The appropriate fractions were combined and evaporated to dryness to give 5 as a white foam (6.7 g, 98%).

¹ H NMR (400 MHz, DMSO-d ₆ ): δ 11.38 (d, J=1.8 Hz, 1H), 7.61 (d, J=8.1 Hz, 1H), 5.71 ( d, J=4.3 Hz, 1H), 5.65 (dd, J=8.0 Hz, J=2.1 Hz, 1H), 4.16 (t, J=5.3 Hz, 1H) , 3.91 (t, J=4.8 Hz, 1H), 3.73 (m, 1H), 3.63 (m, 4H), 3.61 (s, 3H), 3.31 (s, 3H), 2.24 - 2.07 (m, 2H), 0.87 (s, 9H), 0.08 (s, 3H), 0.08 (s, 3H). ¹³ C NMR (126 MHz, DMSO-d6): δ 169.2, 168.9, 163.0, 150.4, 141.2, 141.2, 102.1, 87.7, 81.2, 80 .9, 73.1, 57.5, 52.5, 52.4, 52.3, 48.0, 31.6, 25.6, 17.7, -4.8, -5.1.

HRMS-ESI: calc. for C ₂₁ H ₃₄ N ₂ NaO ₉ Si (M+Na) ⁺ is 509.1931; found value: 509.1929.

Synthesis of 5'-deoxy-5'-C-(dimethylmalonyl)-2'-O-methyluridine (6).

Compound 5 (6.7 g, 13.8 mmol) was stirred with triethylamine trihydrofluoride (11 ml, 202.5 mmol) in 150 ml anhydrous THF in a round bottom flask at room temperature for 48 h. The solvents were evaporated in vacuo until two thirds of the original volume. The residue was purified by silica gel chromatography using 0-10% MeOH in DCM as eluent. The appropriate fractions were combined and evaporated to dryness to give 6 as a white foam (4.5 g, 88%).

¹ H NMR (400 MHz, DMSO-d6): δ 11.37 (s, 1H), 7.56 (d, J=8.1 Hz, 1H), 5.72 (d, J=4.3 Hz , 1H), 5.64 (d, J=8.1 Hz, 1H), 5.24 (d, J=6.3 Hz, 1H), 3.94 (q, J=5.7 Hz, 1H ), 3.86 (t, J=4.8 Hz, 1H), 3.72 (m, 1H), 3.64 (m, 4H), 3.61 (m, 3H), 3.34 (s , 3H), 2.25 - 2.07 (m, 2H). ¹³ C NMR (126 MHz, DMSO-d6): δ 169.2, 169.0, 163.0, 150.3, 141.0, 102.0, 87.4, 81.6, 80.8, 71 .9, 57.6, 52.5, 52.4, 48.0, 31.9.

HRMS-ESI: calc. for C ₁₅ H ₂₀ N ₂ NaO ₉ (M + Na) ⁺ is 395.1067; found value: 395.1070.

Synthesis of 5'-deoxy-5'-C-(dimethylmalonyl)-2'-O-methyluridine-3'-O-(O-(2-cyanoethyl)-N,N-di-isopropyl)phosphoramidite (7).

Compound 6 (3.0 g, 8 mmol) was co-evaporated three times with anhydrous acetonitrile and then dried under vacuum over P ₂ O ₅ overnight. The dry residue was dissolved in 60 ml of anhydrous DCM; diisopropylethylamine (4.5 ml, 24 mmol) and 2-cyanoethyl N,N-diisopropylchlorophosphoramidite (2.2 ml, 10.0 mmol) were added sequentially. After 1 h stirring under argon, another 1.0 ml (4.0 mmol) 2-cyanoethyl N,N-diisopropylchlorophosphoramidite was added and agitation continued for an additional 18 h. The reaction mixture was diluted with 150 ml DCM and washed with 200 ml saturated sodium bicarbonate solution. The organic layer was dried with sodium sulfate and removed by filtration. The solvents were evaporated in vacuo and the crude residue was purified by silica gel chromatography. The eluent was a mixture of hexanes/EtOAc/NEt ₃ (from 66:33:1, v/v/v in a stepwise gradient to a mixture of hexanes/EtOAc/NEt ₃ 33:66:1, v/v/v). The appropriate fractions were combined, evaporated to dryness, co-evaporated with anhydrous acetonitrile and dried under high vacuum to give 7 as a white foam (3.2 g, 56%).

1H NMR (400 MHz, CD3CN, mixture of diastereoisomers): δ 8.97 (s, 1H), 7.36 (m, 1H), 5.78 (d, J=4.2 Hz, 1H), 5.64 (d, J=8.1 Hz, 1H), 4.23 - 3.80 (m, 4H), 3.77 - 3.59 (m, 8H), 3.45 - 3.41 (m, 3H ), 2.68 (t, J=5.9 Hz, 2H), 2.44-2.31 (m, 2H), 1.42-1.00 (m, 12H). ³¹ P NMR (162 MHz, CD3CN, mixture of diastereoisomers): δ 151.8, 151.6. ¹³ C NMR (126 MHz, CD3CN, mixture of diastereoisomers): δ 170.6, 170.2, 163.8, 151.3, 141.3, 119.6, 103.0, 102.9, 89.6, 89.2, 82.9, 82.5, 82.4, 81.8, 81.3, 81.2, 75.3, 75.2, 75.1, 75.0, 59.8, 59, 7, 59.3, 59.1, 58.9, 58.8, 53.3, 53.2, 49.5, 49.4, 44.2, 44.15, 44.1, 44.0, 33.0, 25.0, 24.9, 24.8, 21.0, 20.9.

HRMS-ESI: calc. for C ₂₄ H ₃₈ N ₄ O ₁₀ P (M+H) ⁺ is 573.2326; found value: 573.2321.

Synthesis of 5'-deoxy-5'-C-malonyl-2'-O-methyluridine piperidinium salt (8).

Compound 6 (0.1 g, 0.3 mmol) was stirred with 1 M aqueous piperidine (10 mL, 10 mmol) at room temperature for 24 h. Solvents were evaporated in vacuo and the residue was dissolved in 30% ammonia/ethanol ( 3:1, v/v) and stirred at room temperature for 36 hours. The solvents were evaporated in vacuo and 8 was obtained as a colorless oil (quantitative).

¹ H NMR (400 MHz, D2O): δ 7.75 (d, J=8.1 Hz, 1H), 5.92 (m, 2H), 4.16 (t, J=5.5 Hz, 1H ), 4.06 (t, J=4.7 Hz, 1H), 3.99 (m, 1H), 3.50 (s, 3H), 3.27 (t, J=7.0 Hz, 1H ), 3.17 (t, J=5.7 Hz, 6H), 2.27 - 2.06 (m, 2H), 1.87 - 1.54

- 73 043057 (m, 8H). ¹³ C NMR (126 MHz, D ₂ O): δ 179.5, 179.2, 168.0, 153.0, 142.5, 103.1, 88.5, 83.7, 83.3, 72 .7, 58.9, 55.9,

45.3, 34.7, 23.0, 22.2.

HRMS-ESI (M+H)+: calc. for C ₁₃ H ₁₇ N ₂ O ₉ is 345.0929; found value: 345.0919.

Synthesis of oligonucleotides.

Oligonucleotides were synthesized using an ABI-394 DNA/RNA Synthesizer using modified synthesis runs based on those provided for this device. A solution of 0.25 M 5-(S-ethylthio)-1H-tetrazole in acetonitrile was used as an activator. Phosphoramidites were in the form of 0.15 M solutions in anhydrous acetonitrile. The oxidizing agent was 0.02M I ₂ in THF/pyridine/H ₂ O. N,N-dimethyl-N'-(3-thioxo-3H-1,2,4-dithiazol-5-yl)methanimidamide (DDTT) , at the rate of 0.1 M in pyridine, was used as a sulfating agent. The detritylation reagent was 3% dichloroacetic acid (DCA) in DCM. In the case of 5'phosphate compounds, a chemical phosphorylating reagent from Glen Research (Cat. No. 10-1902-02) was used to introduce the 5'-monophosphate. After completion of the automated synthesis, the solid support was washed with 0.1 M piperidine in acetonitrile for 10 min, and then washed with anhydrous acetonitrile and dried in an argon atmosphere. The oligonucleotide was then manually separated from the support and deprotected using 30% NH4OH/ethanol (3:1, v/v) or 40% methylamine (0.5 ml/µmol solid support) for 6 h at 55 °C or 15 min at 60°C, respectively. The solvent was collected by filtration and the support was washed with DMSO (1.5 ml/µmol solid support).

The 5'-C-malonyl oligonucleotides deposited on the solid support were first treated with 1 M aqueous piperidine solution (1.5 ml/µmol solid support) for 24 h at room temperature, and then the solution was filtered and evaporated to dryness. The residue was dissolved in 30% NH4OH/ethanol (3:1, v/v, 2 ml/µmol solid support) and shaken at room temperature for 36 h and then evaporated to dryness. Crude oligonucleotides were purified by anion exchange HPLC using a linear gradient of 0.22 M to 0.42 M NaClO ₄ in 0.02 M Tris-HCl, pH 8.5/50 vol% acetonitrile over 120-150 min at room temperature . All single strands were purified to >85% purity by HPLC (260 nm) and then desalted by size exclusion chromatography using an AP-2 glass column (20x300 mm, Waters) custom-packed with Sephadex G25 (GE Healthcare), eluted using sterile, nuclease-free water. Hybridization to obtain siRNA duplexes was performed by mixing equimolar amounts of complementary strands to a final concentration of 20 μM in 1x PBS buffer, pH 7.4, and heating in a water bath at 95°C for 5 min, followed by slow cooling to room temperature.

Evaluation of the effect of 5'-C-malonyl modifications on the activity of gene silencing and stability.

Double-stranded RNA 5'-phosphorylation is desirable for efficient incorporation of small interfering RNAs (siRNAs) into the RNA-induced silencing complex (RISC), resulting in RNAi-mediated gene silencing. Endogenous or exogenous siRNAs are usually readily phosphorylated by cytosolic kinase, and in most cases the presence of synthetic 5'monophosphate is not required. However, in some cases of chemically modified siRNAs, metabolically stable mimics of 5'-phosphate can lead to increased stability, increased RISC incorporation, and more efficient gene silencing.

In this example, the effect due to the 5'-C-malonyl fragment, which was introduced as a nucleotide with a modification at the 5' end of the antisense strand of chemically modified siRNAs, was evaluated using solid phase synthesis. 5'-C-malonyl can exist as a dianion at physiological pH, similar to the 5'-monophosphate dianion. The in vitro gene silencing activity, metabolic stability, and RISC incorporation of siRNAs containing a 5'-Smalonyl group in antisense strands were assessed and compared with the corresponding 5'-phosphorylated and non-phosphorylated equivalents.

Cell culture and transfection.

Primary mouse hepatocytes were obtained from Life Technologies and cultured in Williams E Medium with 10% fetal bovine serum (FBS). Transfection was performed by adding 4.9 µl of Opti-MEM plus 0.1 µl of Lipofectamine RNAiMax (Invitrogen) per well to 5 µl of each siRNA duplex at the required concentration per well in a 384-well plate. The mixture was incubated at room temperature for 20 min and 40 μl of complete growth medium containing 5000 cells was added to the mixture containing siRNA. Cells were incubated for 24 h before RNA isolation. A similar procedure was followed when transfecting 10,000,000 cells. Dose-effect experiments were performed using eight 6-fold serial dilutions ranging from 20 nM to 75 pM or 50 nM to 187.5 pM.

Isolation of RNA.

RNA was isolated using a Dynabeads mRNA isolation kit (Invitrogen). Cells were lysed in 75 µl lysis/binding buffer containing 3 µl beads per well and mixed

- 74 043057 for 10 minutes on an electrostatic shaker. Buffers were prepared according to the manufacturer's protocol. The wash steps were performed automatically with a Biotek EL406 using a magnetic plate holder. The beads were washed once (90 μl) with buffer A, once with buffer B and twice with buffer E with aspiration steps between washes.

Synthesis of cDNA.

cDNA synthesis was performed using the High capacity cDNA reverse transcription kit (Applied Biosystems). A mixture of 1 µl 10x buffer, 0.4 µl 25x dNTP, 1 µl random primers, 0.5 µl reverse transcriptase, 0.5 µl RNase inhibitor and 6.6 µl water per reaction was added to each well. The plates were sealed, shaken for 10 minutes on an electrostatic shaker, and then incubated at 37°C for 2 hours. Thereafter, the plates were shaken at 80°C for 8 minutes.

PCR in real time.

cDNA (2 µl) was added to a master mix containing 0.5 µl of mouse GAPDH TaqMan probe (Applied Biosystems, cat. no. 4308313), 0.5 µl of mouse ApoB or PTEN TaqMan probe (Applied Biosystems, cat. no. Mm01545156_m1 and Mm01212532_m1 respectively) and 5 µl master mix with Lightcycler 480 probe (Roche) per well in 50 384-well plates (Roche). Real time PCR was performed on an ABI 7900HT RT-PCR system (Roche) using ΔΔCt(RQ) analysis. Each duplex and concentration was tested in four biological replicates. To calculate relative fold change, real-time data were analyzed using the ΔΔCt method and normalized to analyzes performed on cells transfected with 10 nM non-specific siRNA. IC ₅₀ values were calculated with a 4-way fit model using XLFit.

Table 10. IC ₅₀ values for siRNA with 5'-C-malonyl, 5'-phosphate and 5'-OH in PTEN and ApoB cell silencing assays.

Target FOR miRNA 5'modification Sense thread (5'-3') ^a Antisense strand (5'-3') ^a IC ₅₀ (nM) ¹ 1 ApoW HE С«с UgGaCaUUCaGaAcAaGaA GalNAc u U'c GuGu Uc UgaaUg UcCaGg-g-u 1.0 2 ApoW phosphate С«с UgGaCaUUCaGaAcAaGaA GalNAc Pu U'cUuGuUc UgaaUgUcCaGg'g u 0.1 3 ApoW HE U*g UgAcAaA UAuGgGcAuCaA GalNAc u l>g4 uGcCcG uauUuGuCaCa"a'a 0.5 4 ApoW phosphate U-g UgAcAaA UAuGgGcAuCaA GalNAc Pu U'gAuGcCcAuauUuGuCaCa'a'a 0.1 5 AroW malonate C«c UgGaCaUUCaGaAcAaGaA GalNAc Mu'U'cUuGuUcUgaaUgUcCaGg'g u 0.7 6 AroW malonate U*g UgAcAaA UAuGgGcAuCaA GalNAc Mu G"g4uGcCcTuauGuGuCaCa"a a 0.4 7PTEN HE AaGuAaGgAcCaGaGaCaAdT'd T u Ug Uc Uc UgGuCc UuAc GudT-dT 0.7 8 PTN phosphate AaGuAaGgAcCaGaGaCaAdT-d T PuGgGcGcGgGuCcGu4cGudT»dT 0.2 9 PTN malonate AaGuAaGgAcCaGaGaCaAdT'd T Mu Ug Uc Uc UgGuCc Gtl4c GudT'dT 0.2

Note.

^and P means 5'-monophosphate; M means 5'-malonate (ie 5'-C-malonyl); italic upper case letters and standard lower case letters denote 2'-deoxy-2'-fluoro (2'-F)- and 2'-O-methyl (2'OMe)-modifications of Sugars, respectively; means a phosphorothioate (PS) bond; dT means 2'deoxythymidine nucleotide; GalNAc means hydroxyprolinyl trivalent N-acetylgalactosamine ligand (Nair et al., Multivalent N-Acetylgalactosamine-Conjugated siRNA Localizes in Hepatocytes and Elicits Robust RNAi-Mediated Gene Silencing, J. Am. Chem. Soc. 136, 16958-16961 (2014), incorporated by reference in its entirety).

^b denotes the half-maximal inhibitory concentration (IC ₅₀ ) for gene silencing in primary mouse hepatocytes. All values were obtained as a result of experiments in triplicate.

Stability analysis using tritosomes

Rat liver tritosomes (Xenotech, special order PR14044) were thawed to room temperature and diluted to 0.5 units/ml in 20 mM sodium citrate buffer, pH 5.0. Samples were prepared by mixing 100 μl of tritosome acid phosphatase at 0.5 units/ml with 25 μl of siRNA at 0.4 mg/ml in a microcentrifuge tube. After incubation for 4 h or 24 h in an Eppendorf Thermomixer set at 37°C and at 300 rpm, 300 µl of Phenomenex Lysis Loading Buffer and 12.5 µl of internal standard siRNA at a rate of 0.4 were added to each well. mg/ml. Samples for time 0 were prepared by mixing 100 µl of tritosome acid phosphatase at 0.5 units/ml with 25 µl of siRNA sample at 0.4 mg/ml, 300 µl of Phenomenex Lysis Loading Buffer, and 12.5 µl of internal standard siRNA from calculation of 0.4 mg/ml. siRNA was extracted from the sample for each time point (0 h, 4 h, 24 h) using the Phenomenex Clarity OTX Starter kit.

- 75 043057

kit. The samples were then resuspended with 500 µl of nuclease-free water and 50 µl of the sample were analyzed by LC/MS.

RISC immunoprecipitation and RT-PCR analysis.

Primary mouse hepatocytes (10,000,000 cells) transfected with siRNA were lysed in lysis buffer (50 mM Tris-HCl, pH 7.5, 100 mM NaCl, 1% NP-40, 0.1% SDS) with a protease inhibitor (Sigma-Aldrich ). Lysate concentration was measured using a BCA protein kit (Thermo Scientific). For each reaction, 2 mg of total lysate was used. Anti-Ago2 antibody was purchased from Wako Chemicals (clone no: 2D4). Control mouse IgG was purchased from Santa Cruz Biotechnology (sc-2025). Dynabeads (Life Technologies) were used to precipitate antibodies. Ago2 associated siRNA and endogenous miR122 were measured by Stem-Loop RT followed by TaqMan PCR analysis based on previously published methods.

Computer modeling of the interaction between 5'-deoxy-5'-C-malonyluridine and the human Ago2 MID domain.

The modes of recognition of available crystal structures in complexes formed between MID hAgo2 (amino acids 432-578; residues 440-572 were separated by electron density) and IMP (PDB ID code 3LUJ) and full-length hAgo2 and miR-20a (PDB ID code 4F3T) revealed that the recognition of 5'-terminal phosphates was very similar. The only difference between these two structures is that, in complex with full-length Ago2, the PIWI domain residue (Arg-812) promotes 5'-phosphate recognition. Thus, the UMP:MID complex was used as the basis for modeling the interaction between 5'-malonyluridine and the hAgo2 MID domain. The 3D coordinates of the UMP:MID complex were obtained from the Protein Data Bank (http://www.rcsb.org). Using the UCSF Chimera program (version 1.5.3), all water molecules were removed from the crystal structure, and the 5'phosphate group was converted into a 5'-C-malonyl moiety. Then hydrogen atoms were added and the geometric structure of 5'-deoxy-5'-C-malonyluridine and its spatial orientation, as well as the formation of H-bonds / interactions without formation of bonds in the 5'-phosphate pocket of the hAgo2 MID domain were optimized using the Amber force field (ff12SB and Gasteiger charges for standard amino acids and non-standard amino acid residues, respectively) as implemented in UCSF Chimera.

In FIG. 24A-C are graphs showing the dose effect of PTEN-targeted siRNAs with (A) 5'-OH, (B) 5'-C-malonyl, and (C) 5'-phosphate on primary murine hepatocytes in an in vitro silencing assay. PTEN. All values were obtained as a result of experiments in triplicate.

In FIG. 25 shows enzyme resistance results for siRNAs with 5'-OH, 5'-C-malonyl, and 5'phosphate that were incubated with rat liver tritosomes. Target sequences for siRNA are shown in table. 10. siRNA was incubated at 0.4 mg/ml (approximately 5 mM) for 4 hours and 24 hours, respectively, in the presence of tritosomes. The percentage of full length threads was determined using HPLC. Data were normalized to untreated controls.

In FIG. 26 shows the results of incorporation into RISC siRNA with 5'-OH, 5'-C-malonyl and 5'-phosphate (5'-modification in antisense strands) as determined by Ago2 immunoprecipitation from primary mouse hepatocytes and by amplification with RT- Single strand PCR included in Ago2. Endogenous miR122 levels were determined as a control. Target sequences for siRNA are shown in table. 10. siRNA 7, 8 and 9 were transfected into cells at a concentration of 10 nM. The levels of antisense strands are presented in nM siRNA strands per mg of cell lysate.

The results in these figures show that siRNAs with 5'-C-malonyl retain or increase in vitro gene silencing, high levels of incorporation into Ago2, and provide significantly improved metabolic stability of the antisense strand from siRNA duplexes compared to the corresponding 5'-phosphorylated and non-phosphorylated equivalents.

In silico modeling studies showed a positive fit of the 5'-C-malonyl group within the 5'-phosphate binding pocket in MID hAgo2. Therefore, 5'-C-malonyl, a metabolically stable 5'-phosphate bioisostere, has excellent biomimic properties for siRNA therapeutic applications.

Example 8 Method for Stereoselective Synthesis of 5'-C-Alkyl Nucleosides Using Trialkyl Aluminum and Dialkyl Zinc

General scheme for the synthesis of 5'-C-alkylnucleosides by the addition of weak alkyl nucleophiles to 5'-nucleoside aldehydes

2 3 4

R = TBS, any protecting group or any substituent

X = H, F, OMe, OMOE, ONMA, OPGA (PG - any protecting group), OR (R - any alkyl group) B = unprotected or protected U, 1, C, A, G or any modified nucleobase R ¹ = Me or any alkyl substituent

- 76 043057

Alternatively, instead of AlR' ₃ or ZnR'2 (shown in Scheme 2 above), one can use

SnR'4, TiR'4 and various other metals except Li and Mg with an R' group in this reaction scheme.

Scheme 3. General scheme of stereospecific interconversion of 5'-alkyl nucleoside epimers via anhydro ring closure

RO X

R = IBS any protecting group or any substituent

X = H. F. OH. OME. OMOE. ONMA. OPGA (PG - any protective group), OR (R - any ι weave tn sh 1 rupee g)

B = unprotected!! or protected 11 1 C or any modified pyrimidine nucleobase R' = Me or any alkyl substituent

LG=OMs. Ols or any suitable leaving group

LGX' = MsC 1 TsC 1 or luoop acid chloride or strong acid anhydride

Base = DBU or any basic reagent

Solvent = TTIF dioxane or any water-miscible organic solvent

Scheme 4. Synthesis of pyrimidine-5'-aldehydes 2a-d using Dess-Martin periodinan

(1-2)a: Vx = U, X = OMe (1-2)b: Bx = U,X = F (1-2)c: Vx = T. X = H (1-2)d: Vx \u003d C ^Ac , X \u003d OMe

93-99%;

Aldehyde content: 60-70%

5'-Deoxy-3'-O-[(1,1-dimethylethyl)dimethylsilyl]-2'-O-methyl-5'-oxouridine 2a.

Dess-Martin periodinan (40.7 g, 96 mmol) was added to a stirred and cooled (0°C) solution of 3'-OTBS protected uridine 1a (29.8 g, 80 mmol) in anhydrous DCM (600 ml) in argon atmosphere. The cooling bath was removed and the mixture was stirred at room temperature for 4 h, after which the presence of the starting alcohol 1a could not be detected by TLC. The mixture was cooled to 0° C. and poured into a vigorously stirred mixture of 10% sodium thiosulfate solution (250 ml) and saturated sodium bicarbonate solution (350 ml). After stirring at room temperature for 45 minutes, a pronounced precipitation was observed. The precipitate was filtered off and the solid was washed with DCM (200 ml x 2). The filtrate was placed in a separating funnel, the organic phase was separated and dried over anhydrous sodium sulfate. The solid from the filter funnel was transferred to an Erlenmeyer flask, acetone (450 ml) was added, the suspension was stirred for 15 min, filtered and the solid was washed with acetone (200 mlx2). The acetone extract was evaporated, the residue was combined with the DCM extract, the solvent was evaporated, the residue was dissolved in ACN-acetone 1:1 (200 ml), the solvent was evaporated again and the solid was dried in vacuo to give the crude aldehyde 2a 27.6 g (93%). The aldehyde content was approximately 71% (as determined by the CHO/NH ratio determined by H ¹ NMR in ACN-d ₃ ), which was used in the next step without further purification. The product can be stored without noticeable decomposition at -20°C under argon. 1H NMR of the main component (400 MHz, ACN-d3): δ 0.15 (s, 6H); 0.93(s, 9H); 3.37 (s, 3H); 3.62-3.68 (m, 2H); 3.81 (dd, J=4.6, 5.9 Hz, 1H); 4.48 (d, J=3.4 Hz, 1H); 4.59 (ddd, J=0.4, 3.4, 4.5 Hz, 1H); 5.70 (d, J=8.2 Hz, 1H); 5.94 (d, J=6.0 Hz, 1H); 7.71 (d, J=8.2 Hz, 1H); 9.17 (bs, 1H); 9.68 (d, J=0.5 Hz, 1H).

2',5'-Dideoxy-3'-O-[(1,1-dimethylethyl)dimethylsilyl]-2'-fluoro-5'-oxouridine 2b was prepared similarly from 1b (13.8 g, 38 mmol) and DMP ( 19.5 g, 46 mmol) in anhydrous DCM (550 ml). After stirring overnight at room temperature, the mixture was cooled, quenched and extracted with DCM to give 12.7 g (93%) of crude aldehyde containing approximately 60% of title product 2b. 1H NMR of the main component (400 MHz, ACN-d ₃ ): δ 0.13 (s, 3H); 0.14 (s, 3H); 0.92 (s, 9H); 4.41 (d, J=6.0 Hz, 1H); 4.67 (ddd, J=4.9, 6.0, 13.6 Hz, 1H); 5.15 (ddd, J=2.8, 4.9, 52.5 Hz, 1H); 5.68 (d, J=8.1 Hz, 1H); 5.89 (dd, J=2.8, 18.3 Hz, 1H); 7.55 (d, J=8.1 Hz, 1H); 9.26 (bs, 1H); 9.64 (d, J=1.0 Hz, 1H).

5'-Deoxy-3'-O-[(1,1-dimethylethyl)dimethylsilyl]-5'-oxothymidine 2c was obtained similarly from 1c (17.9 g, 50 mmol) and DMP (25.4 g, 60 mmol) in anhydrous DCM (500 ml). After stirring for 3 hours at 0° C., the mixture was quenched and extracted with DCM to give 20.0 g (quantitative) of crude aldehyde containing about 63% of title product 2c. ¹ H NMR of the main component (400 MHz, ACN-d ₃ ): δ 0.135 (s, 3H); 0.140 (s, 3H); 0.92 (s, 9H); 1.85 (d, J=1.2 Hz, 3H); 2.08-2.24 (m, 2H); 4.38 (d, J=2.2 Hz, 1H); 4.75 (dt, J=2.2, 5.7 Hz, 1H); 6.24 (dd, J=6.2, 7.9 Hz, 1H); 7.55 (d, J=1.3 Hz, 1H); 9.18 (bs, 1H); 9.65 (d, J=0.5 Hz, 1H).

- 77 043057

N-Acetyl-5'-deoxy-3'-O-[(1,1-dimethylethyl)dimethylsilyl]-2'-O-methyl-5'-oxocytidine 2d was obtained similarly from 1d (33.1 g, 80 mmol) and DMP (40.7 g, 96 mmol) in anhydrous DCM (600 ml). After stirring at room temperature for 4 hours, the mixture was quenched and worked up in a similar manner to 2a to give 33.2 g (100%) of crude aldehyde containing about 56% of title product 2d. 1H NMR of the main component (400 MHz, ACN-d ₃ ): δ 0.108 (s, 3H); 0.117 (s, 3H); 0.92 (s, 9H); 2.14 (s, 3H); 3.44 (s, 3H); 3.90-3.94 (m, 1H); 4.49-4.52 (m, 2H); 5.92 (d, J=3.8 Hz, 1H); 7.33 (d, J=7.5 Hz, 1H); 8.10 (d, J=7.5 Hz, 1H); 9.13 (bs, 1H); 9.72(s, 1H).

Synthesis of purine 5'-aldehydes 2e-f using Dess-Martin periodinanan, anhydrous quench

DMP/DCM

0°С - k.t. Extinguishing without e

TBSO x

2e-f

TBSO X

1e-f

quantitative output;

aldehyde content: 50-55% (1-2) e: Bx \u003d A ^Bz , X \u003d OMe (1-2) f: Bx \u003d G ^{I Bu} , X \u003d OMe

N-Benzoyl-5'-deoxy-3'-O-[(1,1-dimethylethyl)dimethylsilyl]-2'-O-methyl-5'-oxoadenosine 2e.

Periodinan Dess-Martin (20.4 g, 48 mmol) was added to a stirred and cooled (0°C) solution of 3'-OTBS protected adenosine 1e (20.0 g, 40 mmol) in anhydrous DCM (300 ml) in argon atmosphere. The cooling bath was removed and the mixture was stirred at room temperature for 3 h, after which the presence of the starting alcohol 1e could not be detected by TLC. Isopropyl alcohol (0.61 ml, 8 mmol) was added and stirring was continued for an additional 1 h. The solvent was removed in vacuo and ethyl acetate (220 ml) was added followed by the slow addition of hexanes (150 ml) over 4 h with stirring. The mixture was stirred at room temperature overnight, filtered and the solid was washed twice with ethyl acetate-hexanes (1:1). The filtrate was evaporated in vacuo and the residue was co-evaporated with dry acetonitrile (300 ml). Acetonitrile (100 ml) was added to form a suspension which was stirred overnight, filtered and the solids washed twice with acetonitrile. The filtrate was evaporated in vacuo to give a white foamy residue which was dried under high vacuum to give 20.2 g (100%) of crude aldehyde containing about 54% of title product 2e, which was used in the next step without further purification. . 1H NMR of the main component (400 MHz, ACN-d ₃ ): δ 0.188 (s, 3H); 0.190 (s, 3H); 0.97 (s, 9H); 3.34 (s, 3H); 4.49 (dd, J=4.3, 6.3 Hz, 1H); 4.51 (dd, J=1.0, 2.9 Hz, 1H); 4.90 (dd, J=3.0, 4.2 Hz, 1H); 6.24 (d, J=6.3 Hz, 1H); 7.54 (t, J=7.3 Hz, 2H); 7.61-7.67 (m, 1H); 7.97-8.03 (m, 2H); 8.44 (s, 1H); 8.66 (s, 1H); 9.50 (bs, 1H); 9.82 (d, J=1.0 Hz, 1H).

N-Isobutyryl-5'-deoxy-3'-O-[(1,1-dimethylethyl)dimethylsilyl]-2'-O-methyl-5'-oxoguanosine 2f was obtained similarly from 1f (19.3 g, 40 mmol) and DMP (20.4 g, 48 mmol) in anhydrous DCM (300 ml). After stirring for 3 hours at room temperature, isopropyl alcohol (0.61 ml, 8 mmol) was added and stirring was continued for an additional 1 hour. The solvent was removed in vacuo and ethyl acetate (225 ml) was added, followed by the slow addition of hexanes (150 ml ) for 30 min with stirring. The mixture was stirred at room temperature for 5 hours, filtered and the solid was washed twice with ethyl acetate-hexanes (1:1.5). The filtrate was evaporated in vacuo and the residue co-evaporated with a mixture of toluene (250 ml) and dry acetonitrile (250 ml) and then with acetonitrile (250 ml). The white foamy residue was dried under high vacuum to give 21.5 g of crude aldehyde containing approximately 53% of title product 2f, which was used in the next step without further purification. 1H NMR of the main component (400 MHz, ACN-d3): δ 0.167 (s, 3H); 0.175 (s, 3H); 0.95 (s, 9H); 1.18 (d, J=6.8 Hz, 3H); 1.19 (d, J=6.8 Hz, 3H); 2.66-2.77 (m, 1H); 3.31 (s, 3H); 4.31 (dd, J=4.3, 7.0 Hz, 1H); 4.48 (dd, J=1.0, 2.3 Hz, 1H); 4.69 (ddd, J=0.4, 2.4, 4.3 Hz, 1H); 6.01 (d, J=7.0 Hz, 1H); 8.03 (s, 1H); 9.45 (s, 1H); 9.79 (d, J=1.1 Hz, 1H); 12.05 (bs, 1H).

Stereoselective addition of weak Me-nucleophiles to nucleoside-5'-aldehydes.

A. General remarks.

Stereoselectivity due to nucleic bases. The tables below show that (S)-epimers of 5'-Me-pyrimidine nucleosides can be synthesized with a high level of stereoselectivity using trimethylaluminum (as shown in the table for AlMe ₃ ), while (S)-epimers of 5'-Mepurines can be synthesized stereoselectively using dimethylzinc (as shown in the table for ZnMe ₂ ).

- 78 043057

2b.f 3b-f 4b-f

2b: B = U

2d: B = C ^Ac

2e: B = A ^Bz

2f: B = G ^bBu

AlMe ₃ (S:R)

Solvent/T (°C) And _With Ac A ^Bz G ^IBu THF (0 to k.t.) 12:1 9:1 1:1 Slow 1:2 DCM (-78 to k.t.) 3.9:1 2:1 1:1 Toluene (-78 to k.t.) 3.4:1 3:1

_ZnMe2 (S:R)

Solvent/T (°C) And _With AC _A Bz G ^IBu THF (0 to k.t.) NR Slow 2:1 DCM/Toluene (-78 to kt) So slow 9:1 3:1

Stereoselectivity due to solvent.

The tables show that (S)-epimers of 5'-Me-pyrimidine nucleosides can be synthesized with a high level of stereoselectivity using trimethylaluminum in THF (as shown in the tables for AlMe ₃ ), while (S)-epimers of 5'-Me- purines can be synthesized stereoselectively using dimethylzinc in a non-coordinating solvent (as shown in the tables for

ZnMe ₂ ). An equimolar mixture of purine stereoisomers can be prepared with trimethylaluminum in either THF (for A-derivatives) or in a non-coordinating solvent (DCM) (for G-derivatives).

AlMe ₃ (S:R)

Solvent/T (°C) A ^Br And Dielectric constant (ε) Toluene (-78 to k.t.) 3:1 3.4:1 2.38 DCM (-78 to k.t.) 2:1 3.9:1 8.93 dioxane 1.9:1 2.25 Ether 1.6:1 4.33 THF (0 to k.t.) 1:1 12:1 7.58

ZnMe2(S:R)

Dependence of stereoselectivity on 2'-substitution: the coordinating and bulkier 2'-OMe substituent provided better selectivity than the smaller, non-coordinating 2'-F or 2'-H.

- 79 043057

b. Procedure for the reaction of 5'-oxo-nucleosides with trimethylaluminum.

2a-f Over-f 4a-f

2a: Bx \u003d U, X \u003d OMe 2b: Bx \u003d U, X \u003d F 2c: Bx \u003d T, X \u003d H 2d: Bx \u003d C ^Ac , X \u003d OMe 2e: Bx \u003d A ^Bz , X \u003d OMe 2f: B \u003d O ^kVi , X \u003d OMe

5'-(S)-C-methyl-3'-O-[(1,1-dimethylethyl)dimethylsilyl]-2'-O-methyluridine 3a and 5'-(R)-C-methyl-3'-O [(1,1-dimethylethyl)dimethylsilyl]-2'-O-methyluridine 4а.

A solution of crude aldehyde 2a containing approximately 68% of the designated compound (25.3 g, <68 mmol) in anhydrous THF (200 ml) was slowly added via tube over approximately 15 min to the stirred and cooled (0° C.) mixture of AlMe ₃ (2 M in heptane, 102 ml, 204 mmol) and anhydrous THF (200 ml) under argon. The cooling bath was removed, the mixture was stirred at room temperature for 17 h, cooled to 0°C and the reaction was quenched by carefully adding 500 ml of a 1:1 mixture of saturated aqueous ammonium chloride and 20% phosphoric acid, followed by the addition of 400 ml of ethyl acetate. The organic phase was separated, washed twice with brine, dried over anhydrous sodium sulfate, and the solvent was removed in vacuo to give 25.6 g of a crude residue. Flash column chromatography of the residue on a CombiFlash column of 330 g silica gel in a gradient (50 to 90%) of ethyl acetate containing 1% triethylamine in hexanes gave 3a (17.5 g, 67%) and a mixture of 3a and 4a (1 .25 g, 5%). The latter was dissolved in 15 ml of hot ethyl acetate followed by the slow addition of 15 ml of hexanes. The mixture was allowed to cool to room temperature, stirred overnight, the precipitate was filtered, washed with ethyl acetate-hexanes 1:2 and dried to give pure compound 4a (0.81 g, 65% by crystallization, 3% by reaction).

3a: 1H NMR (400 MHz, DMSO-d6): δ 0.08 (s, 6H); 0.87 (s, 9H); 1.14 (d, J=6.7 Hz, 3H); 3.33 (s, 3H); 3.68 (dd, J=1.8, 4.4 Hz, 1H); 3.76-3.84 (m, 2H); 4.27 (t, J=4.6 Hz, 1H, H2'); 5.17 (d, J=4.4 Hz, 1H, OH); 5.65 (d, J=8.1 Hz, 1H); 5.83 (d, J=4.7 Hz, 1H, H1'); 8.05 (d, J=8.1 Hz, 1H); 11.3 (s, 1H). ¹³ C NMR (126 MHz, DMSOd6): δ -4.95; -4.82; 17.78; 20.05; 25.61; 57.56; 64.73; 70.57; 82.61; 85.83; 87.96; 101.79; 140.19; 150.48; 163.08. HRMS mass/charge calc. for [C17H30N2O6Si+H]+: 387.1951; found value: 387.1962. 4a: 1H NMR (400 MHz, DMSO-d6): δ 0.09 (s, 6H); 0.88 (s, 9H); 1.10 (d, J=6.6 Hz, 3H); 3.28 (s, 3H); 3.64 (dd, J=2.2, 4.2 Hz, 1H, H4'); 3.77 (dt, J=4.6, 6.5 Hz, 1H, H5'); 3.8 6 (dd, J=4.8, 6.8 Hz, 1H, H2'); 4.4 0 (dd, J=2.2, 4.7 Hz, 1H, H3'); 5.16 (d, J=4.9 Hz, 1H, OH); 5.67 (dd, J=2.2, 8.1 Hz, 1H); 5.86 (d, J=6.8 Hz, 1H, H1'); 7.89 (d, J=8.2 Hz, 1H); 11.36 (d, J=1.8 Hz, 1H). ¹³ C NMR (126 MHz, DMSO-d6): δ 0.60; 0.69; 23.19; 25.19; 62.84; 71.50; 74.48; 87.10; 90.47; 94.80; 107.69; 145.93; 156.11; 168.37. HRMS mass/charge calc. for [CnH30N2O6Si+H] ⁺ : 387.1951; found value: 387.1960.

5'-(S)-C-Methyl-2'-deoxy-3'-O-[(1,1-dimethylethyl)dimethylsilyl]-2'-fluorouridine 3b and 5'-(R)-Cmethyl-2'- deoxy-3'-O-[(1,1-dimethylethyl)dimethylsilyl]-2'-fluoro-uridine 4b was prepared similarly by adding a solution of crude aldehyde 2b containing approximately 55% of the title compound (1.80 g, <5 mmol) in anhydrous THF (10 ml), into a mixture of AlMe ₃ (2 M in heptane, 8 ml, 16 mmol) and anhydrous THF (10 ml) at 0° C. under argon followed by stirring at room temperature overnight. The crude residue (1.88 g) was chromatographed on an 80 g silica gel CombiFlash column with 50% ethyl acetate containing 1% triethylamine in hexanes to give 3b (0.88 g, 47%) and 4b (0.17 g, 9%), as well as a small intermediate mixed fraction. 3b: ¹ H NMR (400 MHz, DMSO-d6): δ 0.088 (s, 3H); 0.094 (s, 3H); 0.86 (s, 9H); 1.19 (d, J=6.5 Hz, 3H); 3.72 (d, J=6.8 Hz, 1H); 3.75-3.83 (m, 1H); 4.31 (ddd, J=4.4, 6.8, 18.3 Hz, 1H, H3'); 5.06 (ddd, J=2.4, 4.4, 53.1 Hz, 1H, H2'); 5.20 (d, J=4.7 Hz, 1H, OH); 5.63 (d, J=8.1 Hz, 1H); 5.91 (dd, J=2.3, 16.9 Hz, 1H, H1'); 7.99 (d, J=8.1 Hz, 1H); 11.4 (s, 1H). ¹³ C NMR (126 MHz, acetone-d6): δ -4.84; -4.53; 18.75; 20.70; 26.14; 66.02; 71.29; 71.42; 87.98; 88.50; 88.77; 93.20; 94.71; 102.58; 141.37; 151.41; 163.74. ¹⁹ F NMR (376 MHz, acetone-ds): δ -207.60 (dt, J=16.6, 53.1 Hz, 1F). HRMS mass/charge calc. For [C16H ₂₇ FN ₂ O ₅ Si+H] ⁺ : 375.1752; found value: 375.1744. 4b: 1H NMR (400 MHz, DMSO-d6): δ 0.096 (s, 3H); 0.102 (s, 3H); 0.87 (s, 9H); 1.11 (d, J=6.7 Hz, 3H); 3.74-3.78 (m, 1H); 3.84-3.94 (m, 1H); 4.43 (dt, J=4.8, 12.2 Hz, 1H, H3'); 5.10 (dt, J=4.2, 52.8 Hz, 1H, H2'); 5.21 (d, J=4.7 Hz, 1H, OH); 5.65 (d, J=8.1 Hz, 1H); 5.94 (dd, J=4.0, 15.8 Hz, 1H, H1'); 7.89 (d, J=8.1 Hz, 1H); 11.4 (s, 1H). ¹³ C NMR (126 MHz, CD3OD): δ -4.56; - 4.10; 19.11; 19.89; 2 6.45; 67.67; 7 0.62; 7 0.74; 88.28; 8 8.55; 89.99; 92.89; 94.40; 103.39; 142.71; 152.28; 165.67. ¹⁹ F NMR (376 MHz, acetone^): δ -208.28 (dt, J=14.3, 52.8 Hz, 1F). HRMS mass/charge calc. for [C ₁₆ H ₂₇ FN ₂ O ₅ Si+H] ⁺ : 375.1752; found value: 375.1760.

5'-(S)-C-methyl-3'-O-[(1,1-dimethylethyl)dimethylsilyl]thymidine 3c and 5'-(R)-C-methyl-3'-O-[(1,1dimethylethyl )dimethylsilyl]thymidine 4c was obtained similarly by adding a solution of crude

- 80 043057 aldehyde 2c, containing approximately 60% of the designated compound (1.21 g, <3.4 mmol) in anhydrous THF (10 ml), in a mixture of AlMe ₃ (2 M in heptane, 6 ml, 12 mmol) and anhydrous THF (10 ml) at 0° C. under argon followed by stirring at room temperature overnight. The crude residue (1.21 g) was chromatographed on 2 consecutive flash silica gel columns (CombiFlash, 80 g and 24 g) with a gradient (80 to 100%) of ethyl ether containing 1% triethylamine in hexanes. Fractions containing the separated epimers were isolated separately, and intermediate mixed fractions were combined and chromatographed on a second column. 0.65 g (52%) 3c and 0.11 g (9%) 4c were obtained, as well as a small intermediate mixed fraction. 3c: 1H NMR (400 MHz, DMSO-d6): δ 0.07 (s, 6H); 0.86 (s, 9H); 1.13 (d, J=6.5 Hz, 3H); 1.76 (d, J=1.0 Hz, 3H); 2.01 (ddd, J=3.0, 6.0, 13.2 Hz, 1H, H2'A); 2.13 (ddd, J=5.9, 7.7, 13.4 Hz, 1H, H2' _in ); 3.58 (t, J=2.7 Hz, 1H, H4'); 3.74-3.83 (m, 1H, H5'); 4.40 (quintet, J=2.8 Hz, 1H, H3'); 5.02 (d, J=4.6 Hz, 1H, OH); 6.15 (dd, J=6.0, 7.7 Hz, 1H, H1'); 7.84 (d, J=1.2 Hz, 1H); 11.27 (s, 1H). ¹³ C NMR (126 MHz, ACN-d3): δ -4.58; -4.39; 12.72; 18.63; 20.65; 26.20; 41.14; 67.55; 73.94; 85.97; 97.74; 111.06; 137.89; 151.76; 165.10. HRMS mass/charge calc. for [C17H30N2O5Si+Na]+: 393.1822; found value: 393.1825. 4c: 1H NMR (400 MHz, DMSO-d6): δ 0.081 (s, 3H); 0.084 (s, 3H); 0.87 (s, 9H); 1.10 (d, J=6.5 Hz, 3H); 1.76 (d, J=1.1 Hz, 3H); 1.95 (ddd, J=1.7, 5.5, 13.1 Hz, 1H, H2'a); 2.14 (ddd, J=5.4, 9.0, 13.2 Hz, 1H, H2'b); 3.55 (dd, J=1.6, 4.7 Hz, 1H, H4'); 3.7 3 (dt, J=4.9, 6.4 Hz, 1H, H5'); 4.49 (dt, J=1.4, 5.3 Hz, 1H, H3'); 5.03 (d, J=5.0 Hz, 1H, OH); 6.15 (dd, J=5.5, 8.9 Hz, 1H, HI'); 7.66 (d, J=1.2 Hz, 1H); 11.29 (s, 1H). ¹³ C NMR (126 MHz, ACN-d3): δ -4.52; -4.25; 12.67; 18.55; 20.21; 26.19; 41.10; 68.26; 72.46; 86.07; 92.35; 111.27; 137.75; 151.83; 165.09. HRMS mass/charge calc. for [C17H ₃₀ N ₂ O ₅ Si+H]+: 371.2002; found value: 371.1992.

N-Acetyl-5'-(S)-C-methyl-3'-O-[(1,1-dimethylethyl)dimethylsilyl]-2'-O-methyl-cytidine 3d was obtained similarly by adding a solution of crude aldehyde 2d containing approx. 56% of the designated compound (2.80 g, <6.8 mmol) in anhydrous THF (20 ml), to a mixture of AlMe3 (2 M in heptane, 12 ml, 24 mmol) and anhydrous THF (20 ml) at 0° C under argon, followed by stirring at room temperature overnight. The crude residue (3.03 g) was chromatographed on 2 consecutive silica gel flash columns (CombiFlash, 80 g and 40 g) with a gradient (70 to 100%) of ethyl acetate in hexanes to give 1.09 g (37 %) less polar (S)-epimer 3d, as well as 0.60 g of a more polar fraction containing a mixture of 3d and 4d, which was not further separated. 3d: ¹ H NMR (400 MHz, DMSO-d6): δ 0.05 (s, 6H); 0.85 (s, 9H); 1.21 (d, J=6.5 Hz, 3H); 2.09 (s, 3H); 3.43 (s, 3H); 3.70-3.76 (m, 2H); 3.77-3.85 (m, 1H); 4.21 (dd, J=4.8, 7.0 Hz, 1H); 5.19 (d, J=4.4 Hz, 1H); 5.83 (d, J=2.0 Hz, 1H); 7.18 (d, J=7.5 Hz, 1H); 8.58 (d, J=7.5 Hz, 1H); 10.90(s, 1H).

Note: N-acetylcytidines are not very stable on a silica gel column in the presence of triethylamine and tend to disproportionate to form N-unprotected and N-diacetylated derivatives. Thus, triethylamine was not used to separate isomers 3d and 4d. However, the addition of triethylamine is useful for better isomer separation in TLC.

N-Benzoyl-5'-(S)-C-methyl-3'-O-[(1,1-dimethylethyl)dimethylsilyl]-2'-O-methyladenosine 3е and N-benzoyl-5'-(R)- C-methyl-3'-O-[(1,1-dimethylethyl)dimethylsilyl]-2'-O-methyladenosine 4e was prepared similarly by adding a solution of crude aldehyde 2e containing approximately 50% of the title compound (16.9 g, <34 mmol) in anhydrous THF (100 ml), into a mixture of AlMe3 (2 M in heptane, 51 ml, 102 mmol) and anhydrous THF (100 ml) at 0°C under argon, followed by stirring at room temperature overnight. The crude residue (15.9 g) was chromatographed on a silica gel flash column (CombiFlash, 220 g) with a gradient (70 to 100%) of ethyl acetate in hexanes to give a purified epimer mixture (~1:1): 8.52 d, 49%. The isomers were further separated by preparative C18 RP-HPLC using a Gilson PLC 2020 purification system: 1 g of the mixture was injected and eluted isocratically with 25 mM triethylammonium bicarbonate and 65% acetonitrile. The appropriate fractions with purity >95% according to HPLC were combined and evaporated to dryness to give 0.15 g of 3e (dr >97%) and 0.25 g of pure stereoisomer 4e. 3f: 1H NMR (400 MHz, DMSO-d6): δ 0.124 (s, 3H); 0.126 (s, 3H); 0.91 (s, 9H); 1.17 (d, J=6.4 Hz, 3H); 3.32 (s, 3H); 3.813.90 (m, 2H); 4.40 (dd, J=4.9, 5.7 Hz, 1H); 4.54 (dd, J=3.2, 4.5 Hz, 1H); 5.19 (d, J=5.7 Hz, 1H); 6.16 (d, J=5.7 Hz, 1H); 7.55 (t split, J=7.8 Hz, 2n); 7.64 (t split, J=7.4 Hz, 1H); 8.03 (d, J=1.4 Hz, 1H); 8.05(s, 1H); 8.76 (s, 1H); 8.80 (s, 1H); 11.23(s, 1H). 4e: 1H NMR (500 MHz, DMSO-d6): δ 0.146 (s, 3H); 0.147 (s, 3H); 0.93(s, 9H); 1.10 (d, J=6.4 Hz, 3H); 3.25 (s, 3H); 3.75 (dd, J=1.1, 5.6 Hz, 1H); 3.89 (sextet, J=5.7 Hz, 1H); 4.60 (dd, J=4.5, 7.4 Hz, 1H); 4.63 (dd, J=1.2, 4.6 Hz, 1H); 5.32 (d, J=4.7 Hz, 1H); 6.11 (d, J=7.4 Hz, 1H); 7.55 (t, J=8.0 Hz, 2H); 7.64 (t split, J=7.5 Hz, 1H); 8.03 (d, J=1.4 Hz, 1H); 8.05 (d, J=0.9 Hz, 1H); 8.76 (s, 1H); 8.77 (s, 1H); 11.23(s, 1H).

N-Isobutyryl-5'-(S)-C-methyl-3'-O-[(1,1-dimethylethyl)dimethylsilyl]-2'-O-methylguanosine 3f and Nisobutyryl-5'-(R)-C- methyl 3'-O-[(1,1-dimethylethyl)dimethylsilyl]-2'-O-methyl-guanosine 4f was prepared similarly by adding a solution of crude aldehyde 2f containing approximately 53% of the title compound (1.63 g, <3 .4 mmol) in anhydrous DCM (10 mL), into a mixture of AlMe3 (2 M in heptane, 10 mL, 20 mmol) and anhydrous DCM (10 mL) at -78°C under argon followed by slow (bath) heating to room temperature overnight. Chromatography was carried out on the crude

- 81 043057 residue (1.59 g) on a silica gel flash column (CombiFlash, 40 g) with a gradient (from 0 to 4%) of methanol in chloroform to give 0.14 g of the less polar (R)-isomer 4f, 0.36 g of an intermediate mixture of 3f and 4f and 0.25 g of the more polar (S) isomer 3f. The intermediate fraction was separated on a second column isocratically (CombiFlash, 40 g) with 3% methanol in chloroform to give an additional 0.20 g 4f and 0.16 g 3f Total yield 3f: 0.34 g (23%) and 4f: 0.41 g (27%). 3f: 1H NMR (400 MHz, DMSO-d ₆ ): δ 0.10 (s, 3H); 0.11 (s, 3H); 0.89 (s, 9H); 1.11 (d, J=6.8 Hz, 6H); 1.12 (d, J=6.4 Hz, 3H); 2.77 (septet, J=6.8 Hz, 1H); 3.29 (s, 3H); 3.76 (t, J=2.6 Hz, 1H); 3.79-3.87 (m, 1H); 4.20 (dd, J=4.8, 6.3 Hz, 1H); 4.44 (dd, J=2.6, 4.6 Hz, 1H); 5.12 (d, J=4.6 Hz, 1H); 5.88 (d, J=6.3 Hz, 1H); 8.36 (s, 1H); 11.61 (s, 1H); 12.10 (s, 1H). 4f: 1H NMR (400 MHz, DMSO-d ₆ ): δ 0.12 (s, 3H); 0.13 (s, 3H); 0.90 (s, 9H); 1.08 (d, J=6.4 Hz, 3H); 1.11 (d, J=6.8 Hz, 6H); 2.76 (septet, J=6.8 Hz, 1H); 3.25 (s, 3H); 3.66 (d, J=5.6 Hz, 1H); 3.76 (sextet, J=6.0 Hz, 1H); 4.36 (dd, J=4.6, 7.8 Hz, 1H); 4.54 (d, J=4.5 Hz, 1H); 5.16 (d, J=5.1 Hz, 1H); 5.83 (d, J=7.8 Hz, 1H); 8.32(s, 1H); 11.60 (s, 1H); 12.10 (s, 1H).

With. Procedure for the stereoselective reaction of purine-5'-oxonucleosides with dimethyl zinc.

N-Benzoyl-5'-(S)-C-methyl-3'-O-[(1,1-dimethylethyl)dimethylsilyl]-2'-O-methyladenosine 3e.

A solution of crude aldehyde 2e containing approximately 55% of the designated compound (1.69 g, <3.4 mmol) in anhydrous DCM (10 ml) was slowly added dropwise over about 20 min to a stirred and cooled (-78°C) a mixture of ZnMe ₂ (2 M in toluene, 6 ml, 12 mmol) and anhydrous DCM (10 ml) under argon. The solution was allowed to slowly warm (in the bath) to room temperature overnight, cooled to 0° C. and quenched by careful addition of 10% phosphoric acid. The organic phase was separated, washed with 5% saline and dried over anhydrous sodium sulfate. The solvent was removed in vacuo and the crude residue (1.57 g) was purified by flash chromatography on a 40 g silica gel CombiFlash column with a gradient (70 to 100%) of ethyl acetate in hexanes to give 3e (0.96 g, 55%) at ~90% diastereomeric purity. The compound was dissolved in 5 ml of diethyl ether, and then hexane (4 ml) was added slowly with stirring, which started crystallization. The mixture was stirred at room temperature overnight, filtered and the solid was washed twice with 1:2 ether-hexanes to give 0.69 g (73% by crystallization) of 3e at ~97% diastereomeric purity.

N-Isobutyryl-5'-(S)-C-methyl-3'-O-[(1,1-dimethylethyl)dimethylsilyl]-2'-O-methylguanosine 3f was prepared similarly by adding a solution of crude aldehyde 2f containing approximately 53 % of the designated compound (1.63 g, <3.4 mmol) in anhydrous DCM (10 mL), to a mixture of ZnMe2 (2 M in toluene, 8.5 mL, 17 mmol) and anhydrous DCM (10 mL) at -78 °C under argon followed by slow (on bath) warming to room temperature overnight. Chromatography of the crude residue (1.56 g) containing 3f and 4f in a ~3:1 ratio was carried out on a silica gel flash column (CombiFlash, 40 g) with 2% methanol in chloroform to give 0.13 g of a mixture of 4f and 3f followed by purification of 3f 0.48 g (32%).

d. Stereospecific interconversion of 5'-alkyl epimers

'-(S)-C-Methyl-5'-O-mesyl-3'-O-[(1,1-dimethylethyl)dimethylsilyl] -2'-O-methyluridine 5 a.

Methanesulfonyl chloride (5.5 ml, 72 mmol) was added dropwise over a period of ~5 min to a chilled (0° C.) and stirred solution of (S)-epimer (3a) (8.32 g, 21.6 mmol) and anhydrous pyridine (5.8 ml, 72 mmol) in anhydrous DCM (80 ml) under argon. The cooling bath was removed, the mixture was stirred at room temperature for 48 h, cooled to 0°C and quenched by careful addition of saturated sodium bicarbonate solution (200 ml). The cooling bath was removed, the mixture was stirred vigorously at room temperature for 1 h, the organic phase was separated, washed successively with 10% phosphoric acid, 5% brine, twice, and dried.

- 82 043057 over anhydrous sodium sulfate. The solvent was removed in vacuo and the residue was dried under high vacuum to give substantially pure 5a (9.79 g, 98%) as an orange foam. 5a:

1H NMR (400 MHz, DMSO-d ₆ ): δ 0.09 (s, 3H); 0.10 (s, 3H); 0.87 (s, 9H); 1.42 (d, J=6.5 Hz, 3H); 3.20 (s, 3H);

3.33 (s, 3H); 3.86 (t, J=4.9 Hz, 1H); 3.89 (t, J=4.9 Hz, 1H); 4.29 (t, J=5.1 Hz, 1H); 4.91 (dt, J=6.4, 11.3 Hz, 1H);

5.68 (dd, J=2.2, 8.1 Hz, 1H); 5.86 (d, J=4.7 Hz, 1H); 7.65 (d, J=8.2 Hz, 1H); 11.42 (s, 1H).

6,9-Epoxy-2Н,6Н-pyrimido[2, 1-b][1,3]oxazocin-2-one-7,8,10-trihydro-9-(R)-methyl-8-О-[ (1,1-dimethylethyl)dimethylsilyl]-7-methoxy-[6K-(6a,7a,8a,9a)] 6a.

A solution of mesylate 5a (2.33 g, 5 mmol) and DBU (1.5 ml, 10 mmol) in anhydrous DMSO (10 ml) was stirred at 60° C. under argon for 27 h, cooled to 0° C. and ethyl acetate (40 ml) was added followed by 5% aqueous NaCl solution (80 ml). The organic phase was separated, washed with a mixture of 1:15% NaCl solution and 10% aqueous phosphoric acid solution, and then with 5% NaCl solution, saturated sodium bicarbonate and saturated NaCl solution. After drying over anhydrous sodium sulfate, the solvent was removed in vacuo to give the crude anhydrous product 6a (1.57 g), which was heated under reflux with 30 ml of diethyl ether for 45 min, cooled to room temperature, stirred for 2 h, white the precipitate was filtered, washed twice with diethyl ether and dried to give 0.88 g (48%) of pure compound 6a. 6a: 1H NMR (400 MHz, DMSOd6): δ 0.06 (s, 3H); 0.7 (s, 3H); 0.84 (s, 9H); 1.36 (d, J=6.7 Hz, 3H); 3.29 (s, 3H); 4.13 (dd, J=0.8, 6.0 Hz, 1H); 4.27(s, 1H); 4.32 (q, J=6.8 Hz, 1H); 4.62 (d, J=5.9 Hz, 1H); 5.78(s, 1H); 5.89 (d, J=7.4 Hz, 1H); 7.95 (d, J=7.4 Hz, 1H). ¹³ C NMR (126 MHz, DMSOA): δ -5.22; -4.81; 16.79; 17.91; 25.59; 57.93; 71.42; 81.77; 86.33; 91.16; 96.24; 109.09; 142.58; 156.29; 170.67.

5'-(R)-C-Methyl-3'-O-[(1,1-dimethylethyl)dimethylsilyl]-2'-O-methyl-uridine 4a (from 5a).

A solution of mesylate 5a (3.72 g, 8 mmol), DBU (2.4 ml, 16 mmol) and water (10 ml) in THF (50 ml) was refluxed under argon for 67 h. vacuum, the residue was partitioned between ethyl acetate (120 ml) and a mixture of 80 ml of 5% NaCl and 30 ml of 10% phosphoric acid, the organic phase was separated, washed twice with 5% NaCl, and then with saturated NaCl. After drying over anhydrous sodium sulfate, the crude residue (3.04 g) was refluxed with a mixture of 30 ml of diethyl ether and 15 ml of hexanes for 1 h, cooled to room temperature, stirred overnight, the white precipitate was filtered and washed twice ether-hexanes 1:1 to give 2.32 g (75%) of 4a.

The method described in this example can be used to synthesize various 5'-alkyl nucleosides for any therapeutic application (eg, antiviral and anticancer applications), including oligonucleotides and small molecules.

Example 9 Steric blocking of phosphorodiester (PO), phosphorothioate (PS), and phosphorodithioate (PS2) bonds by introducing a 4'- and 5'-modified nucleotide at the 3' end of the PO, PS, or PS2 bond

The present inventors have found that the introduction of a nucleotide with a 4' modification and/or a 5' modification at the 3' end of the phosphodiester (PO), phosphorothioate (PS), and/or phosphorodithioate (PS2) linkage of a dinucleotide at any position of a single-stranded or double-stranded oligonucleotide can have a steric effect on the internucleotide bond and, therefore, protect it from the action of nucleases or make it resistant to them.

In this example, the in vitro gene silencing activity of an F7-targeted siRNA that contains 4'-O-methylated or 5'-methylated nucleotides at the selected position was evaluated, and the results are shown in Table 1. eleven.

Table 11

In vitro gene silencing activity in F7-targeted siRNA that contains 4'-O-methylated or 5'-methylated nucleotides at a selected position

10 nM 0.1 nM ID duplex semantic thread Antisense strand Avg. SD Avg. SD AD-60347 CfsasGfgAfoCfaUfCfUfcAf aGfuCfuUfaCfL96 gsUfsaAfgAfcUfuGfagaUfgA fuCfcUfgsgsc 13.2 1.9 81.4 7.2 AD-63931 CfsasGfgAiuCfaUfCIUfcAf aGfuCfUUfaAfL96 usUfsaAfgAfcUfuGfagaUfgA fuCfcUfgsgsc 5.7 0.7 41.2 3.7 AD-69122 CfsasGfgAfoCfaUfCfUfcAf aGfuCfU(Ufm)aAfL96 usUfsaAfgAfcUfuGfagaUfgA fuCfcUfgsgsc 4.9 0.7 51.7 7.7 AD-69123 CfsasGfgAfoCfaUfCfUfcAf aGfuCfU(UfmR)aAfL96 usUfsaAfgAfcUfuGfagaUfgA fuCfcUfgsgsc 5.4 0.9 60.6 9.5

- 83 043057

AD-69124 CfsasGfgAfuCfaUfCfUfcAf aGfuCfU (u5m)aAfL96 usUfsaAfgAfcUfuGfagaUfgA fuCfcUfgsgsc 5.4 1.1 53.9 6.0 AD-69125 CfsasGfgAfuCfaUfCfUfcAf aGfuCfU(u5mR)aAfL96 usUfsaAfgAfcUfuGfagaUfgA fuCfcUfgsgsc 5.5 0.6 55.1 6.0 AD-69126 CfsasGfgAfuCfaUfCfUfcAf aGfuCfU(T5m)aAfL96 usUfsaAfgAfcUfuGfagaUfgA fuCfcUfgsgsc 4.9 0.4 50.7 6.9 AD-69127 CfsasGfgAfuCfaUfCfUfcAf aGfuCfU(T5mR)aAfL96 usUfsaAfgAfcUfuGfagaUfgA fuCfcUfgsgsc 3.5 0.5 34.0 5.5 AD-69128 CfsasGfgAfuCfaUfCfUfcAf aGfuCfU (Ufo)aAfL96 usUfsaAfgAfcUfuGfagaUfgA fuCfcUfgsgsc 4.6 0.3 30.4 4.1 AD-69129 CfsasGfgAfuCfaUfCfUfcAf aGfuCfU(dTo)aAfL96 usUfsaAfgAfcUfuGfagaUfgA fuCfcUfgsgsc 5.3 0.2 40.9 4.3 AD-69130 CfsasGfgAfuCfaUfCfUfcAf aGfuCfUuaAfL96 usUfsaAfgAfcUfuGfagaUfgA fuCfcUfgsgsc 6.6 0.7 51.6 5.5 AD-69131 CfsasGfgAfuCfaUfCfUfcAf aGfuCfUdTaAfL96 usUfsaAfgAfcUfuGfagaUfgA fuCfcUfgsgsc 4.7 0.3 44.8 7.1 AD-63934 CfsasGfgAfuCfaUfCfUfcAf aGfuCfuUfaAfL96 usUfsaAfgAfcUfuGfagaUfgA fUCfcUfgsgsc 5.2 0.7 49.2 8.4 AD-69132 CfsasGfgAfuCfaUfCfUfcAf aGfuCfuUfaAfL96 usUfsaAfgAfcUfuGfagaUfgA fU(Cfm)cUfgsgsc 6.9 1.2 57.7 7.1 AD-69133 CfsasGfgAfuCfaUfCfUfcAf aGfuCfuUfaAfL96 usUfsaAfgAfcUfuGfagaUfgA fU(CfmR)cUfgsgsc 6.5 0.9 58.2 6.1 AD-69134 CfsasGfgAfuCfaUfCfUfcAf aGfuCfuUfaAfL96 usUfsaAfgAfcUfuGfagaUfgA fU(c5m)cUfgsgsc 5.4 0.7 53.1 5.7 AD-69135 CfsasGfgAfuCfaUfCfUfcAf aGfuCfuUfaAfL96 usUfsaAfgAfcUfuGfagaUfgA fU (c5mR)cUfgsgsc 5.5 0.8 47.8 5.2 AD-69136 CfsasGfgAfuCfaUfCfUfcAf aGfuCfuUfaAfL96 usUfsaAfgAfcUfuGfagaUfgA fU(Cfo)cUfgsgsc 5.5 0.8 45.3 4.1 AD-69137 CfsasGfgAfuCfaUfCfUfcAf aGfuCfuUfaAfL96 usUfsaAfgAfcUfuGfagaUfgA fU(dCo)cUfgsgsc 5.1 0.4 48.6 5.9 AD-69138 CfsasGfgAfuCfaUfCfUfcAf aGfuCfuUfaAfL96 usUfsaAfgAfcUfuGfagaUfgA fUccUfgsgsc 5.1 0.8 48.9 4.9 AD-69139 CfsasGfgAfuCfaUfCfUfcAf aGfuCfuUfaAfL96 usUfsaAfgAfcUfuGfagaUfgA fUdTcUfgsgsc 5.1 0.6 49.6 7.1

- 84 043057

(X Oj ^s > ⁴ N -o Xx p F OH 2'-fluoro-5'-(S)-methyl-uridine-3'phosphate (Ufm) ft ° 0.^0 ^F SH 2'-fluoro-5'-(S)-methyl-uridine-3'-phosphorothioate (Ufms) ft O/RU N^O SH 2'-fluoro-5'-(R)-methyl-uridine-3'- Phosphorothioate (UfmRs) ft 0# N^0 0F Oh 2'-fluoro-5'-(R)-methyl-uridine-3'-phosphate (UfmR) (X o№ N^O ν-°χ P Oh 2'-0-methyl-5'-(R)-methyl-uridine- 3'-phosphate (u5mR) I. p0.. Oh 2'-O-methyl-5'-(S)-methyl-uridine-3'-phosphate (u5m) nh ₂ X θΑ/ N^O b HO. p o h 0 2'-O-methyl-5'-(S)-methyl-cytidine 3'-phosphate (c5m) nh ₂ ft o IW N^O X no. po. A o 2'-O-methyl-5'-(R)-methyl-cytidine-3'-phosphate (c5mR) Ϋτ _z no °X°J 0 O=P-OH 5'-(S)-methyl-deoxythymidine-3'- phosphate (T5m) Wow ο ν ο ο Ο=ρ-ΟΗ 5'-(R)-methyl-deoxythymidine-3'-phosphate (T5mR) nh _{2 O} W ⁴ N O A SK PF ^p \ OH 2'-fluoro-5'-(S)-methyl-cytoZine-3'-phosphate (Cfm) νη ₂ Ga 0^ Ρ ^{F ρ} \ SH: 2'-φτορ-5'-(S)-methyl-cytoZine-3'-phosphorothioate (Cfms) nh ₂ A 0^' yoj O^O ^F OH 2'-fluoro-5'-(R)-methyl-cytoZine-3'phosphate (CfmR) νη ₂ Λ CK Ρ F SH 2'-φτορ-5'-(R)-methyl-cytoZine-3'phosphorothioate (CfmRs)

-85043057

Example 10 Chirality-Dependent Glycolic Nucleic Acid (GNA) Activity in siRNA Duplexes.

Chemical modifications in siRNA duplexes are necessary to stabilize these molecules against degradation by nucleases, to facilitate their entry into cells and to influence the formation of active RISC, as well as for RNAi-mediated target silencing. Thermal stability-disrupting modifications introduced at certain positions in the siRNA duplex can increase efficiency by enhancing strand shear and/or dissociation of the sense strand upon incorporation into the RISC complex.

In this example, a three-carbon acyclic nucleic acid analogue, glycolic nucleic acid (GNA), was evaluated in the context of siRNA duplexes. GNA-containing siRNA duplexes were synthesized. The (S)-GNA oligomers formed homoduplexes whose structure was similar to the typical A-form RNA duplex and paired with RNA, but not DNA, within high A/T sequences. The thermostability and nuclease resistance of siRNA duplexes containing (S)- or (R)-GNA were studied. Structural studies using X-ray crystallography have provided additional information about the effect of such GNA substituents in RNA duplexes. In biological studies, the chirality-dependent activity of gene silencing in siRNA duplexes containing GNA was evaluated.

- 86 043057

Duplex ID Sense thread (from 5' to 3') Antisense strand (from 5' to 3') Design AD57727.66 AfsasCfaGfuGfiiUfCfUfiiGfcUfcUfaUfaAfL96 usUfsaUfaGfaGfcAfagaAfcAfcUfgUfiisusu initial AD68368.2 AfsasCfaGfuGfiiUfCfUfiiGfcUfcUfaUfaAfL96 (Tgns)UfsaUfaGfaGfcAfagaAfcAfcUfgUfiisusu AS:gN@Nl AD68369.2 AfsasCfaGfuGfiiUfCfUfiiGfcUfcUfaUfaAfL96 P(Tgns)UfsaUfaGfaGfcAfagaAfcAfcUfgUfususu AS: 5'-p, gN @ N2 AD62896.4 AfsasCfaGfuGfiiUfCfUfiiGfcUfcUfaUfaAfL96 us(Tgns)aUfaGfaGfcAfagaAfcAfcUfgUfususu AS:gN@N2 AD68370.2 AfsasCfaGfuGfiiUfCfUfiiGfcUfcUfaUfaAfL96 usUfs(Agn)UfaGfaGfcAfagaAfcAfcUfgUfususu AS:gN@N3 AD68371.2 AfsasCfaGfuGfiiUfCfUfiiGfcUfcUfaUfaAfL96 usUfsa(Tgn)aGfaGfcAfagaAfcAfcUfgUfususu AS:gN@N4 AD68372.2 AfsasCfaGfuGfiiUfCfUfiiGfcUfcUfaUfaAfL96 usUfsaUf(Agn)GfaGfcAfagaAfcAfcUfgUfususu AS:gN@N5 AD68373.2 AfsasCfaGfuGfiiUfCfUfiiGfcUfcUfaUfaAfL96 usUfsaUfa(Ggn)aGfcAfagaAfcAfcUfgUfiisusu AS:gN@N6 AD68374.2 AfsasCfaGfuGfiiUfCfUfiiGfcUfcUfaUfaAfL96 usUfsaUfaGf(Agn)GfcAfagaAfcAfcUfgUfususu AS:gN@N7 AD68375.2 AfsasCfaGfuGfiiUfCfUfiiGfcUfcUfaUfaAfL96 usUfsaUfaGfa(Ggn)cAfagaAfcAfcUfgUfiisusu AS:gN@N8 AD68376.2 AfsasCfaGfuGfiiUfCfUfiiGfcUfcUfaUfaAfL96 usUfsaUfaGfaGf(Cgn)AfagaAfcAfcUfgUfiisusu AS:gN@N9 AD68377.2 AfsasCfaGfuGfiiUfCfUfiiGfcUfcUfaUfaAfL96 usUfsaUfaGfaGfc(Agn)agaAfcAfcUfgUfiisusu AS: gN@N10 AD68378.2 AfsasCfaGfuGfiiUfCfUfiiGfcUfcUfaUfaAfL96 usUfsaUfaGfaGfcAf(Agn)gaAfcAfcUfgUfususu AS:gN@Nll AD68379.2 AfsasCfaGfuGfiiUfCfUfiiGfcUfcUfaUfaAfL96 usUfsaUfaGfaGfcAfa(Ggn)aAfcAfcUfgUfususu AS: gN@N12 AD68380.2 AfsasCfaGfuGfiiUfCfUfiiGfcUfcUfaUfaAfL96 usUfsaUfaGfaGfcAfag(Agn)AfcAfcUfgUfususu AS: gN@N13 AD68381.2 AfsasCfaGfuGfiiUfCfUfiiGfcUfcUfaUfaAfL96 usUfsaUfaGfaGfcAfaga(Agn)cAfcUfgUfususu AS: gN@N14 AD68382.2 AfsasCfaGfuGfiiUfCfUfiiGfcUfcUfaUfaAfL96 usUfsaUfaGfaGfcAfagaAf(Cgn)AfcUfgUfiisusu AS: gN@N15 AD68383.2 AfsasCfaGfuGfiiUfCfUfiiGfcUfcUfaUfaAfL96 usUfsaUfaGfaGfcAfagaAfc(Agn)cUfgUfususu AS: gN@N16 AD68384.2 AfsasCfaGfuGfiiUfCfUfiiGfcUfcUfaUfaAfL96 usUfsaUfaGfaGfcAfagaAfcAf(Cgn)UfgUfiisusu AS: gN@N17 AD68385.2 AfsasCfaGfuGfiiUfCfUfiiGfcUfcUfaUfaAfL96 usUfsaUfaGfaGfcAfagaAfcAfc(Tgn)gUfususu AS: gN@N18 AD68386.2 AfsasCfaGfuGfiiUfCfUfiiGfcUfcUfaUfaAfL96 usUfsaUfaGfaGfcAfagaAfcAfcUf(Ggn)Ufususu AS: gN@N19 AD68387.2 AfsasCfaGfuGfiiUfCfUfiiGfcUfcUfaUfaAfL96 usUfsaUfaGfaGfcAfagaAfcAfcUfg(Tgn)ususu AS: gN@N20 AD68388.2 AfsasCfaGfuGfuUfCfUfuGfcUfcUfaUfaAfL96 usUfsaUfaGfaGfcAfagaAfcAfcUfgUf(Tgns)usu AS:gN@N21 AD68389.2 AfsasCfaGfuGfiiUfCfUfiiGfcUfcUfaUfaAfL96 usUfsaUfaGfaGfcAfagaAfcAfcUfgUfus(Tgns)u AS: gN@N22 AD68390.2 AfsasCfaGfuGfiiUfCfUfiiGfcUfcUfaUfaAfL96 usUfsaUfaGfaGfcAfagaAfcAfcUfgUfusus(Tgn) AS: gN@N23 AD68391.2 (Agns)asCfaGfuGfiiUfCfUfiiGfcUfcUfaUfaAfL96 usUfsaUfaGfaGfcAfagaAfcAfcUfgUfiisusu S: gN@Nl AD68392.2 Afs(Agns)CfaGfiiGfiiUfCfUfuGfcUfcUfaUfaAfL96 usUfsaUfaGfaGfcAfagaAfcAfcUfgUfiisusu S: gN@N2 AD68393.2 Afsas(Cgn)aGfuGfiiUfCfUfiiGfcUfcUfaUfaAfL96 usUfsaUfaGfaGfcAfagaAfcAfcUfgUfiisusu S: gN@N3 AD68394.2 AfsasCf(Agn)GfiiGfiiUfCfUfuGfcUfcUfaUfaAfL96 usUfsaUfaGfaGfcAfagaAfcAfcUfgUfiisusu S: gN@N4 AD68395.2 AfsasCfa(Ggn)uGfiiUfCfUfiiGfcUfcUfaUfaAfL96 usUfsaUfaGfaGfcAfagaAfcAfcUfgUfiisusu S: gN@N5 AD68396.2 AfsasCfaGf(Tgn)GfiiUfCfUfuGfcUfcUfaUfaAfL96 usUfsaUfaGfaGfcAfagaAfcAfcUfgUfiisusu S: gN@N6 AD68397.2 AfsasCfaGfu(Ggn)uUfCfUfiiGfcUfcUfaUfaAfL96 usUfsaUfaGfaGfcAfagaAfcAfcUfgUfiisusu S:gN@N7 AD68398.2 AfsasCfaGfiiGf(Tgn)UfCfUfuGfcUfcUfaUfaAfL96 usUfsaUfaGfaGfcAfagaAfcAfcUfgUfiisusu S: gN@N8 AD68399.2 AfsasCfaGfuGfii(Tgn)CfUfiiGfcUfcUfaUfaAfL96 usUfsaUfaGfaGfcAfagaAfcAfcUfgUfiisusu S: gN@N9 AD68400.2 AfsasCfaGfuGfiiUf(Cgn)UfiiGfcUfcUfaUfaAfL96 usUfsaUfaGfaGfcAfagaAfcAfcUfgUfiisusu S:gN@N10 AD- AfsasCfaGfuGfuUfCf(Tgn)uGfcUfcUfaUfaAfL96 usUfsaUfaGfaGfcAfagaAfcAfcUfgUfiisusu S:gN@Nll

- 87 043057

62900.1 AD68401.2 AfsasCfaGfiiGfuUfCfUf(Tgn)GfcUfcUfaUfaAfL96 usUfsaUfaGfaGfcAfagaAfcAfcUfgUfnsusu S:gN@N12 AD68402.2 AfsasCfaGfuGfuUfCfUfii(Ggn)cUfcUfaUfaAfL96 usUfsaUfaGfaGfcAfagaAfcAfcUfgUfnsusu S:gN@N13 AD68403.2 AfsasCfaGfuGfuUfCfUfuGf(Cgn)UfcUfaUfaAfL96 usUfsaUfaGfaGfcAfagaAfcAfcUfgUfnsusu S:gN@N14 AD68404.2 AfsasCfaGfuGfuUfCfUfiiGfc(Tgn)cUfaUfaAfL96 usUfsaUfaGfaGfcAfagaAfcAfcUfgUfnsusu S:gN@N15 AD68405.2 AfsasCfaGfuGfuUfCfUfuGfcUf(Cgn)UfaUfaAfL96 usUfsaUfaGfaGfcAfagaAfcAfcUfgUfiisusu S:gN@N16 AD68406.2 AfsasCfaGfuGfuUfCfUfiiGfcUfc(Tgn)aUfaAfL96 usUfsaUfaGfaGfcAfagaAfcAfcUfgUfiisusu S:gN@N17 AD68407.2 AfsasCfaGfuGfuUfCfUfuGfcUfcUf(Agn)UfaAfL96 usUfsaUfaGfaGfcAfagaAfcAfcUfgUfususu S:gN@N18 AD68408.2 AfsasCfaGfuGfuUfCfUfiiGfcUfcUfa(Tgn)aAfL96 usUfsaUfaGfaGfcAfagaAfcAfcUfgUfnsusu S:gN@N19 AD68409.2 AfsasCfaGfuGfuUfCfUfuGfcUfcUfaUf(Agn)AfL96 usUfsaUfaGfaGfcAfagaAfcAfcUfgUfiisusu S:gN@N20 AD68410.2 AfsasCfaGfuGfuUfCШfuGfcUfcUfaUfa(Agn)L96 usUfsaUfaGfaGfcAfagaAfcAfcUfgUfnsusu S:gN@N21 AD57727.66 AfsasCfaGfuGfuUfCfUfiiGfcUfcUfaUfaAfL96 usUfsaUfaGfaGfcAfagaAfcAfcUfgUfususu parent AD- 69078.1 AfsasCfaGfuGfuUfCШfuGfcUfcUfaUfa(Agn)L96 (Tgns)UfsaUfaGfaGfcAfagaAfcAfcUfgUfiisusu AS: gN@Nl; S: gN@N21 AD- 69079.1 AfsasCfaGfuGfuUfCfUfuGfcUfcUfaUf(Agn)AfL96 us(Tgns)aUfaGfaGfcAfagaAfcAfcUfgUfususu AS: gN@N2; S:gN@N20 AD- 69080.1 AfsasCfaGfuGfuUfCfUfiiGfcUfcUfa(Tgn)aAfL96 usUfs(Agn)UfaGfaGfcAfagaAfcAfcUfgUfususu AS: gN@N3; S: gN@N19 AD- 69081.1 AfsasCfaGfnGfuUfCfUfuGfcUfcUf(Agn)UfaAfL96 usUfsa(Tgn)aGfaGfcAfagaAfcAfcUfgUfususu AS: gN@N4; S: gN@N18 AD69082.1 AfsasCfaGfuGfuUfCfUfiiGfcUfc(Tgn)aUfaAfL96 usUfsaUf(Agn)GfaGfcAfagaAfcAfcUfgUfususu AS: gN@N5; S: gN@N17 AD69083.1 AfsasCfaGfuGfuUfCfUfuGfcUf(Cgn)UfaUfaAfL96 usUfsaUfa(Ggn)aGfcAfagaAfcAfcUfgUfususu AS:gN@N6; S: gN@N16 AD69084.1 AfsasCfaGfuGfuUfCfUfiiGfc(Tgn)cUfaUfaAfL96 usUfsaUfaGf(Agn)GfcAfagaAfcAfcUfgUfususu AS: gN@N7; S: gN@N15 AD69085.1 AfsasCfaGfuGfuUfCfUfuGf(Cgn)UfcUfaUfaAfL96 usUfsaUfaGfa(Ggn)cAfagaAfcAfcUfgUfususu AS: gN@N8; S: gN@N14 AD69086.1 AfsasCfaGfuGfuUfCfUfii(Ggn)cUfcUfaUfaAfL96 usUfsaUfaGfaGf(Cgn)AfagaAfcAfcUfgUfususu AS: gN@N9; S: gN@N13 AD- AfsasCfaGfuGfuUfCfUf(Tgn)GfcUfcUfaUfaAfL96 usUfsaUfaGfaGfc(Agn)agaAfcAfcUfgUfususu AS: gN@N10;

- 88 043057

69087.1 S: gN@N12 AD- 69088.1 AfsasCfaGfuGfuUfCf(Tgn)uGfcUfcUfaUfaAfL96 usUfsaUfaGfaGfcAf(Agn)gaAfcAfcUfgUfiisusu AS: gN@Nll; S: gN@Nll AD69089.1 AfsasCfaGfuGfuUf(Cgn)UfiiGfcUfcUfaUfaAfL96 usUfsaUfaGfaGfcAfa(Ggn)aAfcAfcUfgUfiisusu AS: gN@N12; S: gN@N10 AD69090.1 AfsasCfaGfuGfu(Tgn)CfUfiiGfcUfcUfaUfaAfL96 usUfsaUfaGfaGfcAfag(Agn)AfcAfcUfgUfiisusu AS: gN@N13; S: gN@N9 AD- 69091.1 AfsasCfaGfuGf(Tgn)UfCfUfuGfcUfcUfaUfaAfL96 usUfsaUfaGfaGfcAfaga(Agn)cAfcUfgUfususu AS: gN@N14; S: gN@N8 AD- 69092.1 AfsasCfaGfu(Ggn)uUfCfUfuGfcUfcUfaUfaAfL96 usUfsaUfaGfaGfcAfagaAf(Cgn)AfcUfgUfiisusu AS: gN@N15; S:gN@N7 AD69093.1 AfsasCfaGf(Tgn)GfuUfCfUfuGfcUfcUfaUfaAfL96 usUfsaUfaGfaGfcAfagaAfc(Agn)cUfgUfususu AS: gN@N16; S: gN@N6 AD- 69094.1 AfsasCfa(Ggn)uGfuUfCfUfiiGfcUfcUfaUfaAfL96 usUfsaUfaGfaGfcAfagaAfcAf(Cgn)UfgUfiisusu AS: gN@N17; S:gN@N5 AD- 69095.1 AfsasCf(Agn)GfuGfiiUfCfUfuGfcUfcUfaUfaAfL96 usUfsaUfaGfaGfcAfagaAfcAfc(Tgn)gUfususu AS: gN@N18; S:gN@N4 AD- 69096.1 Afsas(Cgn)aGfuGfuUfCfUfiiGfcUfcUfaUfaAfL96 usUfsaUfaGfaGfcAfagaAfcAfcUf(Ggn)Ufiisusu AS: gN@N19; S: gN@N3 AD69097.1 Afs(Agns)CfaGfuGfiiUfCfUfuGfcUfcUfaUfaAfL96 usUfsaUfaGfaGfcAfagaAfcAfcUfg(Tgn)ususu AS: gN@N20; S: gN@N2 AD- 69098.1 (Agns)asCfaGfuGfuUfCfUfiiGfcUfcUfaUfaAfL96 usUfsaUfaGfaGfcAfagaAfcAfcUfgUf(Tgns)usu AS: gN@N21; S: gN@Nl AD69099.1 AfsasCfaGfuGfuUfCfUfiiGfc(Tgn)cUfaUfaAfL96 usUfsaUfa(Ggn)aGfcAfagaAfcAfcUfgUfususu AS: gN@N6; S: gN@N15 AD- 69100.1 AfsasCfaGfuGfuUfCfUfiiGfcUfc(Tgn)aUfaAfL96 usUfsaUfa(Ggn)aGfcAfagaAfcAfcUfgUfususu AS: gN@N6; S: gN@N17 AD- 69101.1 AfsasCfaGfuGfuUfCfUfuGfcUf(Cgn)UfaUfaAfL96 usUfsaUfaGf(Agn)GfcAfagaAfcAfcUfgUfiisusu AS: gN@N7; S:gN@N16 AD- 69102.1 AfsasCfaGfuGfuUfCfUfiiGfcUfc(Tgn)aUfaAfL96 usUfsaUfaGf(Agn)GfcAfagaAfcAfcUfgUfiisusu AS: gN@N7; S: gN@N17 AD- 69103.1 AfsasCfaGfuGfuUfCfUfiiGfc(Tgn)cUfaUfaAfL96 usUfs(Agn)UfaGfaGfcAfagaAfcAfcUfgUfiisusu AS: gN@N3; S: gN@N15 AD- 69104.1 AfsasCfaGfuGfuUfCfUfuGfcUf(Cgn)UfaUfaAfL96 usUfs(Agn)UfaGfaGfcAfagaAfcAfcUfgUfiisusu AS: gN@N3; S: gN@N16 AD- 69105.1 AfsasCfaGfuGfuUfCfUfiiGfcUfc(Tgn)aUfaAfL96 usUfs(Agn)UfaGfaGfcAfagaAfcAfcUfgUfiisusu AS: gN@N3; S: gN@N17 AD- 69106.1 AfsasCfaGfuGfuUfCfUfiiGfc(Tgn)cUfaUfaAfL96 usUfsaUfaGfa(Ggn)cAfagaAfcAfcUfgUfususu AS: gN@N8; S: gN@N15 AD69107.1 AfsasCfaGfuGfuUfCfUfuGfcUf(Cgn)UfaUfaAfL96 usUfsaUfaGfa(Ggn)cAfagaAfcAfcUfgUfususu AS: gN@N8; S: gN@N16 AD- 69108.1 AfsasCfaGfuGfuUfCfUfiiGfcUfc(Tgn)aUfaAfL96 usUfsaUfaGfa(Ggn)cAfagaAfcAfcUfgUfususu AS: gN@N8; S: gN@N17 AD- 69109.1 Afs(Agns)CfaGfiiGfuUfCfUfuGfcUfcUfaUfaAfL96 usUfsaUfa(Ggn)aGfcAfagaAfcAfcUfgUfususu AS: gN@N6; S: gN@N2 AD- 69110.1 Afs(Agns)CfaGfuGfiiUfCfUfuGfcUfcUfaUfaAfL96 usUfsaUfaGf(Agn)GfcAfagaAfcAfcUfgUfiisusu AS: gN@N7; S: gN@N2 AD- 69111.1 AfsasCfaGf(Tgn)GfuUfCfUfuGfcUfcUfaUfaAfL96 usUfsaUfa(Ggn)aGfcAfagaAfcAfcUfgUfususu AS: gN@N6; S: gN@N6 AD- 69112.1 AfsasCfaGf(Tgn)GfuUfCfUfuGfcUfcUfaUfaAfL96 usUfsaUfaGf(Agn)GfcAfagaAfcAfcUfgUfiisusu AS: gN@N7; S: gN@N6 AD- 69113.1 AfsasCfaGfu(Ggn)uUfCfUfuGfcUfcUfaUfaAfL96 usUfsaUfa(Ggn)aGfcAfagaAfcAfcUfgUfususu AS: gN@N6; S: gN@N7 AD- 69114.1 AfsasCfaGfu(Ggn)uUfCfUfuGfcUfcUfaUfaAfL96 usUfsaUfaGf(Agn)GfcAfagaAfcAfcUfgUfiisusu AS: gN@N7; S:gN@N7 AD- 69115.1 AfsasCfaGfuGfuUfCfUfiiGfcUfcUfaUfa(Agn)L96 usUfsaUfa(Ggn)aGfcAfagaAfcAfcUfgUfususu AS: gN@N6; S: gN@N21 AD- 69116.1 AfsasCfaGfuGfuUfCfUfiiGfcUfcUfaUfa(Agn)L96 usUfsaUfaGf(Agn)GfcAfagaAfcAfcUfgUfiisusu AS: gN@N7; S: gN@N21 AD- 69117.1 AfsasCfaGfuGfuUfCf(Tgn)uGfcUfcUfaUfaAfL96 us(Tgns)aUfaGfaGfcAfagaAfcAfcUfgUfususu AS: gN@N2; S: gN@Nll AD- 69118.1 AfsasCfaGfuGfuUfCfUf(Tgn)GfcUfcUfaUfaAfL96 us(Tgns)aUfaGfaGfcAfagaAfcAfcUfgUfususu AS: gN@N2; S: gN@N12 AD- 69119.1 AfsasCfaGfuGfuUfCf(Tgn)uGfcUfcUfaUfaAfL96 usUfsaUfa(Ggn)aGfcAfagaAfcAfcUfgUfususu AS: gN@N6; S: gN@Nll AD- 69120.1 AfsasCfaGfuGfuUfCf(Tgn)uGfcUfcUfaUfaAfL96 usUfsaUfaGf(Agn)GfcAfagaAfcAfcUfgUfiisusu AS: gN@N7; S: gN@Nll AD- 69121.1 AfsasCfaGfuGfuUfCfUfuGfcUf(Cgn)UfaUfaAfL96 us(Tgns)aUfaGfaGfcAfagaAfcAfcUfgUfususu AS: gN@N2; S: gN@N16

The results are shown in FIG. 27-30.

In FIG. 27 is a graph showing in vitro TTR knockdown using single nucleotide (S)-GNA modified siRNA. TTR mRNA levels were measured in primary mouse hepatocytes after incubation with 10 nM siRNA for 24 h. TTR mRNA levels were assessed using RT-qPCR and normalized to PBS-treated cells. All results were the mean of four measurements. In FIG. 27 shows the effect of single nucleotide (S)-GNA administration on in vitro siRNA activity.

In FIG. 28A is a graph showing in vitro TTR knockdown using siRNA modified with one (S)-GNA base pair. TTR mRNA levels were measured in primary mouse hepatocytes after incubation with 10 nM siRNA for 24 h. TTR mRNA levels were assessed using RT-qPCR and normalized to PBS-treated cells. All results were the mean of four measurements. In FIG. 28B shows combination duplexes in which sense and antisense strands containing (S)-GNA single nucleotides were paired as GNA:RNA heterologous base pairs. In FIG. 28 shows the effect of introducing a single base pair of (S)-GNA on in vitro siRNA activity.

In FIG. 29 is a graph showing in vivo serum TTR levels in mice. Animals received a single dose of siRNA of 2.5 mg/kg. At the indicated time points before and after dosing, animals were bled and serum samples were analyzed using a sandwich ELISA assay using an HRP-conjugated antibody and 3,3',5,5'tetramethylbenzidine for reading at 450 nm. All samples were analyzed in duplicate and each result was the mean of the mice in each cohort (n=3). In FIG. 29 depicts the effect of in vivo gene silencing using GNA-modified siRNA duplexes on mouse serum TTR levels.

In FIG. 30 is a graph showing in vivo quantification of TTR mRNA levels. Animals received a single dose of siRNA of 2.5 mg/kg. At the indicated times after dosing, RNA extraction was performed from the liver homogenate. TTR mRNA was assessed as described above by RT-qPCR using the ΔΔCt method with GAPDH as control transcript and normalized to PBS-treated animals. In FIG. 30 depicts the effect of in vivo gene silencing using GNA-modified siRNA duplexes on mouse liver mRNA levels.

The results shown in the above figures demonstrate that the introduction of GNA resulted in a position-dependent violation of the thermal stability of the resulting duplex. The degree of impairment depended on the nucleotide; at the same time, the substitution of the A or U nucleotide resulted in a significantly lower ΔT _μ compared to the substitution of the G or C nucleotides with GNA. The introduction of individual GNA nucleotides into the primer site of siRNA duplexes resulted in similar levels of TTR mRNA knockdown in vitro. In addition, siRNA containing GNA base pairs within the seed region, as well as duplexes obtained by combining, showed higher levels of in vitro knockdown than the corresponding parental siRNA.

In vivo gene silencing correlated well with in vitro results for duplexes containing one GNA substitution. A double substitution with GNA resulted in a loss of in vivo silencing activity when compared to single substitution siRNAs.

All US patents, US patent application publications, foreign patents, foreign patent applications and non-patent publications mentioned in this specification are incorporated herein by reference in their entirety. Aspects of the embodiments may be modified if ideas from various patents, applications, and publications are to be used to provide additional embodiments.

These and other changes may be made to the embodiments in light of the detailed description above. In general, in the following claims, the terms used are not to be construed as limiting the claims to the specific embodiments disclosed in this specification and the claims, but are to be construed to include all possible embodiments together with the full scope of equivalents applicable to such formula points. Accordingly, the claims are not limited to this disclosure.

Claims (32)

1. A double-stranded RNA (dsRNA) agent that is capable of repressing the expression of a target gene, comprising a sense strand and an antisense strand, each strand containing 14 to 40 nucleotides, wherein the dsRNA agent is represented by formula (I) where each of B1, B1', B2', B3' and B4' is independently a nucleotide having a modification selected from the group consisting of 2'-OMe and 2'-fluoro; each of B2 and B3 is a nucleotide containing a 2'-OMe modification; C1 is a glycolic nucleic acid placed at the site opposite the seed site (positions 2-8) of the antisense strand; each of T1, T1', T2' and T3' is independently a 2'-fluoro-modified nucleotide; T1' is at position 14 from the 5' end of the antisense strand, aq ² is 1; T3' is at position 2 from the 5' end of the antisense strand, aq ⁶ and q ⁷ are each 1; length n ² is 3 nucleotides; the length of each of n ¹ , n ³ and q ¹ independently is from 4 to 15 nucleotides; the length of each of n ⁵ and q ³ is independently 1-6 nucleotides; the length of q ⁵ is independently 0-10 nucleotides; and the length of each of n ⁴ and q ⁴ is independently 0-3 nucleotides, where: (a) the sense strand contains at least one phosphorothioate bond, and (b) one of the T1 nucleotides is: or (i) in a position on the sense strand that is opposite position 11 from the 5' end of the antisense strand; or (ii) at position 11 of the sense strand, counting from the 5' end of the sense strand. 2. A dsRNA agent according to claim 1, wherein T2' is at positions 6-10 from the 5' end of the antisense strand, aq ⁴ is 1. 3. The dsRNA agent according to claim 1, wherein each of B1, B1', B2', B3' and B4' has 2'-OMe modifications. 4. The dsRNA agent according to claim 1, where n ⁴ is 0. 5. The dsRNA agent of claim 1, wherein the antisense strand contains two blocks of two phosphorothioate or methylphosphonate internucleotide bonds separated by 14, 15, 16, 17, or 18 phosphate internucleotide bonds. 6. The dsRNA agent of claim 5, wherein the antisense strand contains two blocks of two phosphorothioate or methylphosphonate internucleotide bonds separated by 16-18 phosphate internucleotide bonds. 7. A dsRNA agent according to claim 1, wherein the sense strand contains 19-22 nucleotides and the antisense strand contains 19-25 nucleotides. 8. The dsRNA agent according to claim 1, wherein the sense strand contains 21 nucleotides and the antisense strand contains 23 nucleotides. 9. The dsRNA agent of claim 1, wherein the dsRNA agent has a 3' and/or 5' overhang(s) that is 1-10 nucleotides in length. 10. The dsRNA agent of claim 1, wherein the nucleobase at the nucleotide at position 1 from the 5' end of the antisense strand in the duplex is selected from the group consisting of adenine, uracil, and thymine. 11. A dsRNA agent according to claim 1 conjugated to at least one ligand. 12. The dsRNA agent of claim 11, wherein the at least one ligand improves nuclease resistance of the dsRNA agent. 13. The dsRNA agent of claim 1 further comprising at least one asialoglycoprotein receptor (ASGPR) ligand. 14. The dsRNA agent of claim 13, wherein the ASGPR ligand is attached to the 5' or 3' end of the sense strand. 15. The dsRNA agent of claim 13, wherein the ASGPR ligand is attached to the 3' end of the sense strand. 16. The dsRNA agent of claim 13, wherein the ASGPR ligand is one or more GalNAc derivatives attached via a bivalent or trivalent branched linker. 17. The dsRNA agent according to claim 16, wherein the ASGPR ligand is 18. The dsRNA agent of claim 1, wherein the dsRNA has a blunt end at the 5' end of the antisense strand. 19. A dsRNA agent according to claim 1, wherein the antisense strand contains only 4- - 91 043057 four 2'-F modifications at positions 2, 6, 14 and 16 of the antisense strand from the 5' end of the antisense strand. 20. The dsRNA agent of claim 1, wherein the antisense strand contains only six 2'-F modifications at positions 2, 6, 8-9, 14, and 16 of the antisense strand from the 5' end of the antisense strand. 21. The dsRNA agent according to claim 1, wherein at least one of the first 1, 2, 3, 4 or 5 base pairs within the duplex regions, in the direction from the 5' end of the antisense strand, is selected from the group, consisting of A:U, G:U, T:C and mismatched pairs. 22. The dsRNA agent according to claim 1, wherein formula (I) further comprises 5'-vinylphosphonate (VP). 23. The dsRNA agent of claim 1, wherein formula (I) further comprises a 2'-deoxythymidine linked via a phosphorodithioate (PS ₂ ) bond at the 5' end of the antisense strand or sense strand. 24. A dsRNA agent according to claim 1, wherein B4' is 2'-OMe. 25. A dsRNA agent according to claim 1, having a duplex region length of 19-21 base pairs. 26. A dsRNA agent according to claim 1, having two blunt ends at both ends of the dsRNA duplex. 27. A pharmaceutical composition for suppressing the expression of a target gene, comprising a dsRNA agent according to any one of the preceding claims in combination with a pharmaceutically acceptable carrier or excipient. 28. A method for suppressing the expression of a target gene, comprising the step of administering a dsRNA agent according to any one of claims 1 to 26 in an amount sufficient to suppress the expression of the target gene. 29. The method of claim 28, wherein the dsRNA agent is administered via subcutaneous or intravenous administration. 30. A method for delivering a polynucleotide to a specific target in a subject by administering a dsRNA agent according to any one of claims 1-26. 31. The method of claim 30, wherein said step of administration is via routes of administration comprising intramuscular, intrabronchial, intrapleural, intraperitoneal, intraarterial, intralymphatic, intravenous, subcutaneous, cerebrospinal, or combinations thereof. 32. A method for delivering a polynucleotide to a specific target in a subject, the method comprising delivering the dsRNA agent according to any one of claims 1 to 26 by subcutaneous administration to the subject, whereby the polynucleotide is delivered to the specific target in the subject.