EA046694B1 - LIPID COMPOSITION BASED ON CATIONIC LIPID AND NUCLEIC ACID - Google Patents

Description

Claim of priority

The application for this invention claims priority to U.S.S.N. applications. 61/185800, filed June 10, 2009, and U.S.S.N. 61/244834, filed September 22, 2009, the contents of each of which are incorporated herein by reference in their entirety.

Field of technology

The invention relates to the field of delivery of therapeutic agents using lipid particles. In particular, the present invention is directed to the creation of cationic lipids and lipid particles containing these lipids, which are useful for the in vivo delivery of nucleic acids, as well as compositions of nucleic acids with lipid particles suitable for therapeutic use in vivo. In addition, the present invention is directed to methods for preparing these compositions, as well as methods for introducing nucleic acids into cells using these compositions, for example, for the treatment of various disease conditions.

State of the art

Therapeutic nucleic acids include, for example, small interfering RNA (SRNA), microRNA (smRNA), antisense oligonucleotides, ribozymes, plasmids and immunostimulatory nucleic acids. These nucleic acids act through different mechanisms. In the case of siRNA or miRNA, these nucleic acids can cause a decrease in the intracellular abundance of specific proteins through a process called RNA interference (RNA interference). Once siRNA or ttRNA is introduced into the cell cytoplasm, these double-stranded RNA constructs can bind to a protein called RISC. The sense strand of siRNA or sRNA is removed from the RISC complex, leaving RISC with a template that can recognize and bind to mRNA having a complementary sequence to the sequence bound by siRNA or sRNA. After binding to complementary mRNA, the RISC complex cleaves the mRNA and releases the cleaved strands. RNA can provide suppression of specific proteins by targeting specific destruction of the corresponding mRNAs, which serve as templates for protein synthesis.

The therapeutic use of RNA is extremely broad, since siRNA and ttRNA constructs with any nucleotide sequence directed against the target protein can be synthesized. To date, siRNA constructs have demonstrated the ability to specifically inhibit target proteins in both in vitro and in vivo models. In addition, siRNA constructs are currently being evaluated in clinical studies.

However, at present, two problems arise when using siRNA or sh1RNA constructs, firstly, their sensitivity to cleavage by nucleases in plasma and, secondly, their limited ability to gain access to the intracellular compartment, where they can bind to RISC, when systemic administration in the form of free siRNA or ttRNA. These double-stranded constructs can be stabilized by the introduction of chemically modified nucleotide linkers within the molecule, such as phosphothiolate groups. However, these chemical modifications provide only limited protection against nuclease degradation and may reduce the activity of the construct. Intracellular delivery of siRNA or A1RNA can be facilitated by the use of delivery systems such as polymers, cationic liposomes, or by chemical modification of the construct, for example, by covalently attaching cholesterol molecules. However, improved delivery systems are needed to increase the potency of siRNA and tRNA molecules and reduce or eliminate the need for chemical modification.

Antisense oligonucleotides and ribozymes can also inhibit translation of mRNA into protein. In the case of antisense constructs, these single-stranded deoxyribonucleic acids have a sequence complementary to the mRNA sequence of the target protein and can bind to the mRNA by Watson-Crick base pairing. This binding prevents translation of the target mRNA and/or triggers the degradation of mRNA transcripts by RNase H. As a result, antisense oligonucleotides have great potential for specific action (ie, inhibition of a specific disease-associated protein). These compounds have now shown potential in several in vitro and in vivo models, including models of inflammatory disease, cancer and HIV (for a review, see Agrawal, Trends in Biotech. 14:376-387, 1996). Antisense strands can also influence cellular activity by specifically hybridizing to chromosomal DNA. More rigorous human clinical evaluation of several antisense drugs is currently underway. Targets of these drugs include the bcl2 and apolipoprotein B genes and their mRNA products.

Immunostimulatory nucleic acids include deoxyribonucleic acids and ribonucleic acids. In the case of deoxyribonucleic acids, certain sequences or motifs have been shown to cause immunostimulation in mammals. These sequences or motifs include CpG motifs, pyrimidine-rich sequences and palindromic sequences. It is believed that the CpG motif in deoxyribonucleic acids is specifically recognized by an endosomal receptor, Toll-like receptor 9 (TLR-9), which then triggers both innate and adaptive immune stimulation pathways. Some immunosuppressants have also been reported

- 1 046694 stimulating ribonucleic acid sequences. These RNA sequences are thought to trigger immune activation by binding to Toll-like receptors 6 and 7 (TLR-6 and TLR-7). In addition, double-stranded RNAs have also been reported to be immunostimulatory and are believed to exert activation through binding to TLR-3. One well-known problem in the use of therapeutic nucleic acids relates to the stability of the phosphodiester internucleotide linkage and the sensitivity of this linker to nucleases. The presence of exonucleases and endonucleases in serum results in rapid degradation of nucleic acids containing phosphodiester linkers, and therefore therapeutic nucleic acids may have very short half-lives in the presence of serum or within cells. (Zelphati, O., et al., Antisense. Res. Dev. 3:323-338 (1993) and Thierry, A. R., et al., p. 147-161 in Gene Regulation: Biology of Antisense RNA and DNA (Eds Erickson, R.P. and Izant, J.G.; Raven Press, NY (1992) The basic phosphodiester chemical bond found in natural nucleic acids is not currently used in therapeutic nucleic acid development due to this and other known problems.

This problem has been partially addressed by chemical modifications that reduce degradation in the serum and within the cell. Modifications of the internucleotide phosphodiester bridge (eg, use of a phosphorothioate, methylphosphonate, or phosphoramidate compound), nucleotide bases (eg, 5-propynyl-pyrimidines), or sugars (eg, 2'-modified sugars) have been tested (Uhlmann E., et al., Antisense : Chemical Modifications. Encyclopedia of Cancer, Vol. X., p 64-81 Academic Press Inc. (1997). Others have attempted to improve stability by using 2'-5' sugar couplings (see, for example, US Pat. No. 5,532,130). Other changes were also made. However, none of these solutions have been found to be completely satisfactory, and free in vivo therapeutic nucleic acids still have very limited effectiveness.

In addition, as noted above for siRNA and miRNA, problems remain with the limited ability of therapeutic nucleic acids to cross cell membranes (see, Vlassov, et al., Biochim. Biophys. Acta, 1197:95-1082 (1994)) and problems associated with systemic toxicity, such as complement-mediated anaphylaxis, altered clotting properties and cytopenia (Galbraith, et al., Antisense Nucl. Acid Drug Des. 4:201-206, 1994).

To improve efficacy, researchers have also used lipid-based delivery systems to deliver chemically modified or unmodified therapeutic nucleic acids. In Zelphati, O. and Szoka, F.C., J. Contr. Rel. 41:99-119 (1996), the authors refer to the use of anionic (conventional) liposomes, pH-sensitive liposomes, immunoliposomes, fusogenic liposomes and aggregates of cationic lipids with antisense. Similarly, siRNAs have been systematically administered in cationic liposomes, and these nucleinolipid particles have been reported to provide improved inhibition of target proteins in mammals, including non-human primates (Zimmermann et al., Nature, 441:111-114 (2006)). Despite recent progress, there remains a need for improved compositions of lipids and therapeutic nucleic acids suitable for general therapeutic use. Preferably, these compositions should encapsulate nucleic acids with high efficiency, have a high drug:lipid ratio, protect the encapsulated nucleic acid from degradation and clearance in serum, be suitable for systemic delivery, and provide intracellular delivery of the encapsulated nucleic acid. In addition, these lipid nucleic acid particles must be well tolerated and provide an adequate therapeutic index such that treatment of a patient at an effective dose of the nucleic acid is not associated with significant toxicity and/or risk to the patient. The present invention is aimed at creating such compositions, methods for producing compositions and methods for using compositions for the introduction of nucleic acids into cells, including for the treatment of diseases.

The essence of the invention

The present invention is directed to the provision of improved lipid compositions containing a cationic lipid of formula I, where the formula I\O_„—or a pharmaceutically acceptable salt thereof and a therapeutic agent, wherein said therapeutic agent is a nucleic acid, wherein the nucleic acid is mRNA.

The present invention further includes, in other related embodiments, a method of delivering a therapeutic agent into a cell, comprising administering to a subject a lipid composition of the present invention.

- 2 046694

Brief description of the figures

In fig. 1 is a bar graph depicting the effect of lipid compositions containing DLin-M-C3-DMA on FVII silencing using the mouse as a model organism.

In fig. 2 is a bar graph showing the dose-dependent effect of MC3 in rats for various liposome formulations.

In fig. Figure 3 presents a bar graph that demonstrates the dependence on ApoE of the effectiveness of compositions containing MC3. Wild-type mice, but not ApoE knockout mice, showed a dose-dependent decrease in FVII protein levels. In fig. 3 also depicts a graph demonstrating that the ApoE dependence of MC3 liposome formulations and the lack of silencing in ApoE knockout mice treated with MC3 can be effectively normalized by premixing with ApoE.

In fig. 4 is a bar graph that shows the effects of variations in the molar percentage of MC3 in the liposome composition, as well as the effects of variations in the neutral lipid (eg, replacing the neutral lipid with DSPC, DMPC, and DLPC).

In fig. 5 is a bar graph demonstrating that increasing degrees of PEG shielding reduce non-GalNAc mediated silencing in mice.

In fig. 6 is a bar graph demonstrating that increasing degrees of PEG shielding reduce non-GalNAc mediated silencing in rats.

In fig. 7 is a bar graph showing the effectiveness of liposome formulations having different mole percentages of MC3, with and without GalNAc.

In fig. 8 is a bar graph showing that the activity of GalNAc-carrying liposomes is absent in asialoglycoprotein receptor (ASGPR) knockout mice.

In fig. Figure 9 shows a dose-response curve as a function of % residual FVII and dose (mg/kg) for the composition the preparation of which is described in Example 17.

In fig. 10 shows the pKa titration curve for the cationic lipid of formula I, obtained as described in example 18.

Detailed Description of the Invention

This document describes an improved lipid composition that can be used, for example, as a delivery agent, for example a nucleic acid-based agent, such as RNA-based constructs, into a cell or body of a subject. Also described herein are methods of administering improved lipid compositions containing an RNA-based construct to an animal, and in some embodiments, methods of assessing target gene expression. In some embodiments, the improved lipid composition includes a targeting lipid (eg, a targeting lipid described herein, such as a GalNAc or folate containing lipid).

Lipids.

The present invention is directed to the creation of improved lipid compositions containing a cationic lipid of formula I, a neutral lipid, a sterol and PEG or a PEG-modified lipid, wherein formula I

In one embodiment of the present invention, the lipid is a racemic mixture.

In one embodiment of the present invention, the lipid is enriched in one diastereomer, eg, the lipid contains at least 95%, at least 90%, at least 80%, or at least 70% excess of the diastereomer.

In one embodiment of the present invention, the lipid is chirally pure, for example, a single isomer.

In one embodiment of the present invention, the lipid is enriched in one isomer.

In one embodiment, the compositions of the invention are comprised of at least 75%, at least 80%, or at least 90%. In one embodiment of the present invention, the composition contains from about 25 to about 75% by mole of a cationic lipid of formula I, for example, from about 35 to about 65%, from about 45 to about 65%, about 60%, about 57.5 %, about 50% or about 40% in molar terms.

In one embodiment of the present invention, the composition contains from about 0.5 to about 15% by molar neutral lipid, for example, from about 3 to about 12%, from about 5 to about 10%, or about 15%, about 10%, or approximately 7.5% in molar terms.

In one embodiment of the present invention, the composition contains from about 5 to about 50% molar sterol (for example, from about 15 to about 45%, from example

- 3 046694 but 20 to about 40%, about 40%, about 38.5%, about 35% or about 31% molar). In one embodiment of the present invention, the sterol is cholesterol.

In one embodiment of the present invention, the composition contains from about 0.5 to about 20% by molar value of PEG or PEG-modified lipid (e.g., from about 0.5 to about 10%, from about 0.5 to about 5%, about 1.5%, about 0.5%, about 1.5%, about 3.5%, or about 5% molar).

In one embodiment of the present invention, the compositions of the invention include 25-75% cationic lipid of formula I, 0.5-15% neutral lipid, 5-50% sterol and 0.5-20% PEG or PEG-modified lipid on a molar basis .

In one embodiment of the present invention, the compositions of the invention include 35-65% cationic lipid of formula I, 3-12% neutral lipid, 15-45% sterol and 0.5-10% PEG or PEG-modified lipid on a molar basis.

In one embodiment of the present invention, the compositions of the invention comprise 45-65% cationic lipid of formula I, 5-10% neutral lipid, 25-40% sterol and 0.5-10% PEG or PEG-modified lipid on a molar basis.

In one embodiment of the present invention, the compositions of the invention comprise about 60% cationic lipid of Formula I, about 7.5% neutral lipid, about 31% sterol, and about 1.5% PEG or PEG-modified lipid on a molar basis. In one preferred embodiment, the cationic lipid is a compound of Formula I, the neutral lipid is DSPC, the sterol is cholesterol, and the PEG-lipid is PEG-DMG (also referred to herein as PEG-C14 or C14-PEG). In one embodiment of the present invention, the PEG or PEG-modified lipid contains a PEG molecule with an average molecular weight of 2000 Da. In another embodiment, the PEG or PEG-modified lipid comprises a PEG molecule with an average molecular weight of less than 2000, such as about 1500 Da, about 1000 Da, or about 500 Da.

In one embodiment of the present invention, the PEG or PEG-modified lipid is a compound of the following formula VI:

with a PEG molecule with an average molecular weight of 2000 Da.

In one embodiment of the present invention, the PEG or PEG-modified lipid is PEG-distearoyl-glycerol (PEG-DSG, also referred to herein as PEG-C ₁₈ or C ₁₈ -PEG).

In one embodiment of the present invention, the compositions of the invention comprise about 50% cationic lipid of Formula I, about 10% neutral lipid, about 38.5% sterol, and about 1.5% PEG or PEG-modified lipid on a molar basis. In one preferred embodiment, the cationic lipid is a compound of Formula I, the neutral lipid is DSPC, the sterol is cholesterol, and the PEG lipid is PEG-DMG (also referred to herein as PEG-C14 or C14-PEG). In one embodiment of the present invention, the PEG or PEG-modified lipid is PEG-distyryl-glycerol (PEG-DSG, also referred to herein as PEG-C ₁₈ or C ₁₈ -PEG). In one embodiment of the present invention, the PEG or PEG-modified lipid is PEG-DPG (PEG-dipalmitoylglycerol). In one embodiment of the present invention, the PEG or PEG-modified lipid contains a PEG molecule with an average molecular weight of 2000 Da.

In one embodiment of the present invention, the compositions of the invention include about 50% cationic lipid of formula I, about 10% neutral lipid, about 35% sterol, about 4.5% PEG or PEG-modified lipid, and about 0.5% targeting lipid in molar ratio. In one preferred embodiment, the cationic lipid is a compound of formula I, the neutral lipid is DSPC, the sterol is cholesterol, the PEG lipid is PEG distearoyl glycerol (PEG-DSG, also referred to herein as PEG-C ₁₈ or C ₁₈ -PEG), and the targeting lipid is GalNAc3-PEG-DSG.

In one embodiment of the present invention, the compositions of the invention include about 50% cationic lipid of formula I, about 10% neutral lipid, about 35% sterol, about 4.5% PEG or PEG-modified lipid, and about 0.5% targeting lipid in molar ratio. In one preferred embodiment, the cationic lipid is a compound of formula I, the neutral lipid is DSPC, the sterol is cholesterol, the PEG lipid is PEG-DMG (also referred to herein as PEG-C14 or C ₁₄ -PEG).

In one embodiment of the present invention, the compositions of the invention comprise about 40% cationic lipid of Formula I, about 15% neutral lipid, about 40% sterol, and about 5% PEG or PEG-modified lipid on a molar basis. In one preferred embodiment, the cationic lipid is a compound of formula I, neutral lipid

- 4 046694 is DSPC, the sterol is cholesterol, the PEG lipid is PEG-DMG (also referred to herein as PEG-C ₁₄ or C ₁₄ -PEG).

In one embodiment of the present invention, the compositions of the invention comprise about 50% cationic lipid of Formula I, about 10% neutral lipid, about 35% sterol, and about 5% PEG or PEG-modified lipid on a molar basis. In one preferred embodiment, the cationic lipid is a compound of formula I, the neutral lipid is DSPC, the sterol is cholesterol, the PEG lipid is PEG-DMG (also referred to herein as PEG-C ₁₄ or C ₁₄ -PEG).

In one embodiment of the present invention, the compositions of the invention comprise about 57.2% cationic lipid of formula I, about 7.1% neutral lipid, about 34.3% sterol, and about 1.4% PEG or PEG-modified lipid on a molar basis . In one preferred embodiment, the cationic lipid is a compound of formula I, the neutral lipid is DPPC, the sterol is cholesterol, the PEG lipid is PEG-cDMA (PEG cDMA is further discussed in Heyes et al. (J. Controlled Release, 107, 276-287 (2005) )).

GalNAc3-n3r^Gr. In one embodiment of the present invention, the PEG or PEG-modified lipid is a compound of formula VI or PEG-DSG, wherein the PEG molecule has an average molecular weight of 2000 Da.

In one embodiment of the present invention, the compositions of the invention comprise about 57.5% cationic lipid of Formula I, about 7.5% neutral lipid, about 31.5% sterol, and about 3.5% PEG or PEG-modified lipid on a molar basis . In one preferred embodiment, the cationic lipid is a compound of formula I, the neutral lipid is DSPC, the sterol is cholesterol and the PEG lipid is PEG-DMG.

In one embodiment of the present invention, the lipid^RNA ratio is at least about 0.5:1, at least about 1:1, at least about 2:1, at least about 3:1, at least about 4:1, at least about 5:1, at least about 6:1, at least about 7:1, at least about 8:1, at least about 10:1, at least about 11 :1, at least about 12:1 to at least about 15:1. In one embodiment of the present invention, the lipidPRNA ratio is from about 1:1 to about 20:1, from about 3:1 to about 15:1, from about 4:1 to about 15:1, from about 5:1 to approximately 13:1. In one embodiment of the present invention, the lipidNRNA ratio is from about 0.5:1 to about 15:1.

One feature of the invention is that the improved lipid composition also includes a targeting lipid. In some embodiments, the targeting lipid comprises a GalNAc residue (ie, an N-galactosamine residue). For example, a targeting lipid containing a GalNAc residue may include that described in USSN 12/328669, filed April 12, 2008, which is incorporated by reference in its entirety. The targeting lipid may also include any other lipid (eg, targeting lipid) known to those skilled in the art, for example, as described in USSN 12/328669 or International Publication No. WO 2008/042973, the contents of each of which are incorporated by reference in their entirety. . In some embodiments, the targeting lipid includes multiple GalNAc residues, such as two or three GalNAc residues. In some embodiments, the targeting lipid includes multiple, for example two or three, N-acetylgalactosamine (GalNAc) residues. In some embodiments, the lipid in the targeting lipid is 1,2-di-O-hexadecyl-sn-glyceride (ie, DSG). In some embodiments, the targeting lipid includes a PEG moiety (e.g., a PEG moiety having a molecular weight of at least about 500 Da, such as about 1000, 1500, 2000 Da, or more), e.g., the targeting moiety is attached to the lipid through a PEG moiety .

In some embodiments, the targeting lipid comprises a folate moiety. For example, a targeting lipid containing a folate moiety may include that described in USSN 12/328,669, filed April 12, 2008, which is incorporated herein by reference in its entirety. In another embodiment of the present invention, the targeting lipid containing a folate moiety may comprise a compound of formula V.

Examples of targeted lipids are represented by Formula L below:

(Directed action group)-L-lipid.

formula L wherein the Targeting Group is any targeting group that is known to those skilled in the art and/or described herein (eg, cell surface receptor);

n is an integer from 1 to 5 (for example, 3);

L is a linker group;

the lipid is a lipid such as described herein (e.g., a neutral lipid, such

- 5 046694 kim as DSG).

In some embodiments of the present invention, the linker group contains a PEG moiety.

In some embodiments, the targeting lipid is Compound 2, 3, 4, or 5 as follows:

GalNAc3-flCr

Ca^Ac3-PEG-DSG

Formula III (GalNAc) ₃ -II3r-LCO

Formula IV

Folate-PEG-DSPE

1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-[folate(polyethylene glycol)-2000](ammonium salt) Formula V

- 6 046694

Folate-PEG3400-DSG

MW-4761

Formula VII

In some embodiments, the targeting lipid is present in the composition in an amount of from about 0.001 to about 5% (e.g., about 0.005, 0.15, 0.3, 0.5, 1.5, 2, 2.5, 3, 4 or 5%) on a molar basis. In some embodiments of the present invention, the targeting lipid is present in the composition in an amount of from about 0.005 to about 1.5%. In some embodiments of the present invention, the targeting lipid is included in the composition described herein.

In some embodiments, the lipid composition also contains an antioxidant (eg, a radical scavenger). The antioxidant may be present in the composition, for example, in an amount of from about 0.01 to about 5%. The antioxidant may be hydrophobic or hydrophilic (eg, lipid-soluble or water-soluble). In some embodiments, the antioxidant is a phenolic compound, such as butylated hydroxytoluene, resveratrol, coenzyme Q10 or other flavinoid, or a vitamin, such as vitamin E or vitamin C. Other examples of antioxidants include lipoic acid, uric acid, a carotene such as β-carotene or retinol (vitamin A), glutathione, melatonin, selenium and ubiquinol.

In some embodiments, the targeting lipid receptor (eg, GalNAc-containing lipid) is an asialoglycoprotein receptor (ie, ASGPR).

In one embodiment of the present invention, the compositions according to the invention are prepared using an extrusion method or a flow mixing method.

An extrusion method (also called a preformed method or a batch method) is a method in which empty liposomes are first prepared (ie, without nucleic acid), and then nucleic acid is added to the empty liposomes. Extrusion of liposome formulations through a fine porous polycarbonate membrane or asymmetric ceramic membrane results in a relatively well-defined size distribution. Typically, the suspension is circulated through the membrane one or more times until the desired size distribution of liposome complexes is achieved. Liposomes can be extruded through successive membranes of decreasing pore size to achieve gradual reduction in liposome size. In some cases, the resulting lipid-nucleic acid compositions can be used without separation by size. These methods are disclosed in US patent No. 5008050; 4927637; 4737323; Biochim Biophys Acta. 1979 Oct 19; 557(1):9-23; Biochim. Biophys. Acta. 1980 Oct 2; 601(3):559-7; Biochim. Biophys. Acta. 1986 Jun 13; 858(1):161-8 and Biochim. Biophys. Acta. 1985, 812, 55-65, the contents of which are incorporated by reference in their entirety.

The in-line mixing method is a method in which both lipids and nucleic acid are added to the mixing chamber in parallel. The mixing chamber may be a simple tee connector or any other mixing chamber known to those skilled in the art. These methods are disclosed in US patent No. 6534018 and 6855277; US Publication 2007/0042031 and Pharmaceuticals Research, Vol. 22, No. 3, Mar. 2005, p. 362-372, which are incorporated by reference in their entirety.

It is understood that the compositions of the invention can be prepared by any other methods known to those skilled in the art.

In a further embodiment, representative compositions containing the compound

- 7 046694 formula I, are described in table. 1.

Table 1

MSZ DSFH Cholesterol PEG 60 7.5 31 1.5 50 10 38.5 1.5 40 20 38.5 1.5 50 10 38.5 1.5 50 10 38.5 1.5 40 20 38.5 1.5 60 7.5 21 1.5 50 10 38.5 1.5 50 10 38.5 1.5 20 40 (DMPH) 38.5 1.5 thirty thirty 38.5 1.5 10 50 (DMPH) 38.5 1.5 thirty thirty (DMPH) 38.5 1.5 10 51 (DLPH) 38.5 1.5 20 40 (DLPH) 38.5 1.5 40 20 38.5 1.5 40 10 40 10 60 10 20 10 40 20 37 3 60 10 27 3

In one embodiment, specific compositions containing a compound of Formula I are described as follows:

Lipid ratio (in mole percent).

Lipid^RNA ratio.

50/10/38.5/1.5 (MC3: DSPC:cholesterol: PEG-DMG).

Lipid^RNA ratio ~11;

40/15/40/5 (MC3:DSPC:cholesterol:PEG-DMG).

Lipid^RNA ratio ~11;

0/10/35/4.5/0.5% (MC3 :DSPC:cholesterol:PEG-DSG (C 18-PEG): GalNAc3 -PEG-DSG).

Lipid^RNA ratio ~11;

0/10/30/9.5/0.5% (MC3 :DSPC:cholesterol:PEG-DSG:GalNAc3 -PEG-DSG).

Lipid^RNA ratio ~11;

0/10/35/5% (MC3: DSPC:cholesterol: PEG-DSG).

Lipid^RNA ratio ~11;

50/10/38.5/1.5 (MC3 :DPPC:cholesterol:PEG-DMG).

Lipid^RNA ratio ~11.

40/15/40/5 (MC3:DPPC:cholesterol:PEG-DMG).

Lipid^RNA ratio ~11;

0/10/35/4.5/0.5% (MC3 :DPPC:cholesterol:PEG-DSG:GalNAc3 -PEG-DSG).

Lipid^RNA ratio ~11;

50/10/30/9.5/0.5% (MC3:DPPC:cholesterol:PEG-DSG:GalNAc3-PEG-DSG).

Lipid^RNA ratio ~11;

50/10/35/5% (MC3:DPPC:cholesterol:PEG-DSG)

Lipid^RNA ratio ~11;

50/10/38.5/1.5 (MC3: DSPC:cholesterol: PEG-DMG).

Lipid^RNA ratio ~7;

- 8 046694

50/10/38.5/1.5 (MS3:DSPC:cholesterol:PEG-DSG).

Lipid^RNA ratio ~10;

50/10/38.5/1.5 (MS3:DSPC:cholesterol:PEG-DMG).

Lipid to gRNA ratio ~12;

50/10/35/5% (MC3 :DSPC:cholesterol:PEG-DMG).

Lipid:siRNA ratio ~8№

50/10/35/5% (MC3 :DSPC:cholesterol:PEG-DMG).

Lipid:siRNA ratio ~10.

In one embodiment, the compositions of the invention are at least 75%, at least 80%, or at least 90% captured.

In one embodiment of the present invention, the compositions of the invention further comprise an apolipoprotein. As used herein, the term apolipoprotein or lipoprotein means apolipoproteins known to one skilled in the art, and variants and fragments thereof, and agonists, apolipoprotein analogs, or fragments thereof, described below.

Suitable apolipoproteins include, but are not limited to, ApoA-I, ApoA-II, ApoA-IV, ApoA-V, and ApoE and their active polymorphic forms, isoforms, variant and mutant forms, and fragments or truncated forms. In some embodiments, the apolipoprotein is a thiol-containing apolipoprotein. Thiol-containing apolipoprotein means an apolipoprotein, variant, fragment or isoform containing at least one cysteine residue. The most common thiol-containing apolipoproteins are ApoA-I Milano (ApoA-I _M ) and ApoA-I Paris (ApoA-I _p ), which contain a single cysteine residue (Jia et al., 2002, Biochem. Biophys. Res. Comm. 297: 206-13; Bielicki and Oda, 2002, Biochemistry, 41:2089-96). ApoA-II, ApoE2 and ApoE3 are also thiol-containing apolipoproteins. Isolated ApoE and/or its active fragments and polypeptide analogues, including its recombinantly produced forms, are described in US patent No. 5672685; 5525472; 5473039; 5182364; 5177189; 5168045; 5116739; the disclosures of which are incorporated by reference in the description. ApoE3 is described in Weisgraber, et al., Human E apoprotein heterogeneity: cysteine-arginine interchanges in the amino acid sequence of the apo-E isoforms, J. Biol. Chem. (1981), 256:9077-9083 and Rail, et al., Structural basis for receptor binding heterogeneity of apolipoprotein E from type III hyperlipoproteinemic subjects, Proc. Nat. Acad. Sci. (1982), 79:4696-4700. See also GenBank accession number K00396.

In some embodiments, the apolipoprotein may be in its mature form, in the form of a preproapolipoprotein, or in the form of a proapolipoprotein. Homo- and heterodimers (where applicable) of pro- and mature ApoA-I (Duverger et al., 1996, Arterioscler. Thromb. Vase. Biol. 16(12):1424-29), ApoA-I Milano (Klon et al. , 2000, Biophys. J. 79:(3), 1679-87; Franceschini et al., 1985, J. Biol. Chem. 260:1632-35), ApoA-I Paris (Daum et al., 1999, J Mol. Med. 77:614-22), ApoA-II (Shelness et al., 1985, J. Biol. Chem. 260(14):863746; Shelness et al., 1984, J. Biol. Chem. 259 (15):9929-35), ApoA-IV (Duverger et al., 1991, Euro. J. Biochem. 201(2):373-83) and ApoE (McLean et al., 1983, J. Biol. Chem. 258(14):8993-9000) may also be used within the scope of the invention.

In some embodiments, the apolipoprotein may be a fragment, variant, or isoform of an apolipoprotein. The term fragment means any apolipoprotein having an amino acid sequence shorter than that of the native apolipoprotein, the fragment retaining the activity of the native apolipoprotein, including lipid-binding properties. By variant we mean substitutions or changes in the amino acid sequences of an apolipoprotein, and the substitutions or changes, for example the addition or deletion of amino acid residues, do not destroy the activity of the native apolipoprotein, including lipid-binding properties. Thus, a variant may comprise a protein or peptide having an amino acid sequence substantially identical to the native apolipoprotein described herein, in which one or more amino acid residues have been conservatively replaced by amino acids with similar chemical properties. Examples of conservative substitutions include replacing at least one hydrophobic residue, such as isoleucine, valine, leucine or methionine, with another. Similarly, the present invention contemplates, for example, the exchange of at least one hydrophilic residue, such as, for example, exchange between arginine and lysine, between glutamine and asparagine, and between glycine and serine (see US patent No. 6004925, 6037323 and 6046166).

The term isoform means a protein having the same, broader or partial function and similar, identical or partial sequence, which may or may not be the product of the same gene and is usually tissue specific (see Weisgraber 1990, J. Lipid Res. 31( 8):1503-11; Hixson and Powers 1991, J. Lipid Res. 32(9):1529-35; J. Biol. 260(2):703-6; al., 1986, J. Biol. Chem. 261(9):3911-4; Gordon et al., 1984, J. Biol. Chem. 259(1):46874; 6):831-40; Aviram et al., 1998, Arterioscler. Vase. 18(10):161724; ; Billecke et al., 2000, Drug Metab. 28(11):133542; J. Biol. Chem. 2000, 2000; Chem. 260(4):2258-64; J. Biol. Chem. 255(21):10464-71, 1995, J. Lipid Res. :80- 9 046694

8; Sacre et al., 2003, FEBS Lett. 540(1-3):181-7; Weers, et al., 2003, Biophys. Chem. 100(1-3):481-92; Gong et al., 2002, J. Biol. Chem. 277(33):29919-26; Ohta et al., 1984, J. Biol. Chem. 259(23):14888-93 and US Patent No. 6372886).

In some embodiments, the methods and compositions of the present invention include the use of chimeric apolipoprotein constructs. For example, a chimeric apolipoprotein construct can be composed of an apolipoprotein domain with high lipid-binding activity associated with an apolipoprotein domain having ischemia reperfusion protective properties. A chimeric apolipoprotein construct may be a construct that includes distinct regions within an apolipoprotein (ie, a homologous construct), or a chimeric construct may be a construct that includes distinct regions of different apolipoproteins (ie, heterologous constructs). Compositions containing a chimeric construct may also include segments that are variants of apolipoproteins, or segments designed to have a specific property (eg, lipid binding, receptor binding, enzymatic properties, enzyme activating properties, antioxidant or redox properties) (see . Weisgraber 1990, J. Lipid Res. 31(8):1503-11; Hixson and Powers 1991, J. Lipid Res. 32(9):1529-35; 260(2):703-6; Hoeg et al., 1986, J. Biol. Chem. 261(9):3911-4; Gordon et al., 1984, J. Biol. -74; Powell et al., 1987, Cell, 50(6):831-40; Aviram et al., 1998, Arterioscler Vasc. 18(10):1617-24; 1998, J. Clin. Invest. 101(8):1581-90; Drug Metab. 28(11):1335-42; 275(43):33435-42;Steinmetz and Utermann 1985, J. Biol. Chem. 260(4):2258-64; 71; Dyer et al., 1995, J. Lipid Res. 36(1):80-8; Sorenson et al., 1999, Arterioscler. Thromb. Vasc. Biol. 19(9):2214-25; Palgunachari 1996, Arterioscler. Throb. Vasc. Biol. 16(2):328-38: Thurberg et al., J. Biol. Chem. 271(11):6062-70; Dyer 1991, J Biol. Chem. 266(23):150009-15; Hill 1998, J Biol. Chem. 273(47):30979-84).

Apolipoproteins useful in the invention also include recombinant, synthetic, semi-synthetic or purified apolipoproteins. Methods for producing apolipoproteins or equivalents used in the invention are well known to those skilled in the art. For example, apolipoproteins can be isolated from plasma or natural products using, for example, density gradient centrifugation or immunoaffinity chromatography, or produced synthetically, semi-synthetically, or using recombinant DNA techniques known to those skilled in the art (see, for example, Mulugeta et al. , 1998, J. Chromatogr. 798(1-2):83-90; Chung et al., 1980, J. Lipid Res. 21(3):284-91; 28(8):913-29; Persson, et al., 1998, J. Chromatogr. 711:97-109; US patents No. 5059528, 5834596, 5876968 and 5721114 and PCT publications WO 86/04920 and WO 87/02062 ).

Apolipoproteins useful in the invention further include apolipoprotein agonists, such as peptides and peptide analogs, which mimic the activity of ApoA-I, ApoA-I Milano (ApoA-IM), ApoA-I Paris (ApoA-IP), ApoA-II, ApoA -IV and ApoE. For example, the apolipoprotein may be any of those described in US Pat. Nos. 6,004,925, 6,037,323, 6,046,166, and 5,840,688, the contents of which are incorporated by reference in their entirety.

Apolipoprotein agonist peptides or peptide analogues can be synthesized or produced using any peptide synthesis technique known to those skilled in the art, including, for example, those described in US Pat. Nos. 6,004,925, 6,037,323, and 6,046,166. solid phase synthesis technique originally described by Merrifield (1963, J. Am. Chem. Soc. 85:2149-2154). Other peptide synthesis techniques can be found in Bodanszky et al., Peptide Synthesis, John Wiley & Sons, ^2nd Ed., (1976) and other literature readily available to one skilled in the art. A summary of polypeptide synthesis techniques can be found in Stuart and Young, Solid Phase Peptide. Synthesis, Pierce Chemical Company, Rockford, 111. (1984). Peptides can also be synthesized by methods using solution synthesis, as described in The Proteins, Vol. II, ^3rd Ed., Neurath et al., Eds., p. 105-237, Academic Press, New York, NY (1976). Protecting groups suitable for use in various peptide synthesis processes are described in the above texts and also in McOmie, Protective Groups in Organic Chemistry, Plenum Press, New York, NY (1973). The peptides of the present invention can also be obtained by chemical or enzymatic cleavage from a longer form, for example, apolipoprotein AI.

In some embodiments, the apolipoprotein may be a mixture of apolipoproteins. In one embodiment of the present invention, the apolipoprotein may be a homogeneous mixture, i.e. one type of apolipoprotein. In another embodiment, the apolipoprotein may be a heterogeneous mixture of apolipoproteins, i.e. a mixture of two or more different apolipoproteins. Embodiments of heterogeneous mixtures of apolipoproteins may include, for example, a mixture of an apolipoprotein from an animal source and an apolipoprotein of semi-synthetic origin. In some embodiments, the heterogeneous mixture may include, for example, a mixture of ApoA-I and ApoA-I Milano. In some embodiments, the heterogeneous mixture may include, for example, a mixture of ApoA-I Milano and ApoA-I Paris. Mixtures, at

- 10 046694 suitable for use in the methods and compositions according to the invention will be obvious to a person skilled in the art.

If apolipoprotein is derived from natural sources, it may be derived from a plant or animal source. If the apolipoprotein is derived from an animal source, the apolipoprotein can come from any species. In some embodiments, the apolipoprotein can be obtained from an animal source. In some embodiments, the apolipoprotein can be obtained from a human source. In preferred embodiments of the invention, the apolipoprotein is from the same species as the individual to whom the apolipoprotein is administered.

In one embodiment of the present invention, the target gene is selected from the group consisting of a factor VII gene, Eg5, PCSK9, TPX2, apoB, SAA, TTR, RSV, PDGFe gene, Erb-B gene, Src gene, CRK gene, GRB2 gene, RAS, MEKK gene, JNK gene, RAF gene, Erk1/2 gene, PCNA(p21) gene, MYB gene, JUN gene, FOS gene, BCL-2 gene, cyclin D gene, VEGF gene, EGFR gene, cyclin A gene, cyclin E gene, WNT-1 gene, β-catenin gene, c-MET gene, PKC gene, NFKB gene, STAT3 gene, survivin gene, Her2/Neu gene, topoisomerase I gene, topoisomerase II a gene, p73 gene, p21 gene (WAF1/CIP1), p27(KIP1) gene, PPM1D gene, RAS gene, caveolin I gene, MIB I gene, MTAI gene, M68 gene, mutations in tumor suppressor genes, p53 tumor suppressor gene and combinations thereof. In one embodiment of the present invention, the target gene is a gene expressed in the liver, such as the factor VII (FVII) gene. The effect on the expression of a target gene, such as the FVII gene, is assessed by measuring the level of FVII in a biological sample, such as serum or tissue sample. For example, the level of FVII can be determined in the blood, for example, as measured by quantifying FVII activity. In one embodiment of the present invention, the level of mRNA in the liver can be assessed. In another preferred embodiment, at least two types of assessment are performed, eg, both a protein level (eg, in the blood) and an mRNA level measurement (eg, in the liver) are performed simultaneously.

In one embodiment of the present invention, the agent is a nucleic acid, such as double-stranded RNA (dsRNA).

In another embodiment of the present invention, the nucleic acid agent is single-stranded DNA or RNA, or double-stranded DNA or RNA, or a DNA-RNA hybrid. For example, double-stranded DNA may be a structural gene, a gene containing regulatory and terminator regions, or a self-replicating system such as viral or plasmid DNA. The double-stranded RNA may be, for example, dsRNA or other RNA interference reagent. The single-stranded nucleic acid may be, for example, an antisense oligonucleotide, a ribozyme, a microRNA, or a triplex-forming oligonucleotide.

In another embodiment of the present invention, at various time points after administration of the candidate agent, a biological sample, such as a fluid sample, such as blood, plasma or serum, or a tissue sample, such as a liver tissue sample, is collected from the test subject and tested for the effect of the agent on the level of expression of the target protein or mRNA. In one particularly preferred embodiment, the candidate agent is a dsRNA that is directed against FVII, and the biological assay is tested for an effect on FVII protein or mRNA levels. In one embodiment of the present invention, plasma levels of FVII protein are quantified, such as using an immunohistochemical assay or a chromogenic assay. In another embodiment of the present invention, FVII mRNA levels in the liver are tested using a quantitative assay such as a branched probe or Northern blot or RT-PCR method.

In one embodiment of the present invention, an agent, for example a composition comprising an improved lipid composition, is assessed for toxicity. In another embodiment of the present invention, a model subject may be monitored for physical effects, such as changes in mass or cellular behavior.

In one embodiment of the present invention, the method further includes exposure to an agent, for example, a composition containing an improved lipid composition, for further evaluation. Further evaluation may include, for example, (i) repeating the evaluation described above, (ii) repeating the evaluation described above with different numbers of animals or with different doses, or (iii) evaluation in a different way, for example, evaluation in a different animal model e.g. non-human primate.

In another embodiment of the present invention, a decision is made as to whether or not to include the agent and the improved lipid composition in further studies, such as clinical trials, depending on the observed effect of the candidate agent on protein or mRNA levels in the liver. For example, if a dsRNA candidate is observed to reduce protein or mRNA levels by at least 20, 30, 40, 50% or more, then the agent may be considered for clinical trials.

In another embodiment of the present invention, a decision is made as to whether or not to include the agent and the improved lipid composition in the pharmaceutical composition

- 11 046694 tion, depending on the observed effect of the candidate agent and aminolipid on protein or mRNA levels in the liver. For example, if a dsRNA candidate is observed to reduce protein or mRNA levels by at least 20, 30, 40, 50, or more, then the agent may be considered for clinical trials.

Another feature is that the invention provides a method for evaluating an improved lipid composition for its suitability for delivering a therapeutic agent into a cell. In some embodiments, the invention provides a method of evaluating an improved lipid composition for its suitability for delivering an RNA-based construct, such as a FVII-targeting dsRNA. The method includes creating a composition that contains a FVII-targeting dsRNA and a candidate aminolipid, administering the composition to a rodent, for example, a mouse, assessing FVII expression as a function of at least one of a blood FVII level or a liver FVII mRNA level, which provides an assessment of a candidate aminolipid. In some embodiments of the present invention, the method further includes comparing the expression of a target gene with a preselected reference value.

Compositions that include lipid-containing components, such as liposomes, are described in more detail below. Examples of nucleic acid-based agents include double-stranded RNA, antisense oligonucleotides, ribozymes, microRNAs, immunostimulatory oligonucleotides, or triplex-forming oligonucleotides. These agents are also described in more detail below.

Alkyl means a straight-chain or branched, non-cyclic or cyclic, saturated aliphatic hydrocarbon containing from 1 to 24 carbon atoms. Examples of saturated, straight chain alkyls include methyl, ethyl, n-propyl, n-butyl, n-pentyl, n-hexyl and the like; and saturated branched alkyls include isopropyl, sec-butyl, isobutyl, tert-butyl, isopentyl and the like. Examples of saturated cyclic alkyls include cyclopropyl, cyclobutyl, cyclopentyl, cyclohexyl and the like; and unsaturated cyclic alkyls include cyclopentenyl and cyclohexenyl and the like.

Alkenyl means alkyl, as defined above, containing at least one double bond between adjacent carbon atoms. Alkenyls include both cis and trans isomers. Examples of straight chain and branched alkenyls include ethylenyl, propylenyl, 1-butenyl, 2-butenyl, isobutylenyl, 1-pentenyl, 2-pentenyl, 3-methyl-1-butenyl, 2-methyl-2-butenyl, 2,3- dimethyl-2butenyl, etc.

Alkynyl means any alkyl or alkenyl, as defined above, which additionally contains at least one triple bond between adjacent carbon atoms. Examples of straight-chain and branched alkynyls include acetylenyl, propynyl, 1-butynyl, 2-butynyl, 1-pentynyl, 2-pentynyl, 3-methyl-1-butynyl and the like.

Acyl means any alkyl, alkenyl or alkynyl wherein the carbon atom at the point of attachment is replaced by an oxo group, as defined below. For example, -C(=O)alkyl, -C(=O)alkenyl and -C(=O)alkynyl are acyl groups.

The term aryl means an aromatic monocyclic, bicyclic or tricyclic hydrocarbon ring system where any ring atom may be substituted. Examples of aryl moieties include, but are not limited to, phenyl, naphthyl, anthracenyl, and pyrenyl.

Heterocycle means a monocyclic 5 to 7 membered or bicyclic 7 to 10 membered heterocyclic ring which may be saturated, unsaturated or aromatic and which contains from 1 to 2 heteroatoms independently selected from nitrogen, oxygen and sulfur, and where the nitrogen and sulfur heteroatoms may optionally be oxidized and the nitrogen heteroatom may optionally be quaternary, including bicyclic rings in which any of the above heterocycles are conjugated to a benzene ring. The heterocycle can be attached via any heteroatom or carbon atom. Heterocycles include heteroaryls as defined below. Heterocycles include morpholinyl, pyrrolidinonyl, pyrrolidinyl, piperidinyl, piperisinyl, hydantoinyl, valerolactamyl, oxiranyl, oxetanyl, tetrahydrofuranyl, tetrahydropyranyl, tetrahydropyridinyl, tetrahydroprimidinyl, tetrahydrothiophenyl, tetrahydrothiopyranyl, tetrahydrothiophenyl, tetrahydrothiopyranyl, etc.

The terms optionally substituted alkyl, optionally substituted alkenyl, optionally substituted alkynyl, optionally substituted acyl and optionally substituted heterocycle mean that when there is a substitution, at least one hydrogen atom is replaced by a substituent. In the case of an oxo substituent (=O), two hydrogen atoms are replaced. In this sense, substituents include oxo group, halogen, heterocycle, -CN, -OR ^x , -NR ^x R ^y , -NR ^x C(=O)R ^y , -NR ^x SO2R ^y , -C(=O)R ^x , -C(=O)OR ^x , -C(=O)NR ^x R ^y , -SONR ^x and -SOnNR ^x R ^y , where n is 0, 1 or 2, R ^x and R ^y are the same or different and, independently, are hydrogen, alkyl, or heterocycle, and each of said alkyl or heterocyclic substituents may be further substituted by one or more of oxo, halogen, -OH, -CN, alkyl, -OR ^x , heterocycle, -NR ^x R ^y , -NR ^x C(=O)R ^y , -NR ^x SO2R ^y , -C(=O)R ^x , -C(=O)OR ^x , -C(=O)NR ^x R ^y , -SOnRx and -SOnNRxRy.

The term heteroaryl means an aromatic monocyclic 5-8 membered, bicyclic 8-12 membered or tricyclic 11-14 membered ring system having 1-3 heteroatoms, if

- 12 046694 is monocyclic, 1-6 heteroatoms if it is bicyclic, or 1-9 heteroatoms if it is tricyclic, wherein said heteroatoms are selected from O, N or S (for example, carbon atoms and 1-3, 1-6 or 1-9 heteroatoms N, O or S, if monocyclic, bicyclic or tricyclic, respectively), and any ring atom may be substituted. Heteroaryl groups described herein may also contain conjugated rings sharing a common carbon-carbon bond.

The term alkylheterocycle means heteroaryl where at least one of the ring atoms is substituted with alkyl, alkenyl or alkynyl.

The term substituted means the replacement of one or more hydrogen radicals in a given structure with a radical of a specific substituent, including, but not limited to, halogen, alkyl, alkenyl, alkynyl, aryl, heterocyclyl, thiol, alkylthio, oxo, thioxy, arylthio, alkylthioalkyl, arylthioalkyl, alkylsulfonyl, alkylsulfonylalkyl , arylsulfonylalkyl, alkoxy, aryloxy, aralkoxy, aminocarbonyl, alkylaminocarbonyl, arylaminocarbonyl, alkoxycarbonyl, aryloxycarbonyl, haloalkyl, amino group, trifluoromethyl, cyano group, nitro group, alkylamino, arylamino, alkylaminoalkyl, arylaminoalkyl, aminoalkylamino, group, alkoxyalkyl, carboxyalkyl, alkoxycarbonylalkyl, aminocarbonylalkyl, acyl , aralkoxycarbonyl, carboxylic acid, sulfonic acid, sulfonyl, phosphonic acid, aryl, heteroaryl, heterocyclic and aliphatic group. It is understood that the substituent may be further substituted.

Halogen means fluorine, chlorine, bromine and iodine.

The terms alkylamine and dialkylamine refer to the radicals -IN(alkyl) and -N(alkyl)2, respectively.

The term alkyl phosphate means -O-P(Q')(Q)-O-R, wherein Q' and Q are each independently O, S, N(R)2, optionally substituted with an alkyl or alkoxy group; and R is optionally substituted alkyl, ω-aminoalkyl, or <^(substituted)aminoalkyl.

The term alkyl phosphorothioate means an alkyl phosphate where at least one of Q' or Q is S.

The term alkylphosphonate means an alkyl phosphate where at least one of Q' or Q is alkyl.

The term hydroxyalkyl means an -O-alkyl radical.

The term alkyl heterocycle means an alkyl where at least one methylene has been replaced by a heterocycle.

The term ω-aminoalkyl means the radical -alkyl-IN ₂ . The term 'b-(substituted)aminoalkyl' means ω-aminoalkyl where at least one of the H atoms of N has been replaced by an alkyl.

The term ω-phosphoalkyl means -alkyl-O-PiO')O)-O-K, where Q' and Q are each independently O or S and R is optionally substituted alkyl.

The term ω-thiophosphoalkyl means ω-phosphoalkyl where at least one of Q' or Q is S.

In some embodiments, the methods of the invention may require the use of protecting groups. The protecting group methodology is well known to those skilled in the art (see, for example, Protective Groups in Organic Synthesis, Green, T.W. et. al., WileyInterscience, New York City, 1999). Briefly, protecting groups in the context of the present invention are any group that reduces or eliminates unwanted reactivity of a functional group. A protecting group can be added to a functional group to mask its reactivity during certain reactions, and then removed to release the original functional group. In some embodiments of the present invention, an alcohol protecting group is used. An alcohol protecting group is any group that reduces or eliminates unwanted reactivity of an alcohol functional group. Protecting groups can be added and removed using methods well known to those skilled in the art.

Lipid particles.

Agents and/or aminolipids to be tested in the liver cell model screening described herein can be contained in lipid particles. Lipid particles include, but are not limited to, liposomes. As used herein, a liposome is a structure having lipid-containing membranes enclosing an aqueous interior region. Liposomes may have one or more lipid membranes. The invention involves both single-layer liposomes, which are called unilamellar, and multilayer liposomes, which are called multilamellar. When complexed with nucleic acids, lipid particles can also be lipoplexes, which consist of bilayers of cationic lipids sandwiched between layers of DNA, as described, for example, in Feigner, Scientific American.

The lipid particles may further include one or more additional lipids and/or other components, such as cholesterol. Other lipids may be included in liposome compositions for various purposes, such as preventing lipid oxidation or attaching

- 13 046694 gands on the surface of the liposome. Any of a number of lipids may be present, including amphipathic, neutral, cationic and anionic lipids. Such lipids can be used alone or in combination. Specific examples of additional lipid components that may be present are described below.

Additional components that may be present in the lipid particle include bilayer-stabilizing components such as polyamide oligomers (see, for example, US Pat. No. 6,320,017), peptides, proteins, detergents, lipid derivatives such as PEG-linked phosphatidylethanolamine and PEG-conjugated ceramides (see US patent No. 5885613). In some embodiments, the lipid particle includes a targeting agent, such as a targeting lipid described herein.

The lipid particle may include one or more of a second aminolipid or cationic lipid, a neutral lipid, a sterol, and a lipid selected to reduce aggregation of the lipid particles during their formation, which may result from steric stabilization of the particles preventing charge-induced aggregation during formation.

As used herein, the term cationic lipid is intended to include lipids that have one or two fatty acid or fatty alkyl chains and an amine-containing head group (including an alkylamino or dialkylamino group) that can be protonated to form a cationic lipid at physiological pH. In some embodiments, the cationic lipid is referred to as an aminolipid.

Other cationic lipids may include lipids having other variants of fatty acid groups and other dialkylamino groups, including those in which the alkyl substituents are different (for example, N-ethyl-N-methylamino-, N-propyl-N-ethylamino-, etc.) In general , lipids (eg, cationic lipid) having less saturated acyl chains are more easily separated by size, in particular when complexes are separated by size in the region of less than 0.3 μm, for filter sterilization purposes. Cationic lipids containing unsaturated fatty acids with carbon chain lengths ranging from C1 to _C20 are preferred. Other scaffolds may be used to separate an amino group (eg, the amino group of a cationic lipid) and the fatty acid or fatty alkyl moiety of the cationic lipid. Suitable scaffolds are known to one skilled in the art.

In some embodiments, cationic lipids contain at least one protonatable or deprotonatable group such that the lipid is positively charged at a pH equal to or lower than physiological pH (e.g., pH 7.4) and is neutral at a second pH value, preferably equal to or above physiological pH. Such lipids are also called cationic lipids. It is certainly understood that the addition or removal of protons as a function of pH is an equilibrium process and that the reference to a charged or neutral lipid refers to the nature of the predominant species and does not require that all lipids be present in a charged or neutral form. Lipids having more than one protonatable or deprotonatable group or being zwitterionic are not excluded from use in the invention.

In some embodiments, protonatable lipids (i.e., cationic lipids) have a pKa of the protonatable moiety ranging from about 4 to about 11. The most preferred pKa values are from about 4 to about 7, since such lipids will be in the cationic form at lower pH values during the formulation stage, while the particles will have a substantially (though not completely) neutral surface at a physiological pH of approximately 7.4. One advantage of this pKa value is that at least some of the nucleic acids associated with the outer surface of the particle will lose electrostatic interaction at physiological pH and will be removed by simple dialysis; thereby significantly reducing the susceptibility of particles to clearance.

Examples of lipids that reduce particle aggregation during formation include polyethylene glycol (PEG)-modified lipids, G _M1 monosialogangliosides, and polyamide oligomers (PAO), as described in US Pat. No. 6,320,017). Other compounds with uncharged, hydrophilic, steric barrier moieties that prevent aggregation during formulation, such as PEG, G _M1 or ATTA, can also be attached to lipids for use in the methods and compositions of the invention. ATTA lipids are described, for example, in US Pat. No. 6,320,017, and PEG-lipid conjugates are described, for example, in US Pat. Nos. 5,820,873, 5,534,499, and 5,885,613. Typically, the concentration of the lipid component selected to reduce aggregation is from about 1 to 15 % (molar percentage of lipid).

Examples of lipids that reduce aggregation and/or are suitable for conjugation with nucleic acid agents that can be used in model screening using liver cells include polyethylene glycol (PEG)-modified lipids, G _M1 monosialogangliosides, and polyamide oligomers (PAOs), such as those described in US Pat. No. 6,320,017. Other compounds with uncharged, hydrophilic, steric barrier moieties that prevent aggregation during formulation, such as PEG, G _M1 or ATTA, can also be attached to lipids for use in methods and compositions according to the invention.

- 14 046694

ATTA lipids are described, for example, in US Pat. No. 6,320,017, and PEG-lipid conjugates are described, for example, in US Pat. Nos. 5,820,873, 5,534,499, and 5,885,613. Typically, the concentration of the lipid component selected to reduce aggregation is from about 1 to 15% (molar percentage of lipid).

Specific examples of PEG-modified lipids (or lipid-polyoxyethylene conjugates) that are useful in the invention may have various anchoring lipid moieties to position the PEG moiety on the surface of the lipid vesicle. Examples of suitable PEG-modified lipids include PEG-modified phosphatidylethanolamine and phosphatidic acid, PEG-ceramide conjugates (eg, PEG-CerC ₁₄ or PEG-CerC ₂₀ ). which are described in co-pending application USSN 08/486214, incorporated herein by reference, PEG-modified dialkylamines and PEG-modified 1,2-diacyloxypropan-3-amines. Most preferred are PEG-modified diacylglycerol and dialkylglycerol. In some embodiments of the present invention, the total mole percent content of PEG lipids in the particle composition is about 1.5 mole percent. For example, when a particle contains a plurality of PEG lipids described herein, such as the PEG-modified lipids described above and PEG-containing targeted lipids, the total amount of PEG-containing lipids taken together is approximately 1.5 mol. %.

In embodiments where sterically bulky moieties such as PEG or ATTA are conjugated to a lipid anchor, the choice of lipid anchor depends on what type of association with the lipid particle the conjugate must have. It is well known that tePEG (MW 2000)-diastearoylphosphatidylethanolamine (PEG-DSPE) remains in association with the liposome until the particle is removed from circulation, possibly for a period of days. Other conjugates, such as PEG-CerC ₂₀ , have similar ability to maintain association. However, PEG-CerC ₁₄ is rapidly removed from the composition when exposed to serum, with a half _- life of less than 60 minutes, according to a number of quantitative studies. As clearly demonstrated in US patent application SN 08/486214, at least three parameters influence the rate of exchange: the length of the acyl chain, the degree of saturation of the acyl chain, and the size of the lipid head group creating a steric barrier. Compounds having suitable variations of these characteristics may be useful for the invention. For some therapeutic applications, it may be preferable that the PEG-modified lipid be rapidly removed from the lipid nucleic acid particle in vivo and, therefore, the PEG-modified lipid will have relatively short lipid anchoring moieties. In other therapeutic applications, it may be preferable for the lipid nucleic acid moiety to have a longer plasma persistence time in the circulatory system and therefore the PEG-modified lipid will have relatively longer anchoring lipid moieties. Examples of anchoring lipid moieties include those with a length of from about _C14 to about _C22 , preferably from about _C14 to about _C16 . In some embodiments, the PEG moiety is, for example, ιηΠ3Γ-ΝΗ2. has a size of approximately 1000, 2000, 5000, 10000, 15000 or 20000 Da.

It should be noted that anti-aggregation compounds do not necessarily require conjugation to a lipid to function properly. The presence of free PEG or free ATTA in solution may be sufficient to prevent aggregation. If the particles are stable after preparation of the composition, the PEG or ATTA can be removed by dialysis before administration to the subject.

Neutral lipids, when present in a lipid particle, can be any of a number of types of lipids that exist in either uncharged or neutral zwitterionic form at physiological pH. Such lipids include, for example, diacylphosphatidylcholine, diacylphosphatidylethanolamine, ceramide, sphingomyelin, dihydrosphingomyelin, cephalin and cerebrosides. The selection of neutral lipids for use in the particles described herein generally depends on considerations regarding, for example, the size of the liposomes and the stability of the liposomes in the bloodstream. Preferably, the neutral lipid component is a lipid having two acyl groups (ie, diacylphosphatidylcholine and diacylphosphatidylethanolamine). Lipids having different groups of acyl chains with varying chain lengths and degrees of saturation are available or can be isolated or synthesized using well known methods. In one group of embodiments, lipids containing saturated fatty acids with carbon chain lengths ranging from C ₁₄ to C ₂₂ are preferred. In another group of embodiments, lipids containing mono- or diunsaturated fatty acids with carbon chain lengths ranging from C ₁₄ to C ₂₂ are used. Additionally, lipids containing mixtures of saturated and unsaturated fatty acid chains can be used. Preferably, the neutral lipids used in the invention are DOPE, DSPC, DPPC, POPC or any related phosphatidylcholine. Neutral lipids useful in the present invention may also consist of sphingomyelin, dihydrosphingomyelin, or phospholipids with other head groups such as serine and inositol.

The sterol component of the lipid mixture, if present, may be any of the sterols

- 15 046694 standardly used in the field of production of liposomes, lipid vesicles or lipid particles. The preferred sterol is cholesterol.

Other cationic lipids that carry a net positive charge at near physiological pH values, in addition to those specifically described above, can also be included in the lipid particles of the invention. Such cationic lipids include, but are not limited to:

KH-dioleyl-CN-dimethylammonium chloride (DODAC);

Na(2,3-dioleyloxy)propyl-H^-H-triethylammonium chloride (DOTMA);

H,H-distearyl-H,H-dimethylammonium bromide (DDAB);

H-(2,3-dioleoyloxy)propyl)-H,H,H-trimethylammonium chloride (DOTAP);

1,2-dioleyloxy-3-trimethylaminopropane chloride (DOTAP.C1);

3c-(H-(H',H'-dimethylaminoethane)carbamoyl)cholesterol (DC-Chol);

H-(1-(2,3-dioleyloxy)propyl)-H-2-(sperminecarboxamido)ethyl)-H,H-dimethylammonium trifluoroacetate (DOSPA);

dioctadecylamidoglycyl carboxyspermine (DOGS);

1,2-dioleoyl-n-3-phosphoethanolamine (DOPE);

1,2-dioleoyl-3-dimethylammonium propane (DODAP);

H,H-dimethyl-2,3-dioleyloxy)propylamine (DODMA) and

H-(1,2-dimyristyloxyprop-3-yl)-H,H-dimethyl-N-hydroxyethylammonium bromide (DMRIE).

Additionally, a number of commercially available cationic lipid preparations can be used, such as, for example, LIPOFECTIN (containing DOTMA and DOPE, available from GIBCO/BRL) and LIPOFECTAMINE (containing DOSPA and DOPE, available from GIBCO/BRL). In some embodiments of the present invention, the cationic lipid is an aminolipid.

Anionic lipids suitable for use in the lipid particles of the invention include, but are not limited to, phosphatidylglycerol, cardiolipin, diacylphosphatidylserine, diacylphosphatidic acid, N-dodecanoyl phosphatidylethanolamine, N-succinyl phosphatidylethanolamine, N-glutaryl phosphatidylethanolamine, lysylphosphatidylglycerol and other anionic groups connected to neutrals lipids.

In a large number of embodiments, amphipathic lipids are included in the lipid particles of the invention. The term amphipathic lipids means any suitable material where the hydrophobic portion of the lipid material is oriented inward of the hydrophobic phase and the hydrophilic portion is oriented toward the aqueous phase. Such compounds include, but are not limited to, phospholipids, aminolipids, and sphingolipids. Examples of phospholipids include sphingomyelin, phosphatidylcholine, phosphatidylethanolamine, phosphatidylserine, phosphatidylinositol, phosphatidic acid, palmitoyloleoylphosphatidylcholine, lysophosphatidylcholine, lysophosphatidylethanolamine, dipalmitoylphosphatidylcholine (DPPC), dioleoylphosphatidylcholine (DOPC), distearoylphosphatidylcholine lin (DSPC), dimyristoylphosphatidylcholine (DMPC) or dilinoleoylphosphatidylcholine (DLPC). Other non-phosphorus compounds may also be used, such as sphingolipids, members of the glycosphingolipid family, diacylglycerols and β-acyloxy acids. Additionally, such amphipathic lipids can be easily mixed with other lipids such as triglycerides and sterols.

Programmable fusion lipids are also suitable for inclusion in the lipid particles according to the invention. Such lipid particles have a weak tendency to fuse with cell membranes and release their payload before a specific signaling event occurs. This allows for a more uniform distribution of lipid particles after injection into the body or affected area before they begin to fuse with cells. The signal event can be, for example, a change in pH, temperature, ionic environment or time. In the latter case, the fusion-delaying component or masking component, such as an ATTA-lipid conjugate or a PEG-lipid conjugate, may simply be removed by exchange from the membrane of the lipid particle over time. Examples of anchoring lipid moieties include those with a length of from about _C14 to about _C22 , preferably from about _C14 to about _C16 . In some embodiments, the PEG moiety, such as tPEG-NH2, has a size of about 1000, 2000, 5000, 10000, 15000, or 20000 Da.

By the time the lipid particle is properly distributed in the body, it has lost its essential masking agent, thus becoming fusogenic. For other signaling events, it is desirable to select a signal that is associated with the affected area or target cell, such as an increase in temperature at the site of inflammation.

The lipid particle conjugated to the nucleic acid agent may also include a targeting moiety, such as a cell or tissue type specific targeting moiety. Targeting lipid particles using various targeting moieties such as ligands, cell surface receptors, gly

- 16 046694 coproteins, vitamins (eg, riboflavin) and monoclonal antibodies, have been previously described (see, for example, US patents No. 4957773 and 4603044). Examples of targeting moieties include a targeting lipid, such as the targeting lipid described herein. In some embodiments, the targeting lipid is a GalNAc containing targeting lipid, such as OsHAc3-DSG and GalNAc3-PEG-DSG, as described herein. Targeting moieties may include the entire protein or fragments thereof. Targeting mechanisms generally require that the targeting agents be located on the surface of the lipid particle such that the targeting moiety is available for interaction with a target, such as a cell surface receptor. A number of different targeting agents and methods are known and available to those skilled in the art, including those described, for example, in Sapra, P. and Allen, T.M., Prog. Lipid Res. 42(5):439-62 (2003) and Abra, R.M. et al., J. Liposome Res. 12:1-3, (2002).

Application has been proposed for realizing the targeted action of lipid particles, i.e. liposomes whose surface is coated with hydrophilic polymer chains, such as polyethylene glycol (PEG) chains (Allen, et al., Biochimica et Biophysica Acta, 1237:99-108 (1995); DeFrees, et al., Journal of the American Chemistry Society, 118:6101-6104 (1996); Blume, et al., Biochimica et Biophysica Acta, 1149:180184 (1993); 5.013556; Zalipsky, Bioconjugate Chemistry, 4:296-299 (1993); Zalipsky, FEBS Letters, 353:71-74 (1994); Boca Raton Fl (1995) In one method, the targeting ligand, such as an antibody, is attached to the polar head group of the lipids forming the lipid particle. In another method, the targeting ligand is attached to the distal ends of the PEG chains forming the lipid particle. hydrophilic polymer shell (Klibanov, et al., Journal of Liposome Research, 2:321-334 (1992); Kirpotin et al., FEBS Letters, 388:115-118 (1996)).

Standard methods for attaching target agents can be used. For example, phosphatidylethanolamine, which can be activated to attach target agents, or lipophilic derivatives such as the lipid derivative bleomycin may be used. Antibody-targeting liposomes can be constructed using, for example, protein A-embedded liposomes (see Renneisen, et al., J. Bio. Chem., 265:16337-16342 (1990) and Leonetti, et al. ., Proc. Natl Acad. Sci. (USA), 87:2448-2451 (1990). other proteins specific to cellular components, including antigens associated with neoplasms and tumors. Proteins used as targeting moieties can be attached to liposomes by covalent bonds (see Heath, Covalent Attachment of Proteins to Liposomes, 149 Methods). in Enzymology 111-119 (Academic Press, Inc. 1987)). Other targeting mechanisms include the biotinavidin system.

In one embodiment of the present invention, the lipid particle contains a mixture of the cationic lipid of the present invention, neutral lipids (other than the cationic lipid), a sterol (eg, cholesterol) and a PEG-modified lipid (eg, PEG-DMG or PEG-cDMA). In some embodiments, the lipid mixture consists of or substantially consists of a cationic lipid of the present invention, a neutral lipid, cholesterol, and a PEG-modified lipid. In further preferred embodiments, the lipid particle consists of or substantially consists of the above lipid mixture with molar ratios of about 20-70% DLin-M-C3-DMA:5-45% neutral lipid:20-55% cholesterol:0.515% PEG- modified lipid.

In some embodiments, the lipid particle consists of or consists substantially of DLin-M-C3-DMA, DSPC, cholesterol, and either PEG-DMG or PEG-cDMA, for example, in a molar ratio of about 20-60% DLin-M- C3-DMA:5-25% DSPC:25-55% cholesterol:0.5-15% PEG-DMG or PEG-cDMA. In some embodiments, the lipid molar ratio is about 40/10/40/10 (mol% DLin-M-C3-DMA/DSPC/cholesterol/PEG-DMG or PEG-cDMA), 35/15/40/10 (mol% .% DLin-M-C3-DMA/DSPC/cholesterol/PEG-DMG or PEG-cDMA) or 52/13/30/5 (mol.% DLin-M-C3-DMA/DSPC/cholesterol/PEG-DMG or PEG-cDMA).

In another group of embodiments, the neutral lipid, DSPC, in these compositions is replaced by POPC, DPPC, DOPE or SM.

Compositions and formulations of lipid particles containing a therapeutic agent.

The invention includes compositions containing a lipid particle according to the invention and an active agent, wherein the active agent is associated with the lipid particle. In some embodiments, the active agent is a therapeutic agent. In some embodiments, the active agent is encapsulated within the aqueous interior region of the lipid particle. In other embodiments, the active agent is located in one or more lipid layers of the lipid particle. In other embodiments, the active agent is associated with the outer or inner surface of the lipid

- 17 046694 layers of lipid particles.

Fully encapsulated, as used herein, means that the nucleic acid in the particles is not subject to significant degradation following exposure to serum or during nuclease assays, which would cause significant degradation of free DNA. In a fully encapsulated system, preferably less than 25% of the nucleic acid in the particle is degraded under processing conditions under which 100% of the free nucleic acid would normally be degraded, more preferably less than 10%, and most preferably less than 5% of the nucleic acid in the particle is degraded. In another embodiment, complete encapsulation can be determined by analysis using Oligreen®. Oligreen® is an ultrasensitive fluorescent nucleic acid dye for the quantification of oligonucleotides and single-stranded DNA in solution (available from Invitrogen Corporation, Carlsbad, California, USA). Complete encapsulation also implies that the particles are serum resistant, i.e. that they do not undergo rapid disassembly into components after administration in vivo.

Active agents, as used herein, include any molecule or compound capable of producing a desired effect on a cell, tissue, organ or subject. Such effects may be, for example, biological, physiological or cosmetic.

The active agents may be any type of molecule or compound, including, for example, nucleic acids, peptides and polypeptides, including, for example, antibodies, such as, for example, polyclonal antibodies, monoclonal antibodies, antibody fragments; humanized antibodies, recombinant antibodies, human recombinant antibodies and Primatized™ antibodies, cytokines, growth factors, apoptotic factors, differentiation-inducing factors, cell surface receptors and their ligands; hormones; and small molecules, including small organic molecules or compounds.

In one embodiment of the present invention, the active agent is a therapeutic agent, or a salt or derivative thereof. Derivatives of therapeutic agents may themselves be therapeutically active, or they may be prodrugs that become active after further modification. Thus, in one embodiment, the derivative of the therapeutic agent retains some or all of the therapeutic activity compared to the unmodified agent, and in another embodiment, the derivative of the therapeutic agent has no therapeutic activity.

In various embodiments, therapeutic agents include any therapeutically effective agent or drug, such as anti-inflammatory compounds, antidepressants, stimulants, analgesics, antibiotics, contraceptives, antipyretics, vasodilators, anti-angiogenic agents, cytovascular agents, signal transduction inhibitors, cardiovascular medications such as antiarrhythmic agents, vasoconstrictors, hormones and steroids.

In some embodiments, the therapeutic agent is an oncology drug, which may also be referred to as an antineoplastic drug, an anticancer drug, a tumor drug, an antineoplastic drug, or the like. Examples of oncology drugs that can be used in accordance with the invention include, but are not limited to, adriamycin, alkeran, allopurinol, altretamine, amifostine, anastrozole, araC, arsenic trioxide, azathioprine, bexarotene, biCNU, bleomycin, intravenous busulfan, oral busulfan, capecitabine (Xeloda), carboplatin, carmustine, CCNU, celecoxib, chlorambucil, cisplatin, cladribine, cyclosporine A, cytarabine, cytosine arabinoside, daunorubicin, cytoxan, daunorubicin, dexamethasone, dexrazoxane, dodetaxel, doxorubicin, doxorubicin, DTIC, epirubicin, estramustine, etoposide phosphate , ethos and VP-16, exesestan, fk506, flujaranin, fluoruratyl, 5-Fu, hemcitabin (gemzar), gemtuzumabozogozogosycin, acetate, hydroe, hydroxymochevin, Idarubicin, imatiniba, interferor SRT-111) , Letorozol, Leikororin, Listatin, Leiprolide, Levamisol, Litretinoin, Megastral, Melfalah, L-Pam, Method, Metretrexat, Metoksalen, Mitramitsin, Mitomicin, Mitoxanthon, Azotyan Iprit, Paklitaksel, Pademodemaz, Pentostatin, Stew Porphymer, Preminizon, Prem n , streptozocin, STI-571, tamoxifen, taxotere, temozolamide, teniposide, VM-26, topotecan (Hycamtin), toremifene, tretinoin, ATRA, valrubicin, velban, vinblastine, vincristine, VP 16 and vinorelbine. Other examples of oncology drugs that can be used in accordance with the invention are ellipticin and ellipticin analogs or derivatives, epothilones, intracellular kinase inhibitors and camptothecins.

Lipid-nucleic acid particles.

In some embodiments, the lipid particles of the invention are associated with a nucleic acid to form a lipid-nucleic acid particle. In some embodiments, the nucleic acid is completely encapsulated in a lipid particle. When used herein, the term nucleic acid is intended to include any oligonucleotide or polynucleotide.

- 18 046694

Fragments of up to 50 nucleotides are usually called oligonucleotides, while longer fragments are called polynucleotides. In some embodiments, the oligonucleotides of the invention are 20-50 nucleotides in length.

As used herein, the terms polynucleotide and oligonucleotide refer to a polymer or oligomer of nucleotide or nucleoside monomers consisting of naturally occurring bases, sugars and intersugar (backbone) linkages. The terms polynucleotide and oligonucleotide also include polymers or oligomers containing non-naturally occurring monomers or parts thereof with functional similarity. Such modified or substituted oligonucleotides are often superior to native forms due to properties such as increased cellular uptake and increased stability in the presence of nucleases.

Oligonucleotides are referred to as deoxyribooligonucleotides or ribooligonucleotides. A deoxyribooligonucleotide consists of a 5-carbon sugar called deoxyribose covalently linked to a phosphate at the 5' and 3' carbon positions of the sugar to form an alternating, straight-chain polymer. The ribooligonucleotide consists of a similar repeating structure, where the 5-carbon sugar is ribose.

The nucleic acid present in the lipid-nucleic acid particle in accordance with the present invention includes any known form of nucleic acid. The nucleic acids used in this invention may be single-stranded DNA or RNA, or double-stranded DNA or RNA, or a DNA-RNA hybrid. Examples of double-stranded DNA include structural genes, genes containing control and terminator regions, and self-replicating systems such as viral or plasmid DNA. Examples of double-stranded RNA include siRNA and other RNA interference reagents. Single-stranded nucleic acids include, for example, antisense oligonucleotides, ribozymes, microRNAs, and triplex-forming oligonucleotides.

Nucleic acids according to the invention can have different lengths, depending mainly on the specific form of the nucleic acid. For example, in some embodiments, plasmids or genes may be from 1,000 to 100,000 nucleotide residues in length. In some embodiments, the length of the oligonucleotides can vary from about 10 to 100 nucleotides. In various related embodiments, the oligonucleotides, both single-stranded, double-stranded and triple-stranded, can have a length ranging from about 10 to about 50 nucleotides, from about 20 to about 50 nucleotides, from about 15 to about 30 nucleotides, from about 20 to about 30 nucleotides in length.

In some embodiments, the oligonucleotide (or strand thereof) of the invention specifically hybridizes to or is complementary to a target polynucleotide. Specific hybridizing and complementary are terms used to indicate a significant degree of complementarity such that stable and specific binding occurs between the target DNA or RNA and the oligonucleotide. It is understood that an oligonucleotide does not have to be 100% complementary to the nucleotide sequence of its target to be capable of specific hybridization. An oligonucleotide is capable of specific hybridization when binding of the oligonucleotide to a target interferes with the normal function of the target molecule and causes loss of its useful function or expression, and there is a significant degree of complementarity to avoid nonspecific binding of the oligonucleotide to non-target sequences under conditions in which specific binding is required, i.e. e. under physiological conditions in the case of an in vivo assay or therapeutic treatment, or, in the case of an in vitro assay, under the conditions under which such assay is performed. Thus, in other embodiments, a given oligonucleotide includes 1, 2, or 3 base substitutions relative to the region of the gene or mRNA sequence that is targeted or to which it specifically hybridizes.

Nucleic acids involved in RNA interference.

In some embodiments, the lipid nucleic acid particles of the invention are associated with molecules involved in RNA interference (RNAi). Methods involving RNA interference using molecules involved in RNAi. can be used to disrupt the expression of the gene or polynucleotide in question. Over the past 5 years, small interfering RNAs (siRNAs) have largely replaced antisense oligodeoxyribonucleotides and ribozymes as the next generation of targeted oligonucleotide drugs in development. SiRNAs are RNA duplexes, normally 21–30 nucleotides in length, that can associate with a cytoplasmic multiprotein complex known as the RNAi-inducible silencing complex (RISC). RISC in association with siRNA mediates the degradation of homologous mRNA transcripts, so siRNA can be designed to inhibit protein expression with high specificity. Unlike other technologies that use antisense nucleic acids, siRNAs function through a natural mechanism involved in the control of gene expression by non-coding RNAs. This is believed to be the reason why their activity is more potent in vitro and in vivo than that of antisense oligodeoxyribonucleotides or ribozymes. Various reagents involved in PHKi include

- 19 046694 tea siRNAs directed to clinically relevant targets are currently in pharmaceutical development, as described, for example, in de Fougerolles, A. et al., Nature Reviews, 6:443-453 (2007).

Although the first described molecules involved in RNAg were RNA:RNA hybrids containing RNA as both the sense strand and the antisense strand, it has now been shown that sense DNA:antisense RNA hybrids, sense RNA:antisense DNA hybrids and DNA:DNA hybrids are capable of mediating PHKi (Lamberton, J.S. and Christian, A.T. (2003), Molecular Biotechnology, 24:111-119). Thus, the invention includes the use of RNAg molecules containing any of these different types of double-stranded molecules. In addition, it is understood that the molecules involved in RNA1 can be used and introduced into the cell in various forms. Accordingly, as used herein, molecules involved in RNA1 include any and all molecules capable of inducing a RNA response in cells, including but not limited to double-stranded polynucleotides containing two different strands, i.e. a sense strand and an antisense strand, for example small interfering RNA (SIRNA); polynucleotides containing a hairpin loop of complementary sequences that form a double-stranded region, such as shPNKi molecules, and expression vectors that express one or more polynucleotides capable of forming double-stranded polynucleotides by themselves or in combination with another polynucleotide.

A single-stranded siRNA compound, as used herein, means a single-molecule siRNA compound. It may include duplex regions formed by intrastrand pairing, such as a hairpin structure or a frying pan handle, or may be contained within a junction. Single-stranded siRNA compounds can be antisense to the target molecule.

The single-stranded siRNA junction may be of sufficient length to enter RISC and participate in RISC complex-mediated cleavage of the target mRNA. The single-stranded siRNA junction is at least 14, and in other embodiments, at least 15, 20, 25, 29, 35, 40, or 50 nucleotides in length. In some embodiments, it is less than 200, 100, or 60 nucleotides in length.

Hairpin siRNA junctions will have a duplex region equal to or at least 17, 18, 19, 29, 21, 22, 23, 24, or 25 base pairs. The duplex region may be equal to or less in length than 200, 100, or 50. In some embodiments, the size range of the duplex region is 15 to 30, 17 to 23, 19 to 23, and 19 to 21 base pairs in length. The hairpin may have a single-stranded sticky end or a terminal unpaired region. In some embodiments, the overhangs are 2-3 nucleotides in length. In some embodiments, the sticky end is on the sense side of the hairpin, and in some embodiments, on the antisense side of the hairpin.

A double-stranded siRNA junction, as used herein, means a siRNA junction that contains more than one, and in some cases two, strands in which interstrand hybridization may result in a region of duplex structure.

The antisense strand of a double-stranded siRNA junction may be equal to or at least 14, 15, 16, 17, 18, 19, 25, 29, 40, or 60 nucleotides in length. It may be equal to or less in length than 200, 100 or 50 nucleotides. Sizes range from 17 to 25, 19 to 23, and 19 to 21 nucleotides in length. As used herein, the term antisense strand means a siRNA junction strand that is substantially complementary to the target molecule, such as the target RNA.

The sense strand of a double-stranded siRNA junction may be equal to or at least 14, 15, 16, 17, 18, 19, 25, 29, 40, or 60 nucleotides in length. It may be equal to or less in length than 200, 100 or 50 nucleotides. The size range can be from 17 to 25, 19 to 23, and 19 to 21 nucleotides in length.

The double-stranded portion of a double-stranded siRNA junction may be equal to or at least 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 29, 40, or 60 base pairs in length. It may be equal to or less in length than 200, 100 or 50 base pairs. The range of sizes can be 15-30, 17 to 23, 19 to 23, and 19 to 21 base pairs in length.

In many embodiments, the siRNA compound is of substantial length such that it can be cleaved by endogenous molecules, such as Dicer, to form smaller siRNA compounds, such as siRNA agents.

The sense and antisense strands can be selected such that the double-stranded siRNA junction will include a single-stranded or unpaired region at one or both ends of the molecule. Thus, a double-stranded siRNA compound may contain sense and antisense strands paired in such a way as to form an overhang, for example one or two 5' or 3' overhangs, or a 3' overhang of 1-3 nucleotides in length. The formation of sticky ends can be the result of one chain being longer than the other, or the result of two chains of the same length being offset from each other. In some embodiments there will be

- 20 046694 at least one 3' sticky end. In one embodiment, both ends of the siRNA molecule will have a 3' overhang. In some embodiments, the overhang is 2 nucleotides in length.

In some embodiments, the length of the duplex region is from 15 to 30, or 18, 19, 20, 21, 22 and 23 nucleotides in length, for example, in the range of ssiRNA junctions discussed above. The ssiRNA compounds may resemble in length and structure natural Dicer-processed products derived from long dsiRNAs. Embodiments in which two strands of an ssiRNA compound are joined, for example covalently linked, are also included in the invention. Hairpin or other single-stranded structures providing the required double-stranded region and 3' overhang are also within the scope of the present invention.

The siRNA compounds described herein, including double-stranded siRNA compounds and single-stranded siRNA compounds, can mediate silencing of a target RNA, such as an mRNA, such as a protein-coding gene transcript. For convenience, such mRNAs are also referred to herein as silencing mRNAs. Such a gene is also called a target gene. In general, the RNA targeted for silencing is the product of an endogenous gene or a gene of a pathogenic organism. In addition, RNAs other than mRNA, such as tRNAs and viral RNAs, can also be targets.

As used herein, the phrase "RNAG mediates" refers to the ability to cause sequence-specific silencing of a target RNA. Without following any particular theory, it is believed that the silencing process uses the machinery or process of RNA1 and dRNA, such as ssiRNA compounds of 21 to 23 nucleotides in length.

In one embodiment of the present invention, the siRNA compound is substantially complementary to the target RNA, such as the target mRNA, such that the siRNA compound causes silencing of the production of the protein encoded by the target mRNA. In another embodiment, the siRNA compound is strictly complementary to the target RNA, for example, the target RNA and the siRNA compound are annealed, for example, to form a hybrid consisting solely of Watson-Crick base pairs in the region of strict complementarity. The substantially complementary target RNA may comprise an internal region (eg, at least 10 nucleotides in length) that is strictly complementary to the target RNA. Moreover, in some embodiments, the siRNA compound specifically recognizes a single nucleotide difference. In this case, the siRNA junction is mediated by RNA only if strong complementarity is detected in a given region of single nucleotide difference (eg, within 7 nucleotides).

RNA interference (RNI) can be used to specifically inhibit the expression of target polynucleotides. Double-stranded RNA-mediated suppression of gene and nucleic acid expression can be achieved according to the invention by introducing double-stranded RNA, siRNA or shRNA into cells or organisms. SiRNA can be double-stranded RNA, or a hybrid molecule containing both RNA and DNA, for example, one strand of RNA and one strand of DNA. It has been shown that direct introduction of siRNA into the cell can trigger RNA in mammalian cells (Elshabir, S.M., et al. Nature, 411:494-498 (2001)). Additionally, suppression in mammalian cells occurred at the RNA level and was specific to target genes, with a strong correlation between RNA and protein suppression (Caplen, N. et al., Proc. Natl. Acad. Sci. USA, 98:9746-9747 ( 2001)). In addition, a large number of cell lines, including HeLa S3, COS7, 293, NIH/3T3, A549, HT-29, CHO-KI and MCF-7 cells, have been shown to be sensitive to certain levels of siRNA silencing (Brown, D . et al., TechNotes, 9(1):1-7, available online at www.dot.ambion.dot.com/techlib/tn/91/912.html (9/1/02).

Molecules involved in RNA1 and directed to specific polynucleotides can be easily prepared in accordance with techniques known to those skilled in the art. The structural characteristics of effective siRNA molecules were determined. Elshabir, S.M. et al. (2001), Nature, 411:494-498 and Elshabir, S.M. et al. (2001), EMBO, 20:6877-6888. Accordingly, one skilled in the art will appreciate that a wide variety of different siRNA molecules can be used to target a specific gene or transcript. In some embodiments, the siRNA molecules of the invention are double-stranded and have a length of 16-30 or 18-25 nucleotides, including all integer values in between. In one embodiment, the siRNA is 21 nucleotides in length. In some embodiments, siRNAs have 3' overhangs of 0-7 nucleotides in length or 5' overhangs of 0-4 nucleotides in length. In one embodiment, the siRNA molecule has a 3' overhang that is two nucleotides long. In one embodiment, the siRNA is 21 nucleotides long and has 3' overhangs of two nucleotides (ie, it contains a 19 nucleotide complementary region between the sense and antisense strands). In some embodiments, the overhangs are 3' UU or dTdT overhangs.

In general, siRNA molecules are completely complementary to one strand of the target DNA molecule, since the presence of even just one unpaired base pair has been shown to reduce the efficiency of silencing. In other embodiments, the siRNA may have a modified backbone composition,

- 21 046694 with modifications such as 2'-deoxy or 2'-O-methyl. However, in preferred embodiments, the entire siRNA strand may not be entirely composed of 2'-deoxy or 2'-O-modified bases.

In another embodiment, the present invention is directed to the creation of a cell containing a vector for inhibiting gene expression in the cell. The vector contains a regulatory sequence operably linked to a nucleotide sequence encoding at least one strand of one of the double-stranded RNAs according to the invention.

In one embodiment, siRNA target sites are selected by scanning the sequence of the target mRNA transcript for the presence of AA dinucleotide sequences. Each dinucleotide sequence AA, in combination with the approximately 19 nucleotides located on its 3' side, is a potential target site for siRNA. In one embodiment, the siRNA target sites are preferably not located within the 5' and 3' untranslated regions (UTRs) or regions near the start codon (within about 75 bases), since proteins that bind to the regulatory regions may interfere with the binding of the complex siRNAs and endonucleases (Elshabir, S. et al., Nature, 411:494-498 (2001); Elshabir, S. et al., EMBO J. 20:6877-6888 (2001)). In addition, potential target sites can be compared with a corresponding genomic database such as BLASTN 2.0.5, available on the NCBI server at www.ncbi.nlm. and discard potential target sequences that have significant homology to other coding sequences.

In some embodiments, short hairpin RNAs constitute the nucleic acid component of the lipid nucleic acid particles of the invention. Short hairpin RNA (sIiRNA) is a form of hairpin RNA capable of sequence-specific downregulation of target gene expression. Short hairpin RNAs may have advantages over siRNAs in silencing gene expression, as they are generally more stable and less sensitive to degradation in the cellular environment. Gene silencing mediated by these short hairpin RNAs has been found to operate in a number of normal and tumor cell lines and in mammalian cells, including mouse and human cells. Paddison, P. et al., Genes Dev. 16(8):948-58 (2002). Additionally, lines of transgenic cells carrying chromosomal genes that encode sIiRNA created by genetic engineering methods were obtained. These cells are capable of constitutively synthesizing sIiRNA. thereby facilitating long-term or constitutive gene silencing that can be passed down through generations to descendant cells. Paddison, P. et al., Proc. Natl. Acad. Sci. USA, 99(3):1443-1448 (2002).

shRNAs contain a stem-loop structure. In some embodiments, they may contain stems of varying lengths, typically from 19 to 29 nucleotides in length or any number in this range. In some embodiments, the hairpins contain stems that are 19 to 21 nucleotides in length, while in other embodiments, the hairpins contain stems that are 27 to 29 nucleotides in length. In some embodiments, the loop size ranges from 4 to 23 nucleotides in length, although the loop size can be greater than 23 nucleotides without significantly affecting the silencing activity. shRNA molecules can contain mismatched pairs, such as a G-U mismatched pair between the two strands of the shPHR stem, without reducing efficiency. Indeed, in some embodiments, shRNAs are designed to include one or more G-U pairs in the hairpin stem to stabilize the hairpins, for example, during amplification in bacteria. However, typically, complementarity is required between the portion of the stem that binds the target mRNA (the antisense strand) and the mRNA, and even a single base pair with complementarity failure in this region can interfere with silencing. The 5' and 3' sticky ends are not required as they do not appear to be critical for shPHR function although they may be present (Paddison et al. (2002), Genes & Dev. 16(8):948-58) .

MicroRNA.

MicroRNAs (ttRNAs) are a highly conserved class of small RNA molecules that are transcribed from the genomic DNA of plants and animals but are not translated into proteins. Processed ttRNAs are single-stranded RNA molecules of approximately 17–25 nucleotides (nt) in length that are recruited to the RNA-inducible silencing complex (RISC) and have been identified as key regulators of development, cell proliferation, apoptosis, and differentiation. They are believed to play a role in regulating gene expression by binding to the 3' untranslated region of specific mRNAs. RISC mediates the suppression of gene expression by inhibition of translation, transcript cleavage, or both. RISC is also involved in transcriptional silencing in the nucleus in a wide range of eukaryotes.

The number of ttRNA sequences identified to date is large and continues to grow; illustrative examples of such sequences can be found, for example, in miRBase: microRNA sequences, targets and gene nomenclature, Griffiths-Jones S., Grocock R.J., van Dongen S., Bateman A., Enright A.J. NAR, 2006, 34, Database Issue, D140-D144; The microRNA Registry, Griffiths-Jones S. NAR, 2004, 32, Database Issue, D109-D111; as well as on the Internet at

- 22 046694 microrna.dot.sanger.dot.ac.dot.uk/sequences/.

Antisense oligonucleotides.

In one embodiment, the nucleic acid is an antisense oligonucleotide directed to a target polynucleotide. The term antisense oligonucleotide or simply antisense sequence is intended to include oligonucleotides that are complementary to the sequence of the target polynucleotide. Antisense oligonucleotides are single strands of DNA or RNA that are complementary to the selected sequence. In the case of antisense RNA, they prevent the translation of the complementary RNA strand by binding to it. Antisense DNA can be used to target a specific, complementary (coding or non-coding) RNA. If binding occurs, this DNA/RNA hybrid can be degraded by the enzyme RNase H. In some embodiments, antisense oligonucleotides contain from about 10 to about 50 nucleotides, more preferably from about 15 to about 30 nucleotides. The term also includes antisense oligonucleotides, which may not be strictly complementary to the desired target gene. Thus, the invention can be used in cases where the antisense sequence has been found to have non-specific activities for the target molecule, or where the most preferred antisense sequence for a particular application is one or more mismatched pairs with the target sequence.

Antisense oligonucleotides have been shown to be effective and targeted inhibitors of protein synthesis and, therefore, can be used to specifically inhibit protein synthesis by a target gene. The effectiveness of antisense oligonucleotides in inhibiting protein synthesis is well established. For example, the synthesis of polygalactouronase and muscarinic acetylcholine receptor type 2 is inhibited by antisense oligonucleotides targeting their corresponding mRNA sequences (US patents No. 5,739,119 and 5,759,829). Additionally, examples of inhibition by antisense sequences have been shown for the nuclear protein cyclin, multidrug resistance gene (MDG1), ICAM-1, E-selectin, STK-1, striatal GABAA receptor and human EGF (Jaskulski et al., Science. 1988 Jun 10 ; 240(4858):1544-6; Cancer Commun. 1989; Peris et al. ):310-20; US patents No. 5801154; 5789573; 5718709 and 5610288). Additionally, antisense constructs have been described that are capable of inhibition and can be used to treat various abnormalities of cell proliferation, such as cancer (US Pat. Nos. 5,747,470; 5,591,317 and 5,783,683).

Methods for producing antisense oligonucleotides are known to those skilled in the art and can be easily adapted to produce an antisense oligonucleotide directed to any polynucleotide sequence. The selection of an antisense oligonucleotide sequence specific for a given target sequence is based on analysis of the selected target sequence and determination of secondary structure, T _m , binding energy and relative stability. Antisense oligonucleotides may be selected based on their relative inability to form dimers, hairpins, or other secondary structures that would reduce or prevent specific binding to the target mRNA in the host cell. Highly preferred mRNA target regions include regions containing or adjacent to the AUG translation initiation codon and sequences that are substantially complementary to the 5' regions of the mRNA. These considerations regarding secondary structure analysis and target site selection can be implemented, for example, using the OLIGO primer analysis software, ver. 4 (Molecular Biology Insights) and/or software based on the BLASTN 2.0.5 algorithm (Altschul et al., Nucleic Acids Res. 1997, 25(17):3389-402).

Antagomirs.

Antagomirs are RNA-like oligonucleotides that incorporate various modifications to protect against RNases and impart pharmacological properties such as increased uptake into tissues and cells. They differ from normal RNA by having, for example, complete 2-O-methylation of the sugar, phosphorothioate backbone and, for example, a cholesterol residue at the 3' end. Antagomirs can be used for efficient silencing of endogenous sh1RNAs by forming duplexes consisting of an antagomir and endogenous shRNA, which prevents sh1RNA-induced gene silencing. An example of antagomir-mediated shRNA silencing is the silencing of miR122 described in Krutzfeldt et al., Nature, 2005, 438:685-689, the contents of which are expressly incorporated by reference in their entirety. RNA antagomirs can be synthesized using standard solid-phase oligonucleotide synthesis protocols. See US Patent Application Ser. No. 11/502158 and 11/657341 (the disclosure of each of which is incorporated by reference).

The antagomir may contain subunits of a ligand-conjugated monomer and monomers for the synthesis of oligonucleotides. Examples of monomers are described in US patent application No. 10/916185, filed August 10, 2004. Antagomir may have a ZXY structure, such as described in PCT patent application No. PCT/US2004/07070, filed March 8, 2004. Antagomir may be in complex with amphi

- 23 046694 pathetic remainder. Examples of amphipathic residues for use with oligonucleotide agents are described in PCT patent application No. PCT/US2004/07070, filed March 8, 2004.

Aptamers.

Aptamers are nucleic acid molecules or peptide molecules that bind to the particular molecule in question with high affinity and specificity (Tuerk and Gold, Science, 249:505 (1990); Ellington and Szostak, Nature, 346:818 (1990)). DNA or RNA aptamers have been successfully prepared that bind to many different structures, from large proteins to small organic molecules. See Eaton, Curr. Opin. Chem. Biol. 1:10-16 (1997), Famulok, Curr. Opin. Struct. Biol. 9:324-9(1999) and Hermann and Patel, Science, 287:820-5 (2000). Aptamers may be RNA or DNA based and may contain a riboswitch. A riboswitch is a part of an mRNA molecule that directly binds to a small target molecule, and whose binding to the target molecule affects the activity of a gene. Thus, the riboswitch-containing mRNA is directly involved in regulating its own activity, depending on the presence or absence of its target molecule. In general, aptamers are designed through repeated rounds of in vitro selection or, equally, through SELEX (systematic evolution of ligands by exponential enrichment) so that they bind to various molecular targets such as small molecules, proteins, nucleic acids and even cells, tissues and organisms. The aptamer can be produced using any known method, including synthetic, recombinant and purification methods, and can be used alone or in combination with other aptamers specific for the same target. Additionally, as described more fully herein, the term aptamer specifically includes secondary aptamers that contain a consensus sequence obtained by comparing two or more known aptamers to a given target.

Ribozymes.

In accordance with another embodiment of the present invention, lipid-nucleic acid particles are associated with ribozymes. Ribozymes are complexes of RNA and protein that have a specific catalytic domain that has endonuclease activity (Kim and Cech, Proc. Natl. Acad. Sci. USA. 1987 Dec; 84(24):8788-92; Forster and Symons, Cell. 1987 Apr 24;49(2):211-20). For example, a large number of ribozymes accelerate phosphoester transfer reactions with a high degree of specificity, often cleaving only one of several phosphoester bonds in an oligonucleotide substrate (Cech et al., Cell. 1981 Dec; 27(3 Pt 2):487-96; Michel and Westhof, J. Mol. Biol. 1990 Dec 5; 216(3):585-610; Reinhold-Hurek and Shub, Nature 1992; This specificity was due to the requirement that the substrate bind through specific base-pairing interactions to the internal adapter sequence (IGS) of the ribozyme before the chemical reaction occurs.

At least six major types of naturally occurring enzymatic RNAs are currently known. Each can catalyze the hydrolysis of phosphodiester bonds in RNA in trans (and thus can cleave other RNA molecules) under physiological conditions. In general, enzymatic nucleic acids act by initially binding to a target RNA. This binding occurs through the part of the catalytic nucleic acid designed to bind to the target and located in close proximity to the enzymatic part of the molecule, which cleaves the target RNA. Thus, the enzymatic nucleic acid first recognizes and then binds to the target RNA through complementary base pairing and upon binding to the correct site, carries out catalytic cleavage of the target RNA. Rational cleavage of such target RNA will lead to disruption of the ability to synthesize the encoded protein. After an enzymatic nucleic acid has bound and cleaved its RNA target, it is released from the RNA to find its next target and can repeatedly bind and cleave new targets.

The enzymatic nucleic acid molecule may take the form of a hammerhead structure, a hairpin structure, hepatitis δ virus RNA, a group I intron, or RNA within RNase P (in association with an RNA adapter sequence) or a Neurospora VS RNA motif, for example. Specific examples of hammerhead motifs are described in Rossi et al., Nucleic Acids Res. 1992 Sep 11; 20(17):4559-65. Examples of hairpin motifs are described in Hampel et al. (Eur. Pat. Appl. Publ. No. EP 0360257), Hampel and Tritz, Biochemistry 1989 Jun 13; 28(12):4929-33; Hampel et al., Nucleic Acids Res. 1990 Jan 25; 18(2):299-304 and US Patent No. 5631359. An example of the hepatitis δ virus motif is described in Perrotta and Been, Biochemistry. 1992 Dec 1; 31(47):11843-52; an example of an RNase P motif is described in Guerrier-Takada et al., Cell. 1983 Dec; 35(3 Pt 2):849-57; the VS motif of the Neurospora RNA ribozyme is described by Collins (Saville and Collins, Cell. 1990 May 18; 61(4):685-96; Saville and Collins, Proc. Natl. Acad. Sci. USA. 1991 Oct 1; 88( 19):8826-30; Collins and Olive, Biochemistry 1993 Mar 23; 32(11):2795-9); and an example of a group I intron is described in US Pat. No. 4,987,071. Important features of the enzymatic nucleic acid molecules used in accordance with the present invention are that they

- 24 046694 have a specific substrate binding site that is complementary to one or more regions of the DNA or RNA of the target gene, and that they have a nucleotide sequence within or around this substrate binding site that imparts RNA cleavage activity to the molecule. Thus, ribozyme designs should not be limited to the specific motifs mentioned in this document.

Methods for producing ribozymes directed to any polynucleotide sequence are known to those skilled in the art. Ribozymes can be designed as described in International Patent Application Publication No. WO 93/23569 and International Patent Application Publication No. WO 94/02595, each of which is specifically incorporated herein by reference, and synthesized for in vitro and in vivo as described herein.

Ribozyme activity can be optimized by varying the length of the ribozyme binding arm or by chemically synthesizing ribozymes with modifications that prevent their degradation by serum ribonucleases (see, for example, International Patent Application Publication No. WO 92/07065; International Patent Application Publication No. WO 93/ 15187; International Patent Application Publication No. WO 91/03162; European Patent Application Publication No. 92110298.4; US Patent Application Publication No. WO 94/13688, which describe various chemical modifications that can be made to the residues. sugars in the enzymatic RNA molecule), modifications that enhance their effectiveness in cells, and with the removal of stem II bases to reduce RNA synthesis time and chemical requirements.

Additional oligonucleotide sequences (ODN) of specific nucleic acids suitable for use in the compositions and methods of the invention are described in US Patent Application No. 60/379,343, US Patent Application No. 09/649527; international publication WO 02/069369; international publication No. WO 01/15726; US Pat. No. 6,406,705 and Raney et al., Journal of Pharmacology and Experimental Therapeutics, 298:1185-1192 (2001). In some embodiments, the ODNs used in the compositions and methods of the present invention have a phosphodiester (PO) backbone or a phosphorothioate (PS) backbone and/or at least one methylated cytosine residue within a CpG motif.

Modifications of nucleic acids.

In the 1990s. DNA-based antisense oligodeoxynucleotides (ODNs) and ribozymes (RNAs) represented a promising new paradigm for drug discovery and development, but their in vivo application was hampered by endo- and exonuclease activity and a lack of methods for successful intracellular delivery. The problem associated with degradation has been effectively addressed following extensive research into chemical modifications that prevent the recognition of oligonucleotide (oligo) drugs by nuclease enzymes but do not inhibit their mechanism of action. This research has been so successful that drugs currently in development based on antisense ODNs remain intact in vivo for several days, compared with several minutes for unmodified molecules (Kurreck, J. 2003. Antisense technologies. Improvement through novel chemical modifications. Eur. J Biochem 270:1628-44). However, problems with intracellular delivery and mechanisms of action have so far hampered the development of antisense ODNs and ribozymes into clinical products.

RNA duplexes are more stable to nucleases than single-stranded DNA or RNA, and unlike antisense ODNs, unmodified siRNA exhibits sufficient activity when released into the cytoplasm. Despite this, chemical modifications developed to stabilize antisense ODNs and ribozymes have also been systematically applied to siRNAs to determine how tolerable the chemical modification can be and whether pharmacokinetic and pharmacodynamic activity can be improved. RNA interference caused by siRNA duplexes requires the presence of an antisense and a sense strand, which have different functions. Both strands are required for siRNA to enter RISC, but once loaded the two strands separate and the sense strand is degraded while the antisense strand remains to guide RISC to the target mRNA. Entry into RISC is a structurally less stringent process than recognition and cleavage of target mRNA. Therefore, many different chemical modifications of the sense strand are possible, but only limited changes to the antisense strand are allowed (Zhang et al., 2006).

As is known to those skilled in the art, a nucleoside is a combination of a base and a sugar. Nucleotides are nucleosides that additionally contain a phosphate group covalently attached to a sugar moiety within the nucleoside. In the case of nucleosides that contain a pentofuranose sugar, a phosphate group may be attached to any of the 2', 3' or 5' hydroxyl groups of the sugar. When oligonucleotides are formed, phosphate groups covalently link adjacent nucleosides to each other to form a linear polymer compound. In turn, the corresponding ends of this linear polymer structure can be further connected to form a ring structure. Within the structure of an oligonucleotide, phosphate groups are usually said to form the internucleoside backbone of the oligonucleotide. Normal junction or backbone for RNA and

- 25 046694

DNA is a 3'-5' phosphodiester compound.

The nucleic acid used in the lipid-nucleic acid particle in accordance with the present invention includes any known form of nucleic acid. Thus, the nucleic acid may be a modified nucleic acid of a type previously used to improve nuclease resistance and serum stability. It is noteworthy, however, that suitable therapeutic products can also be obtained using the method of the invention to formulate a mixture of lipid-nucleic acid particles from nucleic acids that do not carry modifications in the phosphodiester linkages of natural nucleic acid polymers, and the use of nucleic acids with unmodified phosphodiester linkages ( i.e. nucleic acids in which all internucleoside linkages are phosphodiester bonds) is a preferred embodiment of the present invention.

Frame modifications.

Antisense RNAs, siRNAs, and other oligonucleotides useful in the present invention include, but are not limited to, oligonucleotides containing modified backbones or unnatural internucleoside linkages. Oligonucleotides having modified backbones include those with a phosphorus atom retained in the backbone and those without a phosphorus atom retained in the backbone. Modified oligonucleotides that do not contain a phosphorus atom in their internucleoside backbone can also be considered oligonucleosides. The backbones of the modified oligonucleotides contain, for example, phosphorothioates, chiral phosphorothioates, phosphorodithioates, phosphotriesters, aminoalkyl phosphotriesters, methyl and other alkylphosphonates, including 3'-alkylenephosphonates and chiral phosphonates, phosphinates, phosphoramidates, including 3'-aminophosphoramidate and aminoalkylphosphoramidates, phosphoramidates, thionoalkylphosphonates, thionoalkylphosphotriesters , phosphoroselenate, methylphosphonate or O-alkyl phosphotriester as internucleoside compounds, and boranophosphates containing normal 3'-5' compounds, their analogues with a 2'-5' connection and those with inverted polarity, where adjacent pairs of nucleoside units are connected 3'-5 ' to 5'-3' or 2'-5' to 5'-2'. Some non-limiting examples of some modifications that may be present in the nucleic acid according to the invention are shown in table. 2.

Also included are various salts, mixtures of salts and free acid forms. Examples of US patents that describe the preparation of the above internucleoside linkages include, but are not limited to, US Patent No. 3,687,808; 4469863; 4476301; 5023243; 5177196; 5188897; 5264423; 5276019; 5278302; 5286717; 5321131; 53999676; 5405939; 5453496; 5455233; 5466677; 5476925; 5519126; 5536821; 5541306;5550111;5563253;5571799;5587361 and 5625050.

In some embodiments, modified oligonucleotide backbones that do not contain a phosphorus atom have backbones formed by short-chain alkyl or cycloalkyl internucleoside linkages, mixed heteroatom and alkyl or cycloalkyl internucleoside linkages, or one or more short-chain heteroatom or heterocyclic internucleoside linkages. This includes, for example, those with morpholino bonds (formed in part with the participation of a sugar residue in the nucleoside); siloxane backbones; sulfide, sulfoxide and sulfone skeletons; formacetyl and thioformacetyl backbones; methyleneformacetyl and thioformacetyl backbones; alkene-containing skeletons; sulfamate backbones; methylene imine and methylene hydrazine backbones; sulfonate and sulfonamide backbones; amide backbones and others, having combinations of N, O, S and CH2 as components. Examples of US patents that describe the above oligonucleosides include, but are not limited to, US Patent No. 5,034,506; 5166315; 5185444; 5214134; 5216141; 5235033; 5264562; 5264564; 5405938; 5434257; 5466677; 5470967; 5489677; 5541307; 5561225; 5596086; 5602240; 5610289; 5602240; 5608046; 5610289; 5618704;5623070;5663312;5633360;5677437 and 5677439.

Phosphorothioate modification of the backbone (Table 2, No. 1), where the non-bridged oxygen atom in the phosphodiester bond is replaced by a sulfur atom, is the very first and most common method used to stabilize nucleic acid-based drugs against degradation by nucleases. It is generally believed that PS modifications can be carried out widely on both siRNA strands without significantly affecting activity (Kurreck, J., Eur. J. Biochem. 270:1628-44, 2003). However, PS oligos are known to associate with proteins in a largely nonspecific manner, leading to toxic effects, especially when administered intravenously. Therefore, PS modifications are usually limited to one or two bases at the 3' and 5' ends. The boranophosphate internucleoside linkage (Table 2, no. 2) is a recently developed modification that is apparently more stable than PS, enhances siRNA activity, and has low toxicity (Hall et al., Nucleic Acids Res. 32:5991-6000, 2004).

- 26 046694

table 2

Chemical modifications applied to siRNA and other nucleic acids

No. Abbreviation Name Website modifications Structure 1 PS Phosphorothioate skeleton '—O w he 1 o X v ^{G he} 2 RV Boranophosphate skeleton W K/w ^,g - 3 N3-MU N 3 -methyl uride in Base «A/** n, ( X W '^:>Εί 4 5-BU 5'-bromouracil Base O Gu V4, V 5 5'-SH 5'-iodouracil Base X A, L ^{s 0} Oh 6 2.6-DP 2.6- diaminopurine Base . X Ί *

- 27 046694

7 2-F 2'-fluoro- Sugar o ¹ \Esse W 8 2-OME 2”-O-methyl Sugar O G-CHs 9 2-O-MY 2'-O-(2methoxyethyl) Sugar i—A ¹ \ Zk? ы, о А - <<: g- chj 10 2-DNP 2'-O-(2,4- dinitrophenyl) Sugar W eleven LNA Locked nucleic acid (methylene bridge connecting the 2'-oxygen atom to the 4'-carbon atom in ribose ring) Sugar 1__ bzl G—?=O 12 2'-amino 2'-amino Sugar ? \Ssie X 13 2'-deoxy 2'-deoxy Sugar 14 4'-thio 4'- thioribonucleotide D Sugar 1 U Bits W "Η

Other useful nucleic acid derivatives include those nucleic acid molecules in which the bridging oxygen atoms (those forming phosphoester bonds) have been replaced by -S-, -NH-, -CH2- and the like. In some embodiments, applied changes to antisense, siRNA, or other nucleic acids will not have a complete effect on the negative charges associated with the nucleic acids. Thus, the present invention contemplates the use of antisense, siRNA, and other nucleic acids in which some of the internucleoside bonds are replaced, for example, by neutral methylphosphonate or phosphoramidate bonds. When neutral internucleoside linkages are used, as in some embodiments, less than 80% of the internucleoside linkages in the nucleic acid are replaced in this way, or less than 50% of the internucleoside linkages in the nucleic acid are replaced in this way.

Modifications to the base.

Base modifications are less common than backbone and sugar modifications. All modifications shown in 0.3-6 appear to stabilize siRNA against nuclease action and have little effect on activity (Zhang, H.Y., Du, Q., Wahlestedt, C., Liang, Z. 2006. RNA Interference with chemically modified siRNA Top Chem 6:893-900.

Accordingly, oligonucleotides may also contain modifications or substitutions of nucleobases (often referred to in the art as simply bases). As used herein, unmodified or natural nucleobases include the purine bases adenine (A) and guanine (G) and the pyrimidine bases thymine (T), cytosine (C), and uracil (U). Modified Nuk

- 28 046694 leobases include other synthetic and natural nucleobases such as 5-methylcytosine (5-me-C or m5c), 5-hydroxymethylcytosine, xanthine, hypoxanthine, 2-aminoadenine, 6-methyl and other alkyl derivatives of adenine and guanine, 2 -propyl and other alkyl derivatives of adenine and guanine, 2-thiouracil, 2-thiothymine and 2-thiocytosine, 5-halo-uracil and cytosine, 5-propynyl-uracil and cytosine, 6-azouracil, cytosine and thymine, 5-uracil ( pseudouracil), 4-thiouracil, 8-halo-, 8-amino-, 8-thiol-, 8-thioalkyl-, 8-hydroxyl- and other 8-substituted adenines and guanines, 5-halo-, in particular 5- bromo-, 5-trifluoromethyl and other uracils and cytosines substituted at the 5th position, 7-methylguanine and 7-methyladenine, 8-azaguanine and 8-azaadenine, 7-deazaguanine and 7-deazaadenine and 3-deazaguanine and 3-deazaadenine.

Certain nucleobases are particularly useful for increasing the binding affinity of the oligomeric compounds of the invention, including 5-substituted pyrimidines, 6-azapyrimidines and N-2, N-6 and O-6 substituted purines, including 2-aminopropyladenine, 5-propynyluracil and 5-propynylcytosine. Substitutions to 5-methylcytosine have been shown to increase the stability of the nucleic acid duplex by 0.6-1.2°C (Sanghvi, Y.S., Crooke, S.T. and Lebleu, B., eds., Antisense Research and Applications 1993, CRC Press, Boca Raton, pages 276-278). They can be combined, in some embodiments, with 2'-O-methoxyethyl sugar modifications. US patents that describe the preparation of some of these modified nucleobases, as well as other modified nucleobases, include, by way of non-limiting examples, the above-mentioned US Patent No. 3,687,808, as well as US Patent No. 4,845,205; 5130302; 5134066; 5175273; 5367066; 5432272; 5457187; 5459255;5484908;5502177;5525711;5552540;5587469;5594121,5596091;5614617 and 5681941.

Sugar modifications.

Most modifications of sugar groups occur at the 2'-OH group of the sugar ring in RNA, which is a standard chemically reactive site (Manoharan, M. 2004. RNA interference and chemically modified small interfering RnAs. Curr. Opin. Chem. Biol. 8:570- 9; Zhang, H.Y., Du, C., Liang, Z. 2006. RNA Interference with chemically modified siRNA. Chem. The 2'-F and 2'-OME modifications (0.7 and 8) are common and both increase stability; the 2'-OME modification does not reduce activity if it is limited to less than 4 nucleotides per strand (Holen, T., Amarzguioui , M., Babaie, E., Prydz, H. 2003. Similar behavior of single- and double-stranded siRNAs suggests they act through the standard RNAg pathway Nucleic Acids Res 31:2401-7). The 2'-O-MOE (0.9) modification is most effective for siRNA when the location of the modified bases is limited to the central region of the molecule (Prakash, T.P., Allerson, C.R., Dande, P., Vickers, T.A., Sioufi, N., Jarres, R. , Baker, B.F., Swayze, E.E., Griffey, R.H., Bhat, B. 2005. Positional effect of chemical modifications on short interference of RNA activity in mammalian cells. J. Chem. Other modifications found to stabilize siRNA without loss of activity are shown in 0.10-14.

Modified oligonucleotides may also contain one or more substituted sugar residues. For example, the present invention includes oligonucleotides that contain one of the following residues at the 2' position: OH; F; O-, S-, or N-alkyl, O-alkyl-O-alkyl, O-, S-, or N-alkenyl, or O-, S-, or N-alkynyl, wherein alkyl, alkenyl and alkynyl may be substituted or unsubstituted with Q-Cu-alkyl or C ₂ -C ₁₀ -alkenyl or alkynyl. Particularly preferred is O|(CH2)nO|mCH _; , O(CH2)nOCH3, O(CH2)2ON(CH3)2, O ₍ CH2)nNH2, O(CH2)nCH3, O(CH2)nONH2 and O( _CH2 ) _nON [( _CH2 ) _nCH3 )] ₂ where n and m are from 1 to about 10. Other preferred oligonucleotides contain one of the following residues at the 2' position: C ₁ -C ₁₀ -lower alkyl, substituted lower alkyl, alkaryl, aralkyl, O-alkaryl or O -aralkyl, SH, SCH3, OCN, Cl, Br, CN, CF3, OCF3, SOCH3, SO2CH3, ONO2, NO2, N3, NH2, heterocycloalkyl, heterocycloalkaryl, aminoalkylamino, polyalkylamino, substituted silyl, RNA-cleaving group, reporter group, intercalator , a group for improving the pharmacokinetic properties of the oligonucleotide or a group for improving the pharmacodynamic properties of the oligonucleotide and other substituents having similar properties. One modification includes 2'-methoxyethoxy (2'-O-CH ₂ CH ₂ OCH ₃ , also known as 2'-O-(2-methoxyethyl) or 2'-MOE) (Martin et al., Helv. Chim. Acta , 1995, 78, 486-504), i.e. alkoxyalkoxy group. Other modifications include 2'-dimethylaminooxyethoxy, i.e. the O(CH ₂ ) ₂ ON(CH ₃ ) ₂ group, also known as 2'-DMAOE and 2'-dimethylaminoethoxyethoxy (2'-DMAEOE).

Additional modifications include 2'-methoxy (2'-O- _CH3 ), 2'-aminopropoxy (2'-OCH ₂ CH ₂ CH ₂ NH ₂ ) and 2'-fluoro (2'-F). Similar modifications can also be introduced at other positions within the oligonucleotide, particularly at the 3' position of the sugar of the 3'-terminal nucleotide or linked 2'-5' oligonucleotides and at the 5' position of the 5'-terminal nucleotide. Oligonucleotides may also contain sugar mimetics, such as cyclobutyl moieties in place of the pentofuranosyl sugar. Examples of US patents that describe the preparation of structures of such modified sugars include, but are not limited to, US Patent No. 4,981,957; 5118800; 5319080; 5359044; 5393878; 5446137; 5466786; 5514785; 5519134; 5567811; 5576427; 5591722; 5597909; 5610300; 5627053; 5639873; 5646265; 5658873;5670633 and 5700920.

In other oligonucleotide mimetics, both sugar and internucleoside bonds, e.g. skeleton, nucleo

- 29 046694 tide units are replaced by new groups, although the bases are retained for hybridization with the corresponding target nucleic acid compound. One such oligomeric compound, an oligonucleotide mimetic that has been shown to have excellent hybridization properties, is called a peptide nucleic acid (PNA). In the PNA compound, the sugar backbone of the oligonucleotide is replaced by an amide-containing backbone, in particular an aminoethylglycine backbone, the nucleobases are retained and are linked directly or indirectly to the nitrogen atoms of the azo groups of the amide part of the backbone. Examples of US patents that describe the preparation of PNA compounds include, but are not limited to, US Patent Nos. 5,539,082, 5,714,331, and 5,719,262. Additional ideas regarding PNA compounds can be found in Nielsen et al. (Science, 1991, 254, 1497-1500).

Some embodiments of the present invention are oligonucleotides with phosphorothioate backbones and oligonucleosides with heteroatomic backbones, and in particular -CH2-NH-O-CH2-, -CH ₂ -N(CH ₃ )-O-CH ₂ - (called methylene ( methylamino) or MMI-backbone), -CH ₂ -ON(CH ₃ )-CH ₂ -, -CH2-N(CH ₃ )-N(CH ₃ )-CH2- and -ON(CH ₃ )-CH ₂ - CH ₂ - (where the natural phosphodiester backbone is represented as -O-P-O-CH2-) according to the above US patent No. 5489677, and amide backbones according to the above US patent No. 5602240. Also preferred are oligonucleotides having the morpholine backbone structure according to the above patent US No. 5034506.

The sugar moiety may also contain one or more carbon atoms that have a stereochemical configuration opposite that of the corresponding carbon atom in the ribose. Thus, the oligonucleotide may include nucleotides containing, for example, arabinose as a sugar. The monomer may have an α-linkage at the 1' position of the sugar, such as α-nucleosides. Oligonucleotides may also include baseless sugars that lack a nitrogenous base at the C-1' atom. These baseless sugars may further contain modifications on one or more of the constituent sugar atoms. Oligonucleotides may also contain one or more sugars present in the L-form, for example L-nucleosides.

Chimeric oligonucleotides.

It is not required that all positions in a given compound be uniformly modified, and in fact more than one of the above modifications may be involved in a single compound or even a single nucleoside within an oligonucleotide. Some preferred oligonucleotides of the present invention are chimeric oligonucleotides. Chimeric oligonucleotides or chimeras in the context of the present invention are oligonucleotides containing two or more chemically different regions, each of which consists of at least one nucleotide. These oligonucleotides typically contain at least one region of modified nucleotides that confer one or more beneficial properties (such as, for example, increased nuclease resistance, increased cellular uptake, increased binding affinity for target RNA), and a region that is substrate for RNase H digestion.

In one embodiment, the chimeric oligonucleotide contains at least one region modified to increase binding affinity to the target. The affinity of an oligonucleotide for its target is routinely determined by measuring the T _m of the oligonucleotide/target pair, which is equal to the temperature at which the oligonucleotide and target dissociate; dissociation is determined spectrophotometrically. The higher the T _m value, the greater the affinity of the oligonucleotide for the target. In one embodiment, the region of the oligonucleotide that is modified to increase binding affinity to the target mRNA contains at least one nucleotide modified at the 2' sugar position, most preferably 2'-Ο-ηλκηλ, 2'-0^km-0 ^km or 2'-fluoro-modified nucleotide. Such modifications are routinely incorporated into oligonucleotides, and these oligonucleotides have been shown to have a higher Tm (ie, higher target binding affinity) than 2'-deoxyoligonucleotides acting against that target. The effect of this increased affinity is to significantly enhance the oligonucleotide's inhibition of target gene expression.

In another embodiment of the present invention, the chimeric oligonucleotide contains a region that serves as a substrate for RNase H. Naturally, it is understood that the oligonucleotides may include any combination of the various modifications described herein.

Another modification of the oligonucleotides of the invention involves chemically attaching to the oligonucleotide one or more residues or conjugates that enhance the activity, cellular distribution, and cellular uptake of the oligonucleotide. Such conjugates and methods for their preparation are known to those skilled in the art.

One skilled in the art will appreciate that for in vivo utility, such as therapeutic efficacy, a good rule of thumb is that if a thiol-modified version of a sequence works in free form, then encapsulated particles of the same sequence, of any chemical nature, will also be effective. Encapsulated particles may also have a wider range of in vivo usefulness, demonstrating effectiveness in conditions and

- 30 046694 model systems that are known not to respond to other antisense oligonucleotide therapies. One skilled in the art will recognize that by using the present invention, it may be found that older model systems will begin to respond to antisense oligonucleotide therapy. Additionally, previously discarded antisense sequences or chemical compounds may be reviewed and found to be effective in the application of the present invention.

The oligonucleotides used in accordance with the present invention can be conveniently and routinely prepared using a well-known solid phase synthesis method. Equipment for this synthesis is sold by several suppliers, including Applied Biosystems. Any other method for such synthesis may also be used; the actual synthesis of oligonucleotides depends on the potential capabilities of the specialist performing it. It is also well known that similar methods can be used to prepare other oligonucleotides, such as phosphorothioate and alkylated derivatives.

Immunostimulatory oligonucleotides.

The nucleic acids associated with the lipid particles of the present invention may be immunostimulatory, including immunostimulatory oligonucleotides (ISS; single- or double-stranded), capable of inducing an immune response when administered to a subject, which may be a mammal or other patient. ISS include, for example, certain palindromic sequences leading to the formation of hairpin secondary structures (see Yamamoto S., et al. (1992), J. Immunol. 148:4072-4076), or CpG motifs, as well as other known characteristic ISS elements (such as multi-G domains, see WO 96/11266).

The immune response can be an innate or adaptive immune response. The immune system is divided largely into the innate immune system and the acquired adaptive immune system in vertebrates, the latter of which is further subdivided into humoral and cellular components. In some embodiments, the immune response may be mucosal.

In some embodiments, the immunostimulatory nucleic acid is immunostimulatory only when administered in combination with a lipid particle and is not immunostimulatory when administered in its free form. In accordance with the present invention, such an oligonucleotide is considered to be immunostimulatory.

Immunostimulatory nucleic acids are considered to be non-sequence specific if they are not required to specifically bind to the target polynucleotide and reduce its expression in order to provoke an immune response. Thus, some immunostimulatory nucleic acids may contain a sequence corresponding to a region of a naturally occurring gene or mRNA, but still be considered non-sequence specific immunostimulatory nucleic acids.

In one embodiment, the immunostimulatory nucleic acid or oligonucleotide contains at least one CpG dinucleotide. The CpG oligonucleotide or dinucleotide may be unmethylated or methylated. In another embodiment of the present invention, the immunostimulatory nucleic acid contains at least one CpG dinucleotide having a methylated cytosine. In one embodiment, the nucleic acid contains one CpG dinucleotide, wherein the cytosine in said CpG dinucleotide is methylated. In a specific embodiment, the nucleic acid contains the sequence 5' TAACGTTGAGGGGCAT 3'.

According to another embodiment, the nucleic acid contains at least two CpG dinucleotides, wherein at least one cytosine in the CpG dinucleotides is methylated. In a further embodiment, each cytosine in the CpG dinucleotides present in the sequence is methylated. In another embodiment of the present invention, the nucleic acid contains a plurality of CpG dinucleotides, wherein at least one of said CpG dinucleotides contains methylated cytosine.

In a specific embodiment, the nucleic acid contains the sequence 5' TTCCATGACGTTCCCTGACGT 3'.

In yet another specific embodiment, the nucleic acid sequence comprises the sequence 5' TCCATGACGTTCCCTGACGT 3', with the two cytosines in bold being methylated. In some embodiments, the ODN is selected from the group of ODNs consisting of ODN 1, ODN 2, ODN 3, ODN 4, ODN 5, ODN 6, ODN 7, ODN 8, and ODN 9, as shown below.

- 31 046694

Table 3

Examples of immunostimulatory oligonucleotides (ODN)

ODN NAME EQ ID ODN SEQUENCE (5'-3'). ODN 1 human c-myc 5 '-TAACGTTGAGGGGCAT-3 *ODN Im 5 '-TAAZGTTGAGGGGCAT-3 ODN 2 5'-TCCATGACGTTCCCTGACGTT-3 *ODN 2m 5 '-TCS ATGAZGTTCCTGAZGTT-3 ODN3 5'-TAAGC ATACGGGGTGT-3 ODN 5 5'-AACGTT-3 ODN 6 5 '-GATGCTGTGTCGGGGTCTCCGGGC-3 ¹ ODN 7 5 '-TCGTCGTTTTGTCGTTTTGTCGTT-3 ¹ ODN 7m 5 '-TZGTZGTTTTGTZGTTTTGTZGTT-3 ¹ ODN 8 5 '-TCC AGGACTTCTCTCAGGTT-3' ODN 9 5 '-TCTCCCAGCGTGCGCC AT-3' ODN 10 mouse intracellular adhesion molecule-1 5'-TGC ATCCCCCAGGCCACCAT-3 ODN 11 intracellular adhesion molecule-1 person 5'-GCCCAAGCTGGCATCCGTCA-3' ODN 12 intracellular adhesion molecule-1 person 5'-GCCCAAGCTGGCATCCGTCA-3' ODN 13 erb-B-2 persons 5'-GGT GCTCACTGC GGC-3' ODN 14 person's s-tus 5'-AASS GTT GAG GGG CAT-3' ODN 15 s-tus person 5'-TAT GCT GTG CCG GGG TCT TCG Goes' ODN 16 5'-GTGCCG GGGTCTTTCGGGC-3' ODN 17 receptor human insulin-like growth factor 1 5'-GGACCCTCCTCCGGAGCC-3' ODN 18 receptor human insulin-like growth factor 1 5'-TCS TCC GGA GCC AGA CTT-3' ODN 19 human epidermal growth factor receptor 5'-AAC GTT GAG GGG CAT-3' ODN 20 epidermal growth factor receptor 5'-CCGTGGTCA TGCTCC-3' ODN 21 vascular endothelial growth factor person 5'-CAG CCTGGCTCACCG CCTTGG-3' ODN 22 Phosphokinase C - a mouse 5'-CAG CCA TGG TTC CCC CCA AC-3' ODN23 5'-GTT CTC GCT GGT GAG TTT CA-3' ODN 24 Sun1-2 people 5'-TCT CCCAGCGTGCGCCAT-3' ODN 25 C-Raf-s man 5'-GTG CTC CAT TGA TGC-3' ODN No. 26 human vascular endothelial growth factor receptor-1 5 '-GAGUUCUGAUGAGGCCGAAAGGCCGAAAGUCUG-3' ODN No. 27 5'-RRCGYY-3' ODN No. 28 5'-AACGTTGAGGGGCAT-3' ODN No. 29 5'-CAACGTTATGGGGGAGA-3' ODN No. 30 human s-tus 5 '-TAACGTTGAGGGGCAT-3'

Z stands for methylated cytosine residue. ODN 14 is a 15-mer oligonucleotide, and ODN 1 is the same oligonucleotide containing thymidine added to the 5' end, resulting in ODN 1 becoming a 16-mer. No differences in biological activity were found between ODN 14 and ODN 1, and they both exhibit similar immunostimulatory activity (Mui et al., 2001).

- 32 046694

Additional oligonucleotide sequences (ODN) of specific nucleic acids suitable for use in the compositions and methods of the invention are described in Raney et al., Journal of Pharmacology and Experimental Therapeutics, 298:1185-1192 (2001). In some embodiments, the ODNs used in the compositions and methods of the present invention have a phosphodiester (PO) backbone or a phosphorothioate (PS) backbone and/or at least one methylated cytosine residue within a CpG motif.

Decoy oligonucleotides.

Since transcription factors recognize their relatively short binding sequences even in the absence of surrounding genomic DNA, short oligonucleotides carrying a specific transcription factor binding consensus sequence can be used as a tool to control gene expression in living cells. This strategy involves the intracellular delivery of such decoy oligonucleotides, which are then recognized and bound by their target factors. Occupation of the DNA-binding site of a transcription factor by a decoy renders the transcription factor unable to subsequently bind to the promoter regions of target genes. Decoys can be used as therapeutic agents, either to inhibit the expression of genes that are activated by a transcription factor, or to increase the expression of genes that are suppressed by transcription factor binding. Examples of the use of decoy oligonucleotides can be found in Mann et al., J. Clin. Invest., 2000, 106: 1071-1075, the contents of which are specifically included in the description as a reference.

Superworld.

Supermir refers to a single-stranded, double-stranded or partially double-stranded oligomer or polymer of ribonucleic acid (RNA) or deoxyribonucleic acid (DNA) or both or modifications thereof, which has a nucleotide sequence substantially identical to shgRNA and is antisense to its target. The term includes oligonucleotides consisting of naturally occurring nucleobases, sugars and covalent internucleoside (backbone) linkages and containing at least one non-naturally occurring moiety that performs similar functions. Such modified or substituted oligonucleotides are superior to native forms due to desirable properties such as, for example, increased cellular uptake, increased affinity for the target nucleic acid, and increased stability in the presence of nucleases. In a preferred embodiment, the superworld does not contain a semantic chain, and in yet another preferred embodiment, the superworld does not self-hybridize to a significant extent. The superworld, which is a feature of the present invention, may have a secondary structure, but it is substantially single-stranded under physiological conditions. A superworld that is substantially single-stranded is single-stranded to the extent that less than about 50% (eg, less than about 40, 30, 20, 10, or 5%) of the superworld forms duplexes within the molecule. The superworld may contain a hairpin segment, for example a sequence, preferably at the 3' end, which may self-hybridize to form a duplex region, for example a duplex region of at least 1, 2, 3 or 4 in length and preferably less than 8, 7, 6 or n nucleotides, for example 5 nucleotides. The duplex region can be attached by a linker, for example a nucleotide linker, for example 3, 4, 5 or 6 dT residues, for example modified dT residues. In another embodiment of the present invention, the superworld forms a duplex with a shorter oligo, for example 5, 6, 7, 8, 9 or 10 nucleotides in length, for example, at one or both 3' and 5' ends or at one end and not terminal or middle part of the superworld.

shRNA mimetics.

ShRNA mimics are a class of molecules that can be used to mimic the ability of one or more shRNAs to cause gene silencing. Thus, the term miRNA mimic represents synthetic non-coding RNAs (i.e., shRNAs not obtained by purification from an endogenous shRNA source) that are capable of entering into the RNAg signaling pathway and regulating gene expression. shGRNA mimics can be engineered as mature molecules (e.g., single-stranded) or mimic precursors (e.g., pri- or pre-shRNA). shRNA mimetics may consist of nucleic acid (modified or unmodified nucleic acids), including oligonucleotides containing, but not limited to, RNA, modified RNA, DNA, modified DNA, locked nucleic acids, or nucleic acids with 2'-O,4'- C-ethylene bridges (ENA) or any combination of the above (including DNA-RNA hybrids). In addition, shRNA mimics may contain conjugates that affect delivery, intracellular compartmentalization, stability, specificity, functionality, strand preference, and/or activity. In one design, shgRNA mimics are double-stranded molecules (eg, with a duplex region of about 16 to about 31 nucleotides in length) and contain one or more sequences with identity to the mature strand of the shgRNA. Modifications may include 2' modifications (including 2'-O-methyl modifications and 2'-E modifications) on one or both strands of the molecule and internucleotide

- 33 046694 modifications (for example, phosphorothioate modifications) that enhance the stability and/or specificity of the nucleic acid. In addition, piRNA mimics may contain sticky ends. Overhangs can consist of 1-6 nucleotides at either the 3' or 5' ends of any strand and can be modified to improve stability or enhance functionality. In one embodiment, the piRNA mimic contains a duplex region ranging from 16 to 31 nucleotides in length and one or more of the following patterns of chemical modifications: the sense strand contains 2'-O-methyl modifications of nucleotides 1 and 2 (counting from the 5' end of the sense oligonucleotide) and all C and U residues; antisense strand modifications may include 2'-P modification of all C and U residues, phosphorylation of the 5' end of the oligonucleotide, and stabilized internucleotide bonds associated with the 3' 2-nucleotide overhang.

Antimirs or piRNA inhibitors.

The terms antimir, microRNA inhibitor, miR inhibitor, or inhibitor are synonymous and refer to oligonucleotides or modified oligonucleotides that interfere with the activity of specific pnRNAs. In general, inhibitors are nucleic acid or modified nucleic acids in nature, including oligonucleotides containing RNA, modified RNA, DNA, modified DNA, locked nucleic acids (LNA), or any combination of the above. Modifications include 2'-modifications (including 2'-O-alkyl modifications and 2'-P modifications) and internucleotide modifications (eg, phosphorothioate modifications), which can affect delivery, stability, specificity, intracellular compartmentalization, or activity. In addition, piRNA inhibitors may contain conjugates that affect delivery, intracellular compartmentalization, stability and/or activity. Inhibitors can take on a variety of configurations, including single-stranded, double-stranded (RNA/RNA or RNA/DNA duplexes), and hairpin structures; in general, piRNA inhibitors contain one or more sequences or portions of sequences that are complementary or partially complementary to the mature strand(s) the target Sh1RNA; in addition, the ppRNA inhibitor may also contain additional sequences located on the 5' or 3' side of the sequence that is reverse complementary to the mature ppRNA. The additional sequences may be reverse complementary to sequences that are located adjacent to the mature miRNA in the cp-piRNA. from which the mature miRNA is derived, or the additional sequences may be random sequences (being a mixture of A, G, C or U residues). In some embodiments, one or both of the additional sequences are random sequences capable of forming hairpins. Thus, in some embodiments of the present invention, the sequence that is reverse complementary to the piRNA is flanked on the 5' side and 3' side by hairpin structures. MicroRNA inhibitors, if double-stranded, may contain mismatched pairs between nucleotides on opposing strands. Additionally, miRNA inhibitors can be coupled to conjugate moieties to facilitate uptake of the inhibitor into the cell. For example, the miRNA inhibitor can be coupled to cholesterol-5-(bis-(4-methoxyphenyl)(phenyl)methoxy)-3-hydroxypentylcarbamate), which allows passive uptake of the miRNA inhibitor into the cell. MicroRNA inhibitors, including hairpin tRNA inhibitors, are described in detail in Vermeulen et al., Double-Stranded Regions Are Essential Design Components Of Potent Inhibitors of RISC Function, RNA, 13:723-730 (2007) and in WO 2007/095387 and WO 2008 /036825, the contents of each of which are fully incorporated into the description as references. One skilled in the art can select a sequence from a database for the desired pRNA and design an inhibitor useful for the methods described herein.

Sh-adapter.

Sh-adaptors inhibit polyA sites and are bifunctional oligonucleotides with a target domain complementary to the site in the terminal exon of the target gene and the Sh-domain, which binds to U1 small nuclear RNA, which is a component of U1 snRNA (Goraczniak, et al., 2008, Nature Biotechnology, 27(3), 257-263, the contents of which are expressly incorporated by reference in their entirety). U1 snRNP is a ribonucleoprotein complex that functions primarily in the early steps of spliceosome formation by binding to the exon-intron boundary of pre-mRNA (Brown and Simpson, 1998, Annu Rev. Plant. Physiol. Plant. Mol. Biol. 49:77- 95). Nucleotides 2-11 at the 5' end of U1 snRNA bind by base pairing to the 5' single-stranded region of the pre-mRNA. In one embodiment, the oligonucleotides of the invention are I-adapgors. In one embodiment, the III adapter may be administered in combination with at least one other iRNA agent.

Modifications of oligonucleotides.

Unmodified oligonucleotides may not be optimal for some applications, for example, unmodified oligonucleotides may be prone to degradation, for example, by cellular nucleases. Nucleases are capable of hydrolyzing phosphodiester bonds of nucleic acids. However, chemical modifications of oligonucleotides can provide improved properties and, for example, can

- 34 046694 can make oligonucleotides more stable to nucleases.

Because oligonucleotides are polymers composed of subunits or monomers, many of the modifications described below occur at a position that is repeated in the oligonucleotide, such as modification of a base, sugar, phosphate moiety, or non-bridged oxygen atom of a phosphate moiety. It is not required that all positions in a given oligonucleotide be uniformly modified, and in fact more than one of the above modifications may be involved in a single oligonucleotide or even in a single nucleoside within an oligonucleotide.

In some cases, modification will occur at all positions in the oligonucleotide under consideration, but in many, in fact most, cases this will not occur. As an example, modification may occur only at the 3' or 5' terminal position, may occur only in the internal region, may occur only in terminal regions, for example, at the terminal nucleotide position or at the extreme 2, 3, 4, 5 or 10 nucleotides of the oligonucleotide. The modification may occur in the double-stranded region, single-stranded region, or both.

The modification may occur only in a region spanning two strands of a double-stranded oligonucleotide, or may occur only in a region spanning one strand of a double-stranded oligonucleotide. For example, phosphorothioate modification at the position of the non-bridged oxygen atom can occur only at one or both ends, and can occur only in terminal regions, for example, at a position on the terminal nucleotide or in the last 2, 3, 4, 5 or 10 nucleotides of the chain , or may occur in regions spanning two chains or one chain, particularly at the ends; The 5' end or ends may be phosphorylated.

A modification described herein may be a single modification or a single type of modification included in a plurality of nucleotides, or a modification may be combined with one or more other modifications described herein. The modifications described herein may also be combined in an oligonucleotide, for example, different nucleotides of the same oligonucleotide may have different modifications described herein.

In some embodiments, it is particularly preferred, for example, to increase stability, to include some nucleobases in the overhanging ends or to include modified nucleotides or nucleotide substitutes in single-stranded overhanging ends, for example, in the 5' or 3' overhanging end 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 will be modified, for example, involving modification described herein. Modifications may include, for example, the use of modifications at the 2' OH group of the ribose sugar, such as the use of deoxyribonucleotides, such as deoxythymidine, instead of ribonucleotides, and modifications of the phosphate group, such as phosphorothioate modifications. Sticky ends may not be homologous to the target sequence.

Specific modifications are discussed in more detail below.

Phosphate group.

The phosphate group carries a negative charge. The charge is evenly distributed between two oxygen atoms that are not part of the bridge. However, the phosphate group can be modified by replacing one of the oxygen atoms with another substituent. One of the results of this modification may be increased resistance of the phosphate backbones of the RNA oligoribonucleotide to nucleolytic degradation. Thus, without being bound by any particular theory, in some embodiments it may be desirable to introduce changes resulting in either uncharged linkers or charged linkers with an asymmetrical charge distribution.

Examples of modified phosphate groups include phosphorothioate, phosphoroselenates, boranophosphates, boranophosphate esters, H-phosphonates, phosphoramidates, alkyl or arylphosphonates, and phosphotriesters. In some embodiments, one of the non-bridged oxygen atoms of the phosphate group in the phosphate backbone moiety may be replaced by any of the following moieties: S, Se, BR3 (R = hydrogen, alkyl, aryl), C (i.e., an alkyl group , aryl group, etc.), H, NR ₂ (R - hydrogen atom, alkyl, aryl) or OR (R - alkyl or aryl). The phosphorus atom in an unmodified phosphate group is not chiral. However, replacing one of the non-bridged oxygen atoms with one of the atoms or groups of atoms listed above converts the phosphorus atom into a chiral one; in other words, the phosphorus atom in a phosphate group modified in this way is a stereogenic center. A stereogenic phosphorus atom can have either an R'' configuration (herein referred to as Rp) or an S'' configuration (herein referred to as Sp).

In phosphorodithioates, both oxygen atoms that are not part of the bridge are replaced by sulfur. The central phosphorus atom in phosphorodithioates is not chiral, which prevents the formation of oligoribonucleotide diastereomers. Thus, without following any particular theory, modifications of both non-bridge oxygen atoms that do not introduce a chiral center, for example, the formation of phosphorodithioates, may be desirable due to the fact that mixtures of diastereomers will not be obtained. Thus, atoms not included in the bridge are acidic

- 35 046694 kind can be independently replaced by any atom or group from S, Se, B, C, H, N or OR (R - alkyl or aryl).

The phosphate linker can also be modified by replacing the bridged oxygen atom (i.e., the oxygen atom connecting the phosphate to the nucleoside) with a nitrogen atom (bridged phosphoramidates), a sulfur atom (bridged phosphorothioates), and a carbon atom (bridged methylenophosphonates). ). Replacement of either of the oxygen atoms included in the bridge, or of both oxygen atoms included in the bridge, can occur. When the bridged oxygen atom is the 3' oxygen atom of a nucleoside, replacement by a carbon atom is preferred. When the bridged oxygen atom is the 5' oxygen atom of the nucleoside, replacement by a nitrogen atom is preferred.

Replacement of phosphate group.

The phosphate group can be replaced by connecting groups that do not contain a phosphorus atom. Without following any particular theory, it is believed that since the charged phosphodiester group is the reaction center for nucleolytic degradation, its replacement with a neutral structural mimetic should provide increased stability to nucleases. On the other hand, without being bound by any particular theory, it may be desirable in some embodiments to introduce changes in which a charged phosphate group is replaced by a neutral moiety.

Examples of residues that can replace the phosphate group include methylphosphonate, hydroxylamino, siloxane, carbonate, carboxymethyl, carbamate, amide, thioether, ethylene oxide linker, sulfonate, sulfonamide, thioformacetal, formoacetal, oxime, methyleneimino, methylenemethylimino, methylenehydrazo, methylenedimethylhydrazo, and methyleneoxymethylimino . Preferred substitutions include methylenecarbonylamino and methylenemethylimino groups.

Modified phosphate internucleoside linkages, where at least one of the oxygen atoms attached to the phosphate has been replaced or the phosphate group has been replaced by a non-phosphorus group, are also called non-phosphodiester backbone linkages.

Replacement of the ribophosphate backbone.

Backbones can also be designed to mimic the structure of an oligonucleotide, where the phosphate linker and ribose sugar are replaced with nuclease-resistant nucleoside or nucleotide substitutes. Without following any particular theory, it is believed that the absence of repeating charged groups in the backbone reduces binding to proteins that recognize polyanions (eg, nucleases). On the other hand, without being bound by any particular theory, it may be desirable in some embodiments to introduce variations in which the bases are connected by a neutral backbone of substitutes. Examples include mofilino, cyclobutyl, pyrrolidine nucleoside substitutes, and peptide nucleic acid (PNA) substitute. A PNA based substitute is preferred.

Modifications to the ends of the molecule.

The 3' and 5' ends of the oligonucleotide can be modified. Such modifications can occur at the 3' end, 5' end, or both ends of the molecule. These involve modification or replacement of the entire terminal phosphate or one or more atoms of the phosphate group. For example, the 3' and 5' ends of the oligonucleotide can be conjugated to other functional molecules, such as labeling moieties such as fluorophores (e.g., pyrene, TAMRA, fluorescein, Cy3 or Cy5 dyes), or protecting groups (based on e.g. sulfur, silicone, boron or ester). Functional molecules can be attached to the sugar through a phosphate group and/or a linker. The terminal linker atom may attach to or replace the connecting atom of the phosphate group of either the C-3' or C-5' O, N, S, or C sugar groups. In another embodiment, the linker may be attached to or replace the terminal atom of a nucleotide substitute (eg, PNA).

If a linker/phosphate-functional linker/phosphate molecule array is located between two strands of double-stranded RNA, the array can replace the RNA hairpin loop in a hairpin RNA agent.

Modifications to the ends of the molecule useful for modulating activity include modification of the 5' end with phosphate or phosphate analogs. For example, in preferred embodiments, the double-stranded RNA antisense strands are phosphorylated at the 5' end or contain a phosphoryl analog at the 5' end. 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-O-P(HO)(O)-O-5'); 5'-triphosphate ((HO)2(O)P-O-(HO)(O)P-O-P(HO)(O)-O-5'); 5'-guanosine cap (7-methylated or unmethylated) (7m-G-0-5'-(H0)(0)P-0-(H0)(0)P-0-P(H0)(0)- 0-5'); 5'-adenosine cap (App) or any modified or unmodified nucleotide cap structure (N-O-5'-(HO)(O)P-O-(HO)(O)P-O-P(HO)(O)-O-5') ; 5'-monothiophosphate (phosphorothioate; (HO)2(S)P-O-5'); 5'-monodithiophosphate (phosphorodithioate; (HO)(HS)(S)P-O-5'), 5'-phosphorothiolate ((HO)2(O)P-S-5'); any additional combination of oxygen/sulfur substitutions in a monophosphate, diphosphate or triphosphate (e.g. 5'-a-thiotriphosphate, 5'-y-thiotriphosphate, etc.), 5'-phosphoramidates ((HO)2(O)P -NH-5',

- 36 046694 (HO)(NH2)(O)P-O-5'), 5'-alkylphosphonates (R = alkyl = methyl, ethyl, isopropyl, propyl, etc., e.g. RP(OH)(O)- O-5'-, (OH)2(O)P-5'-CH2-), 5'-alkyl ether phosphonates (R = alkyl ether = methoxymethyl (MeOCH2-), ethoxymethyl, etc., e.g. RP(OH) (O)-O-5'-).

Modifications to the ends of the molecule can also be useful for monitoring distribution, and in such cases, preferred groups to add include fluorophores, such as fluorescein or an Alexa dye, such as Alexa 488. Modifications to the ends of the molecule can also be useful to increase absorption; useful modifications for this include cholesterol . Modifications to the ends of the molecule can also be useful for cross-linking an RNA agent to another residue; Modifications useful for this include mitomycin C.

Nucleobases.

Adenine, guanine, cytosine and uracil are the most common bases found in RNA. These bases can be modified or replaced to give the RNA improved properties. For example, nuclease-resistant oligoribonucleotides can be prepared using these bases or synthetic and natural nucleobases (eg, inosine, thymine, xanthine, hypoxanthine, nubularine, isoguanisine or tubercidin) and any of the above modifications. In another embodiment, a substituted or modified analogue of any of the above bases may be used, such as non-standard bases, modified bases, non-natural bases and universal bases described herein. Non-limiting examples include 2-aminoadenine, 6-methyl and other alkyl derivatives of adenine and guanine, 2-propyl and other alkyl derivatives of adenine and guanine, 5-halouracil and cytosine, 5-propynyl-uracil and cytosine, 6-azouracil, cytosine and thymine , 5-uracil (pseudouracil), 4-thiouracil, 5-halouracil, 5-(2-aminopropyl)uracil, 5-aminoallyluracyl, 8-halo-, amino-, thiol-, thioalkyl-, hydroxyl- and other substituted 8 adenine and guanine, 5-trifluoromethyl- and other 5-substituted uracils and cytosines, 7-methylguanine, 5-substituted pyrimidines, 6-azapyrimidines and N-2, N-6 and O-6-substituted purines, including including 2-aminopropyladenine, 5-propynyluracil and 5-propynylcytosine, dihydrouracil, 3-deaza-5-azacytosine, 2-aminopurine, 5-alkyluracil, 7-alkylguanine, 5-alkylcytosine, 7-deazaadenine, N,N-dimethyladenine, 2 ,6-diaminopurine, 5-aminoallyluracyl, N3-methyluracil, substituted 1,2,4-triazoles, 2-pyridinone, 5-nitroindole, 3-nitropyrrole, 5-methoxyuracil, uracil-5-hydroxyacetic acid, 5-methoxycarbonylmethyluracil, 5 -methyl-2-thiouracil, 5-methoxycarbonylmethyl-2-thiouracil, 5-methylaminomethyl-2-thiouracil, 3-(3-amino-3-carboxypropyl)uracil, 3-methylcytosine, 5-methylcytosine, N'4-acetylcytosine. 2-thiocytosine, M ⁶ -methyladenine, ^-isopentyladenine, 2-methylthio-^ ³ -isopentenyladenine, N-methylguanines or O-alkylated bases. Additional purines and pyrimidines include those described in US Pat. No. 3,687,808, those described in Concise Encyclopedia Of Polymer Science And Engineering, pages 858-859, Kroschwitz, JI, ed. John Wiley & Sons, 1990, and those described in Englisch et al., Angewandte Chemie, International Edition, 1991, 30, 613.

Cationic groups.

Modifications of oligonucleotides may also include the addition of one or more cationic groups to a sugar, base, and/or phosphorus atom of the phosphate or modified phosphate backbone moiety. The cationic group can be attached to any substitutable atom in a natural, non-standard or universal base. The preferred position is one that does not interfere with hybridization, i.e. does not interfere with the formation of interactions in the form of hydrogen bonds necessary for the formation of base pairs. The cationic group can be attached, for example, through the C2' position of a sugar or similar position in a cyclic or acyclic sugar substitute. Cationic groups may include, for example, protonated amino groups, derivatives such as O-AMINE (AMINE = NH2; alkylamino, dialkylamino, heterocyclyl, arylamino, diarylamino, heteroarylamino or diheteroarylamino, ethylenediamine, polyamino); aminoalkoxy, for example O(CH ₂ ) _n AMINE, (for example, AMINE = NH2; alkylamino, dialkylamino, heterocyclyl, arylamino, diarylamino, heteroarylamino or diheteroarylamino, ethylenediamine, polyamino); amino (for example, NH ₂ ; alkylamino, dialkylamino, heterocyclyl, arylamino, diarylamino, heteroarylamino, diheteroarylamino or amino acid) or NH(CH ₂ CH ₂ NH) _n CH ₂ CH ₂ -AMINE (AMINE = NH ₂ ; alkylamino, dialkylamino, heterocyclyl , arylamino, diarylamino, heteroarylamino or diheteroarylamino).

Location within an oligonucleotide.

Certain modifications may be preferentially included in the oligonucleotide at certain positions, such as in the interior of the strand or at the 5' or 3' end of the oligonucleotide. The preferred position of the modification within the oligonucleotide may impart preferred properties to the agent. For example, preferred locations of certain modifications may confer optimal gene silencing properties or increased resistance to endonuclease or exonuclease activity.

One or more nucleotides within an oligonucleotide may have a 2'-5' junction. One or more nucleotides in an oligonucleotide may have an inverted connection,

- 37 046694 example connection 3'-3', 5'-5', 2'-2' or 2'-3'.

Double-stranded oligonucleotides may include at least one dinucleotide 5'-uridine adenine-3' (5'-UA-3'), where uridine is a 2'-modified nucleotide, or a terminal dinucleotide 5'-uridine-guanine-3' (5 '-UG-3'), with 5'-uridine being a 2'-modified nucleotide, or terminal dinucleotide 5'-cytidine-adenine-3' (5'-CA-3'), with 5'-cytidine being 2'-modified nucleotide, or terminal dinucleotide 5'-uridine-uridine-3' (5'-UU-3'), with 5'-uridine being a 2'-modified nucleotide, or terminal dinucleotide 5'-cytidine-cytidine3 ' (5'-CC-3'), wherein 5'-cytidine is a 2'-modified nucleotide, or terminal dinucleotide 5'-cytidine-uridine-3' (5'-CU-3'), wherein 5' -cytidine is a 2'-modified nucleotide, or terminal dinucleotide 5'-uridine-cytidine-3' (5'-UC-3'), with 5'-uridine being a 2'-modified nucleotide. Double-stranded oligonucleotides containing these modifications are particularly stable to endonuclease activity.

General links.

Oligoribonucleotides and oligoribonucleosides used in accordance with the present invention can be synthesized using solid phase synthesis, see, for example, Oligonucleotide synthesis, a practical approach, Ed. M.J. Gait, IRL Press, 1984; Oligonucleotides and Analogues, A Practical Approach, Ed. F. Eckstein, IRL Press, 1991 (especially Chapter 1, Modern machine-aided methods of oligodeoxyribonucleotide synthesis, Chapter 2, Oligoribonucleotide synthesis, Chapter 3, 2'-O-Methyloligoribonucleotides: synthesis and applications, Chapter 4, Phosphorothioate oligonucleotides, Chapter 5 , Synthesis of oligonucleotide phosphorodithioates, Chapter 6, Synthesis of oligo-2'-deoxyribonucleoside methylphosphonates, and Chapter 7, Oligodeoxynucleotides containing modified bases. Other particularly useful synthetic routes, reagents, blocking groups, and reaction conditions are described in Martin, P., Helv. Chim. Acta, 1995, 78, 486-504; or works referred to in these publications. Modifications described in WO 00/44895, WO 01/75164 or WO 02/44321 may apply in this document. Disclosure of all publications, patents and published patent applications listed in this document. document, included in the description as references.

Links to phosphate groups.

The preparation of phosphinate oligoribonucleotides is described in US Patent No. 5508270. The preparation of alkylphosphonate oligoribonucleotides is described in US Patent No. 4469863. The preparation of phosphoramidite oligoribonucleotides is described in US Patent No. 5256775 or 5366878. The preparation of phosphotriester oligoribonucleotides is described in US Patent No. 50 23243. The preparation of boranophosphate oligoribonucleotides is described in US Pat. No. 5,130,302 and 5,177,198. The preparation of 3'-deoxy-3'-amino phosphoramidate oligoribonucleotides is described in US Pat. No. 5,476,925. 3'-Deoxy-3'-methylenephosphonate oligoribonucleotides are described in An, H., et al., J. Org. Chem. 2001, 66, 2789-2801. The preparation of sulfur-bridged nucleotides is described in Sproat et al. Nucleosides Nucleotides 1988, 7, 651 and Crosstick et al. Tetrahedron Lett. 1989, 30, 4693.

Links to sugar groups.

Modifications related to 2' modifications can be found in Verma, S. et al. Annu. Rev. Biochem. 1998, 67, 99-134 and references given in this work. Specific modifications of ribose can be found in the following works: 2'-fluoro- (Kawasaki et. al., J. Med. Chem., 1993, 36, 831-841), 2'-MOE (Martin, P. Helv. Chim. Acta, 1996, 79, 1930-1938), LNA (Wengel, J. Acc. Chem. Res. 1999, 32, 301-310).

Phosphate Group Substitution Links.

Methylenemethylimino-crosslinked oligoribonucleosides, also referred to herein as MMI-linked oligoribonucleosides, methylenedimethylhydrazo-crosslinked oligoribonucleosides, also referred to herein as MDH-linked oligoribonucleosides, and methylenecarbonylamine-linked oligonucleosides, also referred to herein as amido-3-linked oligoribonucleosides, and methyleneaminocarbonyl-crosslinked Oligonucleosides, also referred to herein as amido-4-cross-linked oligoribonucleosides, as well as mixed-backbone compounds containing, for example, alternating MMI and PO or PS linkages, can be prepared as described in US Pat. Nos. 5,378,825, 5,386,023, 5,489,677, and PCT published patent applications PCT/US92/04294 and PCT/US92/04305 (published as WO 92/20822 and WO 92/20823 respectively). Formoacetal and thioformacetal cross-linked oligoribonucleosides can be prepared as described in US Patent Nos. 5,264,562 and 5,264,564. Ethylene oxide cross-linked oligoribonucleosides can be prepared as described in US Patent No. 5,223,618. Substitutions to siloxane are described in Cormier, J.F. et al., Nucleic Acids Res. 1988, 16, 4583. Carbonate substitutions are described in Tittensor, J.R. J. Chem. Soc. p. 1971, 1933. Substitutions to carboxymethyl are described in Edge, M.D. et al., J. Chem. Soc. Perkin Trans. 1, 1972, 1991. Carbamate substitutions are described in Stirchak, E.P. Nucleic Acids Res. 1989, 17, 6129.

Phosphate-ribose backbone replacement links.

Cyclobutyl sweetener compounds can be prepared as described in U.S. Patent No. 5,359,044. Pyrrolidine sweetener can be prepared as described in U.S. Patent No. 5,519,134. Morpholine sweetener can be prepared as described in U.S. Patents

- 38 046694 No. 5142047 and 5235033, and in the disclosures of other related patents. Peptide Nucleic Acids (PNAs) are known as such and can be prepared according to any of a variety of methods outlined in Peptide Nucleic Acids (PNA): Synthesis, Properties and Potential Applications, Bioorganic & Medicinal Chemistry, 1996, 4, 5-23 . They may also be obtained in accordance with US Patent No. 5,539,083.

Links to modifications to the ends of molecules.

Modifications to the ends of the molecules are described in Manoharan, M. et al. Antisense and Nucleic Acid Drug Development, 12, 103-128 (2002) and references cited in this work.

Nucleobase Links.

N-2-Substituted purine nucleoside amidites can be prepared as described in US Pat. No. 5,459,255. 3-Deazapurine nucleoside amidites can be prepared as described in US Pat. No. 5,457,191. 5,6-Substituted pyrimidine nucleoside amidites can be prepared as described in US Pat. No. 5,614,617. 5-Propynyl pyrimidine nucleoside amidites can be prepared as described in US Pat. No. 5,484,908.

Linkers.

The term linker means an organic moiety that connects two parts of a compound. Typically, linkers contain a direct bond or atom such as an oxygen atom or a sulfur atom, a structural unit such as NR ¹ , C(O), C(O)NH, SO, SO2, SO2NH, or a chain of atoms such as a substituted or unsubstituted alkyl, substituted or unsubstituted alkenyl, substituted or unsubstituted akinyl, arylalkyl, arylalkenyl, arylalkynyl, heteroarylalkyl, heteroarylalkyl, heteroarylalkynyl, heterocyclylalkenyl, heterocyclylalkenyl, heterocyclylalkynyl, aryl, heteroaryl, heterocyclyl, cycloalkenyl, cycloalkenyl, kyl, alkylarylalkenyl, alkylarylalkynyl, alkenylarylalkyl, alkenylarylalkenyl , alkenylarylalkynyl, alkynylarylalkyl, alkynylarylalkenyl, alkynylarylalkynyl, alkylheteroarylalkyl, alkylheteroarylalkynyl, alkylheteroarylalkynyl, alkenylheteroarylalkyl, alkenylheteroarylalkynyl, alkenylheteroarylalkynyl, alkynylheteroarylalkyl, alkynylheteroaryl alkenyl, alkynylheteroarylalkynyl, alkylheterocyclylalkyl, alkylheterocyclylalkenyl, alkylheterocyclylalkenyl, alkenylheterocyclylalkyl, alkenylheterocyclylalkenyl, alkenylheterocyclylalkynyl, alkynylheterocyclylalkyl, alkynylheterocyclylalkenyl, alkynylheterocyclylalkynyl, alkylaryl , alkenyl aryl, alkynylaryl , alkylheteroaryl, alkenylheteroaryl, alkynylheteroaryl, where one or more methylene residues may be interrupted or terminated by an O, S, S(O), SO2, N(R ¹ )2, C(O) atom or group, cleavable by a linker group, substituted or unsubstituted aryl, substituted or unsubstituted heteroaryl, substituted or unsubstituted heterocycle; where R ¹ is a hydrogen atom, acyl, aliphatic or substituted aliphatic group.

In one embodiment, the linker is

-[(P-Q-R)q-X-(P'-Q'-R')q]q-T-, where P, R, T, P', R' and T are each independently absent for each case, is CO, NH, O, S,

OS(O), NHC(O), CH2, CH2NH, CH2O; NHCH(R ^a )C(O), -C(O)-CH(R ^a )-, NH-, CH=NO, o ,s—s.

A or heterocyclyl;

,s—s, ,s—s.

Q and Q' are each independently absent for each case, is -(CH ₂ ) _n -, -C^X^XCH2H-, -(CH2)nC(R1)(R ² )-, -(CH2CH2OXCH2CH2-, or -( CH2CH2OXCH2CH2NH-;

X - absent or a cleavable linker group;

R ^a is H or a side chain of an amino acid;

R ¹ and R ² are each independently for each case H, CH3, OH, SH or N(R ⁿ )2;

R ⁿ is independently for each case H, methyl, ethyl, propyl, isopropyl, butyl or benzyl;

q, q' and q are each independently equal to 0-20 in each case, and the repeating unit may be the same or different;

n independently for each case is equal to 1-20;

m independently for each case is equal to 0-50.

In one embodiment, the linker contains at least one cleavable linker group.

In some embodiments, the linker is a branched linker. The branch point of a branched linker may be at least trivalent, but may be a tetravalent, pentavalent or hexavalent atom or group providing such multiple valencies. In some embodiments, the branch point is -N, -N(Q)-C, -O-C, -S-C, -SS-C, -C(O)N(Q)-C, -OC(O)N(Q )-C, -N(Q)C(O)-C or -N(Q)C(O)O-C; where Q is independently in each case H or optionally substituted alkyl. In another embodiment, the branch point is glycerol or a glycerol derivative.

Cleavable linker groups.

A cleavable linker group is a group that is substantially stable outside the cell,

- 39 046694 but when it enters the target cell, it splits, releasing two parts that the linker holds together. In a preferred embodiment, the cleavable linker group is cleaved at least 10 times or more, preferably at least 100 times faster in the target cell or under a first standard condition (which may, for example, be selected to simulate or sample intracellular conditions) than in the subject's blood, or under a second standard condition (which may, for example, be selected to simulate or sample conditions found in blood or serum).

Cleavable linker groups are sensitive to cleavage agents, such as pH, redox potential, or the presence of degradation molecules. In general, degrading agents are either more prevalent or found at higher levels or activities within the cell than in serum or blood. Examples of such degrading agents include: redox agents that are selected for specific substrates or that do not have substrate specificity, including, for example, oxidative or reductive enzymes or reducing agents such as mercaptans present in the cell that can degrade redox-sensitive a cleavable linker group by reduction; esterase; endosomes or agents capable of creating acidic environmental conditions, such as those resulting in a pH of five or lower; enzymes capable of hydrolyzing or degrading a linker group cleaved under acidic conditions by the mechanism of general acid catalysis, peptidases (which can be substrate-specific) and phosphatases.

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

The linker may include a cleavable linker group that is cleaved by a particular enzyme. The type of cleavable linker group incorporated into the linker may depend on the target cell. For example, liver cell-targeting ligands can be attached to cationic lipids through a linker that contains an ester group. Liver cells contain a lot of esterases, for this reason the linker will be degraded more efficiently in liver cells than in cell types that are not rich in esterases. Other cell types rich in esterases include cells of the lung, renal cortex, and testes.

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

In general, the suitability of a candidate cleavable linker group can be assessed by testing the ability of a degradant agent (or condition) to cleave the candidate linker group. It is also desirable to further test the candidate cleavable linker group for its ability to resist cleavage in the blood or upon contact with other non-target tissues. Thus, the relative sensitivities to cleavage under the first and second conditions can be determined, with the first chosen to model cleavage in the target cell and the second chosen to model cleavage in other tissues or biological fluids, such as blood or serum. The assessment can be carried out in cell-free systems, in cells, in cell culture, in organ or tissue culture, or in the whole animal body. It may be useful to conduct an initial assessment in a cell-free system or culture and confirm this by further assessment in the whole animal. In preferred embodiments, the useful candidate compounds are degraded at least 2-, 4-, 10-, or 100-fold faster in a cell (or in vitro conditions selected to simulate intracellular conditions) compared to blood or serum (or in vitro conditions). , selected so that they simulate the conditions of the extracellular environment).

Redox-sensitive cleavable linker groups.

One class of cleavable linker groups is represented by redox-sensitive cleavable linker groups, which are cleaved upon reduction or oxidation. An example of a reduction-cleavable linker group is a disulfide linker group (-S-S-). To determine whether a candidate cleavable linker group is a suitable reduction cleavable linker group or, for example, suitable for use with a particular interfering RNA moiety and a particular targeting agent, one can rely on the methods described herein. For example, a candidate moiety can be assessed by incubation with dithiothreitol (DTT) or another reducing agent using reagents known to one skilled in the art, which reproduces the rate of degradation that would be observed in a cell, such as a target cell. candidate

- 40 046694 group can also be assessed under conditions that are selected to simulate conditions in blood or serum. In a preferred embodiment, the candidate compounds are degraded in the blood by no more than 10%. In preferred embodiments, the useful candidate compounds degrade at least 2, 4, 10, or 100 times faster in a cell (or in vitro conditions selected to simulate intracellular conditions) compared to blood (or in vitro conditions selected so that they simulate the conditions of the extracellular environment). The rate of degradation of candidate compounds can be determined using standard enzyme kinetics analysis under conditions selected to simulate the intracellular environment and in comparison with conditions selected to simulate the extracellular environment.

Phosphate-based cleavable linker groups.

Phosphate-based cleavable linker groups 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 cellular phosphatases. Examples of phosphate-based linker groups are -O-P(O)(ORk)-O-, -O-P(S)(ORk)-O-, -O-P(S)(SRk)-O-, -S-P(O)(ORk )-O-, -O-P(O)(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(O)(Rk)-O-, -S-P(S)(Rk)-O- , -S-P(O)(Rk)-S-, -O-P(S)(Rk)-S-. Preferred embodiments are

-O-P(O)(OH)-O-, -O-P(S)(OH)-O-, -O-P(S)(SH)-O-, -S-P(O)(OH)-O-, -O-P (O)(OH)-S-, -S-P(O)(OH)-S-, -O-P(S)(OH)-S-, -S-P(S)(OH)-O-, -O-P(O )(H)-O-, -O-P(S)(H)-O-, -S-P(O)(H)-O-, -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 candidate compounds can be assessed using methods similar to those described above.

Linker groups that are cleavable by acid.

Acid-cleavable linker groups are linker groups that are cleaved under acidic conditions. In preferred embodiments, the acid-cleavable linker groups are cleaved in an acidic environment with a pH value of about 6.5 or lower (for example, about 6.0, 5.5, 5.0 or lower) or by the action of agents, such as enzymes, that may act by a general acid mechanism. In the cell, specific low-pH organelles such as endosomes and lysosomes may provide a cleavage-promoting environment for acid-cleavable linker groups. Non-limiting examples of acid-cleavable linker groups include hydrazones, esters, and amino acid esters. Acid-cleavable groups may have the general formula -C=NN-, C(O)O or -OC(O). In a preferred embodiment, the carbon atom attached to the oxygen atom of the ester (alkoxy group) belongs to an aryl group, a substituted alkyl group or a tertiary alkyl group such as dimethylpentyl or tert-butyl. These candidate compounds can be assessed using methods similar to those described above.

Ester-based linker groups.

Ester linker groups that are subject to cleavage are cleaved in cells by enzymes such as esterases and amidases. Non-limiting examples of cleavable ester linker groups include esters of alkylene, alkenylene and alkynylene groups. Cleavable ester linker groups have the general formula -C(O)O- or -OC(O)-. These candidate compounds can be assessed using methods similar to those described above.

Peptide-based cleavable groups.

Peptide-based cleavable linker groups are cleaved in cells by enzymes such as peptidases and proteases. Peptide-based cleavable linker groups are peptide bonds formed between amino acids to form oligopeptides (eg, dipeptides, tripeptides, etc.) and polypeptides. Peptide-based cleavable groups do not include an amido group (-C(O)NH-). The amido group can be formed between any alkylene, alkenylene or alkynene. A peptide bond is a special type of amide bond formed between amino acids to form peptides and proteins. A peptide-based 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 a fully functional amido group. The cleavable peptide-based linker groups have 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 candidate compounds can be assessed using methods similar to those described above.

Ligands.

A wide variety of molecules can be attached to the oligonucleotides and lipids of the present invention. Preferred residues are ligands that are attached, preferably covalently, either directly or indirectly through an insertion element.

In preferred embodiments, the ligand modifies the distribution, direction of action, or lifetime of the molecule into which it is incorporated. In preferred embodiments, the ligand provides increased affinity for a selected target, e.g., a molecule, a cell or cell type, a compartment, e.g., a cellular or organ compartment, a tissue, organ, or region of the body,

- 41 046694 for example, when compared with a form lacking such a ligand. Ligands that provide increased affinity for a selected target are also called targeting ligands. Preferred ligands for conjugation with lipids of the present invention are targeted ligands.

Some ligands may have endosomolytic properties. Endosomolytic ligands promote lysis of the endosome and/or transport of the composition according to the invention, or its components, from the endosome into the cytoplasm of the cell. The endosomolytic ligand may be a polyanionic peptide or a peptidomimetic exhibiting pH-dependent membrane activity and fusogenicity. In some embodiments, the endosomolytic ligand acquires its active conformation at the endosomal pH. An active conformation is a conformation in which the endosomolytic ligand ensures lysis of the endosome and/or transport of the composition according to the invention, or its components, from the endosome into the cytoplasm of the cell. Examples of endosomolytic ligands include GALA peptide (Subbarao et al., Biochemistry, 1987, 26:2964-2972), EALA peptide (Vogel et al., J. Am. Chem. Soc., 1996, 118:1581-1586) and their derivatives (Turk et al., Biochem. Biophys. Acta, 2002, 1559:56-68). In some embodiments, the endosomolytic moiety may contain a chemical group (eg, an amino acid) that will undergo a charge change or protonation in response to a change in pH. The endosomolytic component can be linear or branched. Examples of primary sequences of peptide-based endosomolytic ligands are shown in Table. 4.

Table 4

List of peptides with endosomolytic activity

Name Sequence (N- to C-terminus) Link GALA AALEALAEALEALAEALEALAEAAAAGGC 1 EALA AALAEALAEALAEALAEALAEALAAAAGGC 2 ALEALAEALEALAEA 3 INF-7 GLFEAIEGFIENGWEGMIWDYG 4 Inf NA-2 GLFGAIAGFIENGWEGMIDGWYG 5 diINF-7 GLF EAI EGFI ENGW EGMI DGWYGC GLF EAI EGFI ENGW EGMI DGWYGC 5 diINF3 GLF EAI EGFI ENGW EGMI DGGC GLF EAI EGFI ENGW EGMI DGGC 6 GLF GLFGALAEALAEALAEHLAEALAEALEALAAGGSC 6 GALA-INF3 GLFEAIEGFIENGWEGLAEALAEALEALAAGGSC 6 INF-5 GLF EAI EGFI ENGW EGnl DG K GLF EAI EGFI ENGW EGnl DG 4

n, norleucine.

Links.

1. Subbarao et al., Biochemistry, 1987, 26: 2964-2972.

2. Vogel et al., J. Am. Chem. Soc., 1996, 118: 1581 -1586

3. Turk, M. J., Reddy, J. A. et al. (2002). Characterization of a novel pH-sensitive peptide that enhances drug release from folate-targeted liposomes at endosomal pHs. Biochim. Biophys. Acta 1559, 56-68.

4. Plank, C. Oberhauser, B. Mechtler, K. Koch, C. Wagner, E. (1994). The influence of endosome-disruptive peptides on gene transfer using synthetic virus-like gene transfer systems, J. Biol. Chem. 269 12918-12924.

5. Mastrobattista, E., Koning, G. A. et al. (2002). Functional characterization of an endosome-disruptive peptide and its application in cytosolic delivery of immuno liposome-entrapped proteins. J Biol. Chem. 277, 27135-43.

6. Oberhauser, B., Plank, C. et al. (1995). Enhancing endosomal exit of nucleic acids using pH-sensitive viral fusion peptides. Deliv. Strategies Antisense Oligonucleotide Ther. 247-66.

Preferred ligands may improve properties associated with transport, hybridization, and specificity, and may also improve the nuclease resistance of the resulting natural or modified oligoribonucleotide or polymer molecule containing any combination of monomers described herein and/or natural or modified ribonucleotides.

In general, ligands may contain therapeutic modifiers, for example to enhance absorption; diagnostic compounds or reporter groups, for example for distribution monitoring; cross-linking agents and nuclease resistance-conferring residues. Major examples include lipids, steroids, vitamins, sugars, proteins, peptides, polyamines, and peptide mimetics.

Ligands may include a naturally occurring substance such as a protein (for example, sulfur

- 42 046694 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, such as a synthetic polyamino acid or oligonucleotide (eg, an aptamer). Examples of polyamino acids include polyamino acid polylysine (PLL), poly-L-aspartic acid, poly-L-glutamic acid, styrene-maleic anhydride copolymer, poly-(L-lactide-co-glycolic) copolymer, divinyl ether-maleic anhydride copolymer, N(2-hydroxypropyl)methacrylamide copolymer (HMPA), polyethylene glycol (PEG), polyvinyl alcohol (PVA), polyurethane, poly(2-ethyl acrylic acid), N-isopropylacrylamide polymers, or polyphosphazine. Examples of polyamines include polyethylenimine, polylysine (PLL), spermine, spermidine, polyamine, pseudopeptide polyamine, peptidomimetic polyamine, dendrimeric polyamine, arginine, amidine, protamine, cationic lipid, cationic porphyrin, polyamine quaternary salt, or α-helical peptide.

Ligands may also include targeting moieties, such as cell or tissue targeting agents, such as a lectin, glycoprotein, lipid, or protein, such as an antibody, that binds to a specific cell type, such as kidney cells. The targeted group can be thyrotropin, melanotropin, lectin, glycoprotein, surfactant protein A, mucin carbohydrate, multivalent lactose, multivalent galactose, N-acetyl-galactosamine, N-acetyl-glucosamine, multivalent mannose, multivalent fucose, glycosylated polyamino acids, multivalent galactose, transferrin, bisphosphonate, polyglutamate, polyaspartate, lipid, cholesterol, steroid, bile acid, folate, vitamin B12, biotin, RGD peptide, RGD peptide mimetic or aptamer. In table Figure 5 shows some examples of targeting ligands and their associated receptors.

- 43 046694

Table 5

Targeting ligands and associated receptors

Liver cells Ligand Receptor 1) Parenchyma cell (PC) (hepatocytes) Galactose ASGP-R (receptor asiologlycoproteins) Gal NAc (N-acetyl-galactosamine) ASPG-R Gal NAc receptor Lactose Asialofetuin ASPG-r 2) Sinusoidal endothelial cell (SEC) Hyaluronic acid Hyaluronic acid receptor Pro-collagen Pro-collagen receptor Negatively charged molecules Phagocytic receptors Mannose Mannose receptors N-acetyl-glucosamine Phagocytic receptors Immunoglobulins Fc receptor Lipopolysaccharide CD 14 receptor Insulin Receptor-mediated transcytosis Transferrin Receptor-mediated transcytosis Albumin Non-specific Conjugates of sugars with albumins Mannose 6-phosphate Mannose-6-receptor phosphate 3) Kupffer cell (KC) Mannose Mannose receptors Fucose Fucose receptors Albumin Non-specific Mannose conjugates albumins

Other examples of ligands include dyes, intercalating agents (eg, acridines), cross-linkers (eg, psoralen, mitomycin C), porphyrins (TPPC4, texaphyrin, sapphirin), polycyclic aromatic hydrocarbons (eg, phenazine, dihydrophenazine), non-natural endonucleases (eg , EDTA), lipophilic molecules, for example cholesterol, cholic acid, adamantane acetic acid, 1-pyrene-butyric acid, dihydrotestosterone, 1,3-bis-O(hexadecyl)glycerol, geranyloxyhexyl group, hexadecylglycerol, borneol, menthol, 1, 3-propanediol, heptadecyl group, palmitic acid, myristic acid, O3-(oleoyl)lithocholic acid, O3-(oleoyl)cholic acid, dimethoxytrityl or phenoxazine peptide conjugates (e.g. antennapedia peptide, Tat peptide), alkylating agents, phosphate , amino-, mercapto-, PEG (e.g. PEG-40K), MPEG, [MPEG]2, polyamino-, alkyl, substituted alkyl, radiolabeled markers, enzymes, haptens (e.g. biotin), transport/absorption compounds (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 may be proteins, such as glycoproteins, or peptides, such as molecules having a specific affinity for a co-ligand, or antibodies, such as an antibody that binds

- 44 046694 with a specific cell type, such as a cancer cell, endothelial cell or bone cell. Ligands may also include hormones and hormone receptors. They may also include non-peptide molecules such as lipids, lectins, carbohydrates, vitamins, cofactors, multivalent lactose, multivalent galactose, N-acetyl-galactosamine, N-acetyl-glucosamine, multivalent mannose, multivalent fucose or aptamers. The ligand may be, for example, a lipopolysaccharide, a p38 MAP kinase activator, or an NF-kB activator.

The ligand may be a substance, eg a drug, that can enhance the uptake of the RNA1 agent into the cell, eg by disrupting the cell's cytoskeleton, eg by disrupting cellular microtubules, microfilaments and/or intermediate filaments. The drug substance may be, for example, taxon, vincristine, vinblastine, cytochalasine, nocodazole, jasplakinolide, latrunculin A, phalloidin, swincholide A, indanocin or myoservin.

The ligand may enhance the uptake of the RNA1 agent into the cell by, for example, activating an inflammatory response. Examples of ligands that could have such an effect include tumor necrosis factor α (TNF-α), interleukin-1b or interferon-γ.

In one embodiment of the present invention, the ligand is a lipid or a lipid-based molecule. Such a lipid or lipid-based molecule preferentially binds to a serum protein, such as human serum albumin (HSA). The HSA-binding ligand allows distribution of the conjugate to a target tissue, for example, a non-kidney target tissue of the body. For example, the target tissue may be liver tissue, including liver parenchyma cells. Other molecules that bind to HSA can also be used as ligands. For example, Neproxin or aspirin may be used. The 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 regulate binding to a serum protein, such as HSA.

The lipid-based ligand can be used for modulation, for example to control the binding of the conjugate to the target tissue. For example, a lipid or lipid-based ligand that binds to HSA more strongly 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 significant affinity such that the conjugate will preferentially distribute to non-renal tissues. However, it is preferable that the affinity is not so strong that the binding of HSA to the ligand is irreversible.

In another preferred embodiment, the lipid-based ligand binds weakly or not at all to HSA such that the conjugate will preferentially distribute to the kidneys. Other renal cell-targeting moieties may also be used instead of or in conjunction with a lipid-based ligand.

Another feature is that 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 useful for the treatment of disorders characterized by unwanted proliferation of cells, for example of a malignant or non-malignant type, such as tumor cells. Examples of vitamins include vitamins A, E, and K. Other examples of vitamins include B vitamins, such as folic acid, B12, riboflavin, biotin, pyridoxal, or other vitamins or nutrients taken up by tumor cells. This also includes HAS, low-density lipoprotein (LDL) and high-density lipoprotein (HDL).

Another feature is that the ligand is a cell penetrating agent, preferably a helical cell penetrating agent. Preferably the agent is amphipathic. An example of an agent is a peptide such as tat or antennopedia. If the agent is a peptide, it may be modified, including by peptidyl mimetic, invertomers, non-peptide or pseudo-peptide linkers, or the use of D-amino acids. The helical agent is preferably an α-helical agent, which preferably has a lipophilic and lipophobic phase.

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 that of a natural peptide. The peptide or peptidomimetic residue may be about 5-50 amino acids in length, such as about 5, 10, 15, 20, 25, 30, 35, 40, 45, or 50 amino acids in length (for example, see Table 6).

- 45 046694

Table 6

Examples of cell-penetrating peptides

Penetrating into cell peptide Amino acid sequence Link Penetratin RQIKIWFQNRRMKWKK Derossi et al., J. Biol. Chern. 269:10444, 1994 Tat fragment (48-60) GRKKRRQRRRPPQC Vives et al., J. Biol. Chern., 272:16010, 1997 Signal sequence-based peptide GALFLGWLG A AGS TMG A WS QPKKKR KV Chaloin et al., Biochem. Biophys. Res. Commun, 243:601, 1998 PVEC LLIILRRRIRKQ AN AN S K Elmquist et al., Exp. Cell Res., 269:237, 2001 Transportan GWTLNSAGYLLKINLKALAAALAKKIL Pooga et al., FASEB J., 12:67, 1998 Amphiphilic model peptide KLALKLALKALKAALKLA Oehlke et al., Mol. Then, 2:339,2000 Argg RRRRRRRRR Mitchell et al., J. Pept. Res., 56:318, 2000 Penetrating the bacterial cell wall KFFKFFKFFK LL-37 LLGDFFRKS KEKIGKEFKRIVQRIKDFL RNLVPRTES Cecropin P1 SWLSKTAKKLENSAKKRISEGIAIAIQGGPR a-defensin ACYCRIPACIAGERRYGTCIYQGRLWA FCC β-defensin DHYNCVSSGGQCLYSACPIFTKIQGTC YRGKAKCCK Bactenecin RKCRIVVIRVCR PR-39 RRRPRPPYLPRPRPPPFFPPRLPPRIPPGF PPRFPPRFPGKR-NH ₂ Indolicidin ILPWKWPWWPWRR-NH2

The peptide or peptidomimetic may be, for example, a cell penetrating peptide, a cationic peptide, an amphipathic peptide, or a hydrophobic peptide (eg, consisting primarily of Tyr, Trp, or Phe). The peptide residue may be a dendrimeric peptide, a peptide with limited conformational freedom, or a cross-linked peptide. In another embodiment, the peptide residue may include a hydrophobic membrane translocation sequence (MTS). An example of a peptide containing a hydrophobic MTS is RFGF, which has the amino acid sequence AAVALLPAVLLALLAP.

An RFGF analog (eg, the amino acid sequence AALLPVLLAAP) containing a hydrophobic MTS may also be a targeting moiety. The peptide moiety may be a delivery peptide that transports large polar molecules, including peptides, oligonucleotides and proteins, across cell membranes.

For example, sequences from the HIV Tat protein (GRKKRRQRRRPPQ) and the Drosophila Antennapedia protein (RQIKIWFQNRRMKWKK) have been found to function as delivery peptides. A peptide or peptidomimetic may be encoded by a random DNA sequence, such as a peptide identified in a phage display library or a one-bead-one-compound (SOC) combinatorial library (Lam et al., Nature, 354:82-84, 1991 ). Preferably, the peptide or peptidomimetic attached to the RNA1 agent via an inserted monomer unit is a targeting peptide such as an arginine glycine aspartic acid (RGD) peptide or an RGD mimetic. The peptide residue may have a length ranging from about 5 to about 40 amino acids. Peptide residues may have a structural modification such that it increases stability or affects conformational properties. Any of the structural modifications described below may be used.

The RGD peptide moiety can be used to target a tumor cell, such as an endothelial tumor cell or a breast cancer cell (Zitzmann et al., Cancer

- 46 046694

Res., 62:5139-43, 2002). The RGD peptide can facilitate targeting of an agent involved in PHKi to tumors of a variety of other tissues, including lung, kidney, spleen or liver (Aoki et al., Cancer Gene Therapy, 8:783-787, 2001). Preferably, the RGD peptide will facilitate the targeting of the agent involved in RNA|. on the kidney. The RGD peptide may be linear or cyclic and may be modified, such as glycosylated or methylated, to facilitate targeting of specific tissues. For example, a glycosylated RGD peptide can deliver an agent involved in PHKi to a tumor cell expressing ανβ3 (Haubner et al., Jour. Nucl. Med., 42:326-336, 2001).

Peptides can be used that target markers that are enriched in proliferating cells. For example, RGD-containing peptides and peptidomimetics can be targeted to cancer cells, in particular cells that exhibit ανβ3 integrin. Thus, RGD peptides, cyclic peptides containing RGD, RGD peptides containing D-amino acids, as well as synthetic RGD mimetics can be used. In addition to RGD, other residues that target the ανβ3 integrin ligand can be used. In general, such ligands can be used to control proliferating cells and angiogenesis. Preferred ligand conjugates of this type target PECAM-1, VEGF or other tumor genes, such as the tumor genes described herein.

The cell penetrating peptide is capable of penetrating into a cell, such as a microbial cell, such as a bacterial or fungal cell, or a mammalian cell, such as a human cell. The bacterial cell penetrating peptide may be, for example, an α-helical linear peptide (for example, LL-37 or Ceropine PI), a disulfide bond-containing peptide (for example, α-defensin, β-defensin or bactenecin), or a peptide containing only one or two dominant amino acids (eg PR-39 or indolicidin). The cell-penetrating peptide may also contain a nuclear localization signal (NLS). For example, the cell-penetrating peptide may be a bipartite amphipathic peptide, such as MPG, which is produced by fusing the peptide domain of HIV-1 gp41 and the NLS of the large T antigenavirus SV40 (Simeoni et al., Nucl. Acids Res. 31:2717-2724, 2003 ).

In one embodiment, the targeting peptide attached to the PHKi agent and/or carrier oligomer may be an amphipathic α-helical peptide. Non-limiting examples of amphipathic α-helical peptides include cecropins, lycotoxins, paradaxins, buforin, CPF, bombinin-like peptide (BLP), cathelicidins, ceratotoxins, S. clava peptides, hagfish intestinal antimicrobial peptides (HFIAP), magainins, brevinin-2, dermaseptins , melittins, pleurocidin, H2A peptides, Xenopus peptides, esculentinis-1 and caerins. A number of factors must preferably be taken into account to maintain the integrity of a stable helix. For example, a maximum number of helix-stabilizing residues (eg, Leu, Ala or Lys) and a minimum number of helix-destabilizing residues (eg, proline or cyclic monomer units) will be used. A capping residue should be considered (eg, Gly is an example of an N-capping residue) and/or C-terminal amidation may be used to provide additional hydrogen bonding to stabilize the helix. The formation of salt bridges between residues with opposite charges separated by i±3 or i±4 positions can provide stability. For example, cationic residues such as lysine, arginine, homo-arginine, ornithine or histidine can form salt bridges with the anionic residues glutamate and aspartate.

Peptide and peptidomimetic ligands include those with naturally occurring or modified peptides, such as D- or L-peptides; α-, β- or ω-peptides; N-methyl peptides; azapeptides; peptides having one or more amido groups, i.e. peptides in which the bonds between the monomers are replaced by one or more bonds based on urea, thiourea, carbamate, or sulfonylurea; or cyclic peptides.

The targeting ligand can be any ligand capable of acting on a specific receptor. Examples are: folate, GalNAc, galactose, mannose, mannose-6-phosphate, sugar clusters such as the 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, endothelium, GCPII, somatostatin, LDL and HDL ligands. Ligands can also be nucleic acid based, such as an aptamer. The aptamer may be unmodified or carry any combination of modifications described herein.

Endosomal release agents include imidazoles, poly- or oligoimidazoles, PEI, peptides, fusogenic peptides, polycarboxylates, polycations, masked oligo- or poly-cations or anions, acetals, polyacetals, ketals/polyketals, orthoesters, polymers with or without masked cationic or anionic charges, dendrimers with masked or unmasked cationic or anionic charges.

PK modulator refers to a pharmacokinetic modulator. PK modulator includes lipophile

- 47 046694 nal molecules, bile acids, steroids, phospholipid analogues, peptides, protein binding agents, PEGs, vitamins, etc. Non-limiting examples of PK modulators include cholesterol, fatty acids, cholic acid, lithocholic acid, dialkylglycerides, diacylglyceride, phospholipids, sphingolipids, naproxen, ibuprofen, vitamin E, biotin, etc. It is known that oligonucleotides that contain multiple phosphorothioate linkages also bind to serum protein, so short oligonucleotides, such as oligonucleotides of about 5 bases, 10 bases, 15 bases, or 20 bases in length, containing multiple phosphorothioate linkages in the backbone are also suitable for the present invention in as ligands (for example, PK-modulating ligands).

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

Other ligands useful for the present invention are described in co-pending applications USSN: 10/916185, filed August 10, 2004; USSN: 10/946873, filed September 21, 2004; USSN: 10/833934, filed August 3, 2007; USSN: 11/115989, filed April 27, 2005, and USSN: 11/944227, filed November 21, 2007, which are incorporated in their entirety by reference.

When two or more ligands are present, the ligands may all have the same properties, all may have different properties, or some ligands may 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. In a preferred embodiment, all ligands have different properties.

Ligands can be attached to oligonucleotides at various positions, for example at the 3' end, 5' end and/or at positions within the sequence. In preferred embodiments, the ligand is attached to the oligonucleotides through an intermediate structure. A ligand, or a ligand attached via an intermediate structure, may be present on a monomer when said monomer is recruited into a growing chain. In some embodiments, a ligand may be recruited by combining with a precursor monomer after said precursor monomer has been recruited into the growing chain. As an example, a monomer having, for example, an amino-terminal intermediate connecting structure (ie, no associated ligand), for example, TAP-(CH ₂ )NNH ₂ , can be recruited into a growing sense or antisense strand. At a subsequent stage, i.e. once the precursor monomer is recruited into the chain, a ligand having an electrophilic group, such as a pentafluorophenyl ether or aldehyde group, can subsequently be attached to the precursor monomer by reacting the electrophilic group of the ligand with the terminal nucleophilic group of the connecting structure of the precursor monomer.

In the case of double-stranded oligonucleotides, ligands may be attached to one or both strands. In some embodiments, the double-stranded RNAt agent comprises a ligand conjugated to the sense strand. In other embodiments, the double-stranded RNAt agent comprises a ligand conjugated to an antisense strand.

In some embodiments, the ligand may be conjugated to nitrogenous bases, sugar residues, or internucleoside bonds of the nucleic acid molecule. Conjugation with purine nitrogenous bases or their derivatives can occur at any position, including endocyclic and exocyclic atoms. In some embodiments, the 2-, 6-, 7-, or 8-position of the purine nitrogenous base is connected to a conjugating moiety. Conjugation with pyrimidine nitrogenous bases or their derivatives can also occur at any position. In some embodiments, the 2-, 5-, and 6-positions of the pyrimidine nitrogenous base may be replaced by a conjugating moiety. Conjugation with sugar residues in the nucleoside can occur at any carbon atom. Examples of carbon atoms within a sugar moiety to which a conjugating moiety may be attached include the 2', 3' and 5' carbon atoms. The 1' position may also be connected to a conjugating residue, for example in the absence of a nitrogenous base. Internucleoside bonds can also carry conjugated residues. In the case of phosphorus-containing bonds (eg, phosphodiester, phosphorothioate, phosphorodithioate, phosphoramidate, etc.), the conjugating moiety may be attached directly to the phosphorus atom or to an O, N or S atom bonded to the phosphorus atom. In the case of amino- or amide-containing internucleoside linkages (eg, PNA), the conjugating moiety may be attached to the nitrogen atom of the amino or amido group or to an adjacent carbon atom.

There are many methods for preparing conjugates of oligomeric compounds. In general, an oligomeric compound is attached to a conjugating moiety by reacting a reactive group (eg, OH, SH, amino, carboxyl, aldehyde, etc.) on the oligomeric compound with a reactive group on the conjugating moiety. In some embodiments, one reactive group is electrophilic and the other is nucleophilic.

For example, the electrophilic group may be a carbonyl-containing functional group, and the nucleophilic group may be an amino or thiol group. Methods for the conjugation of nucleic acids and related oligomeric compounds with or without the participation of linker groups in detail

- 48 046694 are described in the literature, such as, for example, the work of Manoharan in Antisense Research and Applications, Crooke and LeBleu, eds., CRC Press, Boca Raton, Fla., 1993, Chapter 17, the contents of which are fully included in the description as links.

Examples of US patents that describe the preparation of oligonucleotide conjugates include, but are not limited to, US Patent Nos. 4828979, 4948882, 5218105, 5525465, 5541313, 5545730, 5552538, 5578717, 5580731, 84, 5109124, 5118802, 5138045, 5414077, 5486603,

5512439, 5578718, 5608046, 4587044, 4605735, 4667025, 4762779, 4789737, 3824941, 4835263, 4876335,

4904582, 4958013, 5082830, 5112963, 5214136, 5149782, 5214136, 245022, 5254469, 5258506, 5262536,

5272250, 5292873, 5317098, 5371241, 5391723, 5416203, 5451463, 5510475, 5512667, 5514785, 5565552,

5567810, 5574142, 5585481, 5587371, 5595726, 5597696, 5599923, 5599928, 5672662, 5688941, 5714166,

6153737, 6172208, 6300319, 6335434, 6335437, 6395437, 6444806, 6484308, 6525031, 6528631, 6559279, the contents of each of which are incorporated herein by reference.

Definitions

For convenience, the meanings of certain terms and phrases used in the specification, examples, and appended claims are set forth below. If there is an apparent conflict between the use of a term elsewhere in this specification and the description given in this section, the definition in this section shall control.

The symbols G, C, A and U each generally represent a nucleotide containing guanine, cytosine, adenine and uracil as a base, respectively. However, it is understood that the term ribonucleotide or nucleotide can also refer to a modified nucleotide, as described in detail below, or to a surrogate residue that replaces it. One skilled in the art will appreciate that guanine, cytosine, adenine, and uracil can be replaced by other residues without significantly changing the base-pairing properties of the oligonucleotide, including the nucleotide bearing such replacement residue. For example, in particular, a nucleotide containing inosine as a base can form a base pair with nucleotides containing adenine, cytosine or uracil. Therefore, nucleotides containing uracil, guanine or adenine can be replaced in the nucleotide sequence according to the invention by a nucleotide containing, for example, inosine. Sequences containing such replacement residues are embodiments of the present invention.

As used herein, Factor VII means Factor VII mRNA, protein, peptide, or polypeptide. As synonyms for the term factor VII, those skilled in the art are familiar with the terms AI132620, Cf7, factor VII precursor, coagulation factor VII, FVII, serum prothrombin conversion accelerator, coagulation protein FVII, and eptacog α.

As used herein, target sequence means the contiguous portion of the nucleotide sequence of an mRNA molecule produced during transcription of a gene, including mRNA that is the product of RNA processing of the primary transcription product.

As used herein, the term chain containing sequence means an oligonucleotide containing a chain of nucleotides arranged in accordance with a named sequence specified using standard nucleotide nomenclature.

As used in this document and unless otherwise specified, the term complementary when used in the context of a nucleotide pair means a classical Watson-Crick pair, i.e. GC, AT or AU. It also covers classical Watson-Crick pairing, in which one or both nucleotides have been modified as described herein, for example by modification of the ribose or modification of the phosphate backbone. It also involves pairing with inosine or another molecule, in which the base-pairing properties are not significantly changed.

As used herein and unless otherwise specified, the term complementary, when used to describe a first nucleotide sequence with respect to a second nucleotide sequence, refers to the ability of an oligonucleotide or polynucleotide containing the first nucleotide sequence to hybridize and form a duplex structure under certain conditions with the oligonucleotide or polynucleotide , containing a second nucleotide sequence, as understood by one skilled in the art. Complementarity may include complete complementarity, a substantial degree of complementarity, and a sufficient degree of complementarity to hybridize under physiological conditions, for example, under physiologically relevant conditions that may occur within an organism. Full complementarity means complementarity, as described above for a single pair, of all first and second sequence pairs. When a sequence is substantially complementary to a second sequence, as used herein, the two sequences may be completely complementary, or they may form one or more, but generally not more than 4, 3, or 2 pairs with base complementarity disrupted by hybridization, while maintaining the ability to hybridize under conditions more appropriate to their primary use. A significant degree of complementarity can also be defined as hybridization under stringent conditions, and stringent conditions may include: 400 mM NaCl, 40 mM PIPES pH 6.4, 1 mM EDTA, 50 or 70°C for 12-16 hours followed by washes . One skilled in the art will be able to determine a set of conditions

- 49 046694 most suitable for testing the complementarity of two sequences in accordance with the main application of the nucleotides being hybridized.

However, if two oligonucleotides are designed to form one or more single-stranded overhangs upon hybridization, such overhangs should not be considered as mismatched pairs for purposes of determining complementarity. For example, a double-stranded RNA containing one 21-nucleotide-long oligonucleotide and another 23-nucleotide-long oligonucleotide, wherein the longer oligonucleotide contains a 21-nucleotide-long sequence that is fully complementary to the shorter oligonucleotide, may still be considered fully complementary for purposes of the present invention.

Complementary sequences, when used herein, may also include or consist entirely of non-Watson-Crick base pairs and/or base pairs formed by non-natural and modified nucleotides, as long as the above requirements regarding their ability to hybridize are met.

The terms complementary, fully complementary, substantially complementary, and complementarity sufficient to hybridize under physiological conditions, such as physiologically relevant conditions that may occur within an organism, may be used herein to refer to base pairing between the sense strand and the antisense strand of double-stranded RNA. or between the antisense strand of a double-stranded RNA and the target sequence, as will be clear from the context of their use.

As used herein, a polynucleotide that is complementary, e.g., substantially complementary, to at least a portion of a messenger RNA (mRNA) molecule means a polynucleotide that is complementary, e.g., substantially complementary, to a contiguous portion of the mRNA in question (e.g., encoding factor VII) . For example, a polynucleotide is complementary to at least a portion of factor VII mRNA if the sequence is substantially complementary to an uninterrupted portion of the mRNA encoding factor VII.

The term double-stranded RNA or dsRNA, as used herein, means a ribonucleic acid molecule or complex of ribonucleic acid molecules having a duplex structure comprising two antiparallel and substantially complementary nucleic acid strands as described above. The two strands forming a duplex structure may be different parts of a single larger RNA molecule, or they may be separate RNA molecules. In case the two strands are parts of one larger molecule and for this reason are connected by a continuous chain of nucleotides between the 3' end of one strand and the 5' end of the corresponding second strand, forming a duplex structure, the connecting RNA strand is called a hairpin loop. When two strands are joined covalently in a manner other than a continuous chain of nucleotides between the 3' end of one strand and the 5' end of the corresponding second strand forming a duplex structure, the connecting structure is called a linker. RNA chains can contain the same or different numbers of nucleotides. The maximum number of base pairs is equal to the number of nucleotides in the shortest dsRNA strand. In addition to the duplex structure, dsRNA may contain one or more nucleotide overhangs. dsRNA, as used herein, is also referred to as small inhibitory RNA, 'sRNA, 'sRNA agent, RNA interference agent, or RNAI agent.

As used herein, nucleotide overhang refers to the unpaired nucleotide or nucleotides that protrude from the dsRNA duplex structure when the 3' end of one dsRNA strand extends beyond the 5' end of the other strand, or vice versa. A blunt or blunt end means that there are no unpaired nucleotides at the end of the dsRNA, i.e. missing nucleotide sticky end. Blunt-ended dsRNA is dsRNA that is double-stranded throughout its entire length, i.e. the nucleotide sticky end is absent at both ends of the molecule.

The term antisense strand means a strand of dsRNA that includes a region substantially complementary to the target sequence. As used herein, the term complementarity region means a region on an antisense strand that is substantially complementary to a sequence, such as a target sequence, as defined herein. If the region of complementarity is not completely complementary to the target sequence, mismatched pairs are most tolerated in the terminal regions and, if present, generally in the terminal region or regions, e.g., within 6, 5, 4, 3, or 2 nucleotides 5' - and/or 3'-end.

The term sense strand, as used herein, means a strand of dsRNA that includes a region substantially complementary to that of the antisense strand.

The term identity means the relationship between two or more polynucleotide sequences determined by comparison of these sequences. Identity also means the degree of sequence similarity between polynucleotide sequences, determined by strand alignment of such sequences. Although there are several ways to measure the identity between two polynucleotide sequences, the term is well

- 50 046694 is known to those skilled in the art (see, for example, Sequence Analysis in Molecular Biology, von Heinje, G., Academic Press (1987); and Sequence Analysis Primer, Gribskov., M. and Devereux, J., eds. , M. Stockton Press, New York (1991)). Substantially identical, as used herein, means that there is a very high degree of homology (preferably 100% sequence identity) between the sense strand of the dsRNA and the corresponding part of the target gene. However, the invention may employ dsRNAs having greater than 90 or 95% sequence identity, and therefore allow for sequence variations that might be expected to occur due to genetic mutations, chain polymorphisms, or evolutionary divergence. Although 100% identity is preferred, dsRNA may contain random single or multiple pairs with complementarity imbalance between the RNA and the target gene.

Cellularity, when referring to dsRNA, means facilitating uptake or absorption by the cell, as understood by one skilled in the art. Absorption or uptake of dsRNA can occur through processes of passive diffusion or active cellular transport or through the use of additional agents or devices. The meaning of this term is not limited to cells in an in vitro system; dsRNA can also be introduced into a cell if the cell is part of a living organism. In this case, the introduction into the cell will include delivery to the body. For example, for in vivo delivery, dsRNA can be injected into a tissue area or administered systemically. Introduction into cells in vitro includes methods known to those skilled in the art, such as electroporation and lipofection.

The terms suppress by silencing and inhibit expression, when referring to the factor VII gene, as used herein mean at least partial suppression of the expression of the factor VII gene, which is reflected by a decrease in the amount of mRNA generated from the factor VII gene that can be isolated from the first cell or a group of cells in which the factor VII gene is transcribed and which has been treated so that expression of the factor VII gene is inhibited, compared with a second cell or group of cells substantially identical to the first cell or group of cells but which has not been treated thus (control cells). The degree of inhibition is usually expressed as follows:

(mRNA in control cells) - (mRNA in treated cells) _______________________________________________________________ _f 100% (mRNA in control cells)

In another embodiment, the degree of inhibition may be expressed in terms of the reduction in an index that is operably associated with factor VII gene transcription, for example, the amount of protein encoded by the factor VII gene and secreted by the cell, or the number of cells exhibiting a particular phenotype, such as apoptosis. As such, factor VII gene silencing can be determined in any cell expressing the target, constitutively or through genetic engineering manipulation, and by any appropriate analytical method. However, if reference information is needed to determine whether a given dsRNA inhibits factor VII gene expression to a certain extent and is therefore a dsRNA of the present invention, the assay method shown in the examples below should serve as such reference information.

For example, in some cases, factor VII gene expression is suppressed by at least about 20, 25, 35, 40 or 50% by administration of a double-stranded oligonucleotide of the invention. In one embodiment, the Factor VII gene is suppressed by at least about 60, 70 or 80% by administration of a double-stranded oligonucleotide of the invention. In a more preferred embodiment, the factor VII gene is suppressed by at least about 85, 90 or 95% by administration of a double-stranded oligonucleotide of the invention.

The terms treat, cure and the like mean getting rid of a disease or disorder or mitigating its course. In the context of the present invention, to the extent that it relates to any of the other conditions listed herein below (for example, Factor VII-mediated conditions other than a thrombotic disorder), the terms treat, treat, and the like mean getting rid of or alleviating the at least one symptom associated with such a condition, or slowing or reversing the progression of such a condition.

A therapeutically significant composition may ameliorate a disease or disorder or a symptom of a disease or disorder when administered at an appropriate dosage.

As used herein, the term factor VII-mediated condition or disease and related terms and phrases means a condition or disease characterized by abnormal, eg, higher than normal, factor VII activity. Abnormal functional activity of factor VII may result from the expression of factor VII in cells that do not normally express factor VII, or from increased expression of factor VII (leading, for example, to the symptom of a viral hemorrhagic fever or blood clot formation). A factor VII-mediated condition or disease may be mediated in whole or in part by abnormal functional activity of factor VII. However, a factor VII-mediated condition or disease is one in which modulation of factor VII leads to a specific effect on the underlying condition or disease (on

- 51 046694 example, a factor VII inhibitor leads to a certain improvement in patient well-being, at least in some patients).

Hemorrhagic fever involves a combination of diseases caused by a viral infection. Fever and gastrointestinal symptoms are usually accompanied by capillary bleeding.

Coagulopathy is any disorder of a subject's blood clotting mechanism.

As used herein, a thrombotic disorder is any disorder, preferably resulting from unwanted expression of FVII, including any disorder characterized by unwanted blood clotting.

As used herein, the phrases "therapeutically effective amount" and "prophylactically effective amount" mean an amount that provides a therapeutic benefit in the treatment, prophylaxis, or relief of viral hemorrhagic fever, or a significant symptom of such disorder, e.g., bleeding, fever, weakness, muscle pain, headache. , inflammation or circulatory shock. The specific amount that is therapeutically effective can be readily determined by the ordinary practitioner and may vary depending on factors known to those skilled in the art, such as, for example, the type of thrombotic disorder, the medical history and age of the patient, the stage of the disease, and the administration of other agents. .

As used herein, the pharmaceutical composition includes a pharmacologically effective amount of dsRNA and a pharmaceutically acceptable carrier. As used herein, a pharmacologically effective amount, a therapeutically effective amount, or simply an effective amount means an amount of RNA effective to produce the desired pharmacological, therapeutic, or prophylactic result. For example, if a given clinical treatment is considered effective if there is at least a 25% reduction in the value of a measurable parameter associated with a disease or disorder, a therapeutically effective amount of a drug for treating the disease or disorder is the amount required to achieve at least 25 % reduction in the value of this parameter.

The term pharmaceutically acceptable carrier means a carrier for administering a therapeutic agent. Such carriers include, but are not limited to, sodium chloride solution, sodium chloride buffer solution, dextrose, water, glycerol, ethanol, and combinations thereof. The term specifically excludes cell culture medium. For drugs administered orally, pharmaceutically acceptable carriers include, but are not limited to, pharmaceutically acceptable excipients such as inert diluents, disintegrating agents, binding agents, lubricants, sweetening agents, flavoring agents, coloring agents and preservatives. Suitable inert diluents include sodium and calcium carbonate, sodium and calcium phosphate and lactose, while corn starch and alginic acid are useful disintegrating agents. Binding agents may include starch and gelatin, and lubricants, if present, will generally be magnesium stearate, stearic acid or talc. If necessary, the tablets may be coated with materials such as glyceryl monostearate or glyceryl distearate to delay absorption from the gastrointestinal tract.

As used herein, a transformed cell is a cell into which a vector has been introduced from which a dsRNA molecule can be expressed.

Characteristics of lipid-nucleic acid particles.

In some embodiments, the invention provides methods and compositions for producing nucleic acid lipid-encapsulated particles, wherein the nucleic acids are encapsulated within a lipid layer. Such lipid-nucleic acid particles containing siRNA oligonucleotides are characterized using various biophysical parameters, including (1) drug-to-lipid ratio; (2) encapsulation efficiency and (3) particle size. High drug to lipid ratios, high encapsulation efficiency, good nuclease resistance and serum stability, and controlled particle size, generally less than 200 nm in diameter, are desirable characteristics. In addition, the nature of the nucleic acid polymer is important, since modification of nucleic acids to confer resistance to nucleases increases the cost of treatment, while in many cases providing only limited resistance. Unless otherwise stated, these criteria are calculated in the present specification as follows.

The nucleic acid to lipid ratio is equal to the amount of nucleic acid in a given volume of production divided by the amount of lipid in the same volume. The ratio can be expressed as moles per mole, mass per mass, or mass per mole. For the final formulation ready for administration, the nucleic acid:lipid ratio is calculated after as much unencapsulated nucleic acid as possible has been removed by dialysis, chromatography and/or enzyme digestion (eg, nucleases).

The efficiency of encapsulation is equal to the ratio of drug to lipid at the beginning

- 52 046694 of the mixture divided by the drug to lipid ratio in the final administerable composition. This is a measure of relative effectiveness. To measure absolute effectiveness, it is necessary to calculate the total amount of nucleic acid added to the initial mixture that will end up in an administerable composition. The amount of lipid lost during the composition preparation process can also be calculated. Efficiency is a measure of the loss and cost of a composition.

Size indicates the size (diameter) of the particles produced. Size distribution can be determined using quasi-elastic light scattering (QELS) on a Nicomp Model 370 Submicron Particle Size Analyzer. Particle sizes less than 200 nm are preferred for distribution in neovascularized (increased permeability) tissues such as neoplasms and areas of inflammation.

Methods for producing lipid particles.

The methods and compositions of the invention employ certain cationic lipids, the synthesis, preparation and characterization of which are described below and in the accompanying examples. In addition, the present invention is directed to methods for producing lipid particles, including those associated with a therapeutic agent, such as a nucleic acid. In the methods described herein, a mixture of lipids is combined with a buffered aqueous solution of nucleic acid to produce an intermediate mixture containing nucleic acid encapsulated in lipid particles, wherein the encapsulated nucleic acids are present in a nucleic acid/lipid ratio of about 3 to about 25% by weight, preferably 5 to 15% by weight. The intermediate mixture may optionally be subjected to particle size separation to produce lipid-encapsulated nucleic acid particles, wherein the lipid moieties are unilamellar vesicles, preferably having a diameter of from 30 to 150 nm, more preferably from about 40 to 90 nm. The pH is then increased to neutralize at least a portion of the surface charges on the lipid-nucleic acid particles, thereby ensuring that the lipid-encapsulated nucleic acid composition contains at least partially neutral surfaces.

As described above, some of these cationic lipids are aminolipids, which carry charges at pH values below the pKa of the amino group, and are substantially neutral at pH values above the pKa. Such cationic lipids are called titratable cationic lipids and can be used in the compositions of the invention using a two-step process. In the first step, lipid vesicles are formed at low pH from titratable cationic lipids and other vesicle components in the presence of nucleic acids. Thus, vesicles encapsulate and retain nucleic acids. In the second stage, the surface charge of the newly formed vesicles can be neutralized by increasing the pH value of the medium to a level above the pKa value of the titratable cationic lipids present, i.e. to physiological pH or higher. In particular, advantageous features of this method include both the easy removal of any surface-adsorbed nucleic acid and the resulting nucleic acid delivery vesicle that has a neutral surface. It is believed that liposomes or lipid particles having a neutral surface will avoid rapid clearance from the circulatory system and will not have certain toxic properties that are associated with cationic liposome preparations. Additional detailed descriptions of these uses of similar titratable cationic lipids in lipid-nucleic acid particle compositions are provided in US Pat. Nos. 6,287,591 and 6,858,225, incorporated herein by reference.

It is further noted that vesicles formed in this manner provide uniformly sized vesicle compositions with a high nucleic acid content. Additionally, the vesicles have a size range of from about 30 to about 150 nm, more preferably from about 30 to about 90 nm.

Without being bound by any particular theory, it is believed that the very high efficiency of nucleic acid encapsulation is the result of electrostatic interaction at low pH values. At acidic pH values (e.g., pH 4.0), the surface of the vesicle is charged and binds some of the nucleic acids through electrostatic interactions. When the external acidic buffer solution is replaced with a more neutral buffer solution (eg, pH 7.5), the surface of the lipid particle or liposome is neutralized, allowing the removal of any external nucleic acid. More detailed information on the method of obtaining formulations is presented in various publications (for example, in US patents No. 6287591 and 6858225).

In view of the above, the present invention is directed to the creation of methods for producing lipid-nucleic acid compositions. In the methods described herein, a mixture of lipids is combined with a buffered aqueous nucleic acid solution to produce an intermediate mixture containing the nucleic acid encapsulated in lipid particles, for example, where the encapsulated nucleic acids are present in a nucleic acid/lipid ratio of about 10 to approximately 20 wt.%. The intermediate mixture may optionally be subjected to separation

- 53 046694 variation in particle size to produce lipid-encapsulated nucleic acid particles, wherein the lipid moieties are unilamellar vesicles, preferably having a diameter of from 30 to 150 nm, more preferably from about 40 to 90 nm. The pH is then increased to neutralize at least a portion of the surface charges on the lipid-nucleic acid particles, thereby ensuring that the lipid-encapsulated nucleic acid composition contains at least partially neutral surfaces.

In some embodiments, the lipid mixture contains at least two lipid components: a first amino lipid component of the present invention that is selected from lipids having a pKa such that the lipid is cationic at pH values below the pKa and neutral at pH values above the pKa, and a second lipid component , which is selected from lipids that prevent particle aggregation during the formation of lipid-nucleic acid particles. In some embodiments, the aminolipid is a novel cationic lipid of the present invention.

When preparing lipid-nucleic acid particles according to the invention, the lipid mixture is typically a solution of lipids in an organic solvent. This lipid mixture can subsequently be dried to form a thin film or lyophilized to form a powder before hydration with an aqueous buffer solution to form liposomes. In another embodiment, in a preferred method, the lipid mixture can be solubilized in a water-miscible alcohol such as ethanol, and this ethanol solution is added to an aqueous buffer solution, resulting in spontaneous formation of liposomes. In most embodiments, the alcohol is used as it is commercially available. For example, ethanol can be used as absolute ethanol (100%), or as 95% ethanol with the remainder being water. This method is described in more detail in US Pat. No. 5,976,567.

In one embodiment of the present invention, the lipid mixture is a mixture of cationic lipids, neutral lipids (other than a cationic lipid), a sterol (eg, cholesterol), and a PEG-modified lipid (eg, PEG-DMG or PEG-cDMA) in an alcohol solvent. In preferred embodiments, the lipid mixture consists substantially of a cationic lipid, a neutral lipid, cholesterol and a PEG-modified lipid in an alcohol, more preferably ethanol. In further preferred embodiments, the first solution consists of the above lipid mixture with molar ratios of about 20-70% cationic lipid:5-45% neutral lipid:20-55% cholesterol:0.5-15% PEG-modified lipid. In even more preferred embodiments, the first solution consists substantially of a lipid selected from table. 1, DSPC, cholesterol and PEG-DMG or PEG-cDMA, more preferably in a molar ratio of about 20-60% cationic lipid:5-25% DSPC:25-55% cholesterol:0.5-15% PEG-DMG or PEG -DMA. In some embodiments, the lipid molar ratio is about 50/10/38.5/1.5 (mol% cationic lipid/DSPC/cholesterol/PEG-DMG, PEG-DSG, or PEG-DPG), 57.2/7. 1/34.3/1.4 (mol.% cationic lipid/DPPC/cholesterol/PEG-cDMA), 40/15/40/5 (mol.% cationic lipid/DSPC/cholesterol/PEG-DMG), 50/ 10/35/4.5/0.5 (mol.% cationic lipid/DSPC/cholesterol/PEG-DSG or GαlNAc3-PEG-DSG), 50/10/35/5 (cationic lipid/DSPC/cholesterol/PEG- DMG), 40/10/40/10 (mol.% cationic lipid/DSPC/cholesterol/PEG-DMG or PEG-cDMA), 35/15/40/10 (mol.% cationic lipid/DSPC/cholesterol/PEG- DMG or PEG-cDMA) or 52/13/30/5 (mol% cationic lipid/DSPC/cholesterol/PEG-DMG or PEG-cDMA). In another group of preferred embodiments, the neutral lipid in these compositions is replaced by POPC, DPPC, DOPE or SM.

According to the invention, the lipid mixture is mixed with a buffered aqueous solution, which may contain nucleic acids. A buffered aqueous solution is typically a solution in which the buffer has a pH value below the pKa value of the protonatable lipid in the lipid mixture. Examples of suitable buffer solutions include citrate, phosphate, acetate and MES buffers. A particularly preferred buffer solution is citrate buffer. Preferred buffer solutions will have a concentration in the range of 11,000 mM per anion, depending on the chemical properties of the nucleic acid that is encapsulated, and optimizing the buffer concentration may be an essential step to achieve high loading levels (see, for example, US patents No. 6287591 and 6858225) .

In another embodiment, pure water acidified to pH 5-6 with chloride, sulfate or the like may be useful. In this case, it may be useful to add 5% glucose or other nonionic solute that balances the osmotic potential across the particle membrane when the particles are dialyzed to remove ethanol, raise the pH, or when mixed with a pharmaceutically acceptable carrier such as isotonic sodium chloride solution. The amount of nucleic acid in the buffer solution may vary, but will typically be from about 0.01 to about 200 mg/ml, more preferably from about 0.5 to about 50 mg/ml.

The lipid mixture and the buffered aqueous solution of therapeutic nucleic acids are combined to form an intermediate mixture. The intermediate mixture is typically a mixture

- 54 046694 lipid particles carrying encapsulated nucleic acids.

Additionally, the intermediate mixture may also contain some portion of nucleic acids that are attached to the surface of lipid particles (liposomes or lipid vesicles) due to the ionic interaction of negatively charged nucleic acids and positively charged lipids on the surface of the lipid particle (aminolipids or other lipids constituting the protonable first lipid component, are positively charged in a buffer solution having a pH value lower than the pKa value of the protonated lipid group). In one set of preferred embodiments, the lipid mixture is an alcohol solution of the lipids, and the volumes of each solution are adjusted such that, when combined, the final alcohol content is from about 20 to about 45% by volume. The method of combining mixtures can include any variety of methods, often depending on the scale of preparation of the composition. For example, if the total volume is approximately 10-20 ml or less, the solutions can be combined in a test tube and mixed using a vortex mixer. Large scale mixing can be carried out in suitable commercial glass containers.

Optionally, lipid-encapsulated therapeutic agent (eg, nucleic acid) complexes that are prepared by mixing the lipid mixture and a buffered aqueous solution of the therapeutic agent (nucleic acid) can be size-sorted to achieve a desired size range and relatively narrow lipid particle size distribution. Additionally, the compositions described herein will be graded by particle size with an average diameter of from about 70 to about 200 nm, more preferably from about 90 to about 130 nm. Several methods are available to sort liposomes by size to obtain the required size. One method for sizing is described in US Pat. No. 4,737,323, which is incorporated herein by reference. Sonication of the liposome suspension, either in a bath or with a probe, results in a progressive reduction in size down to small unilamellar vesicles (SUVs) of less than about 0.05 μm in size. Homogenization is another method based on the use of shear energy to fragment large liposomes to form small ones. In a typical homogenization process, multilamellar vesicles are recycled through a standard emulsion homogenizer until the selected liposome sizes are reached, typically from about 0.1 to 0.5 μm. In both methods, the particle size distribution can be monitored by standard particle sizing using a laser beam. Some of the methods in this document use extrusion to produce uniformly sized vesicles.

Extrusion of liposome formulations through a fine porous polycarbonate membrane or asymmetric ceramic membrane results in a relatively well-defined size distribution. Typically, the suspension is circulated through the membrane one or more times until the desired size distribution of liposome complexes is achieved. Liposomes can be extruded through successive membranes of decreasing pore size to achieve gradual reduction in liposome size. In some cases, the resulting lipid-nucleic acid compositions can be used without separation by size.

In some embodiments, the methods of the present invention further include the step of neutralizing at least some of the surface charges on the lipid portion of the lipid nucleic acid compositions. By neutralizing at least partially the surface charges, unencapsulated nucleic acid is released from the surface of the lipid particle and can be removed from the composition using standard methods. Preferably, non-encapsulated and surface-absorbed nucleic acids are removed from the final composition by exchanging buffer solutions. For example, replacing the citrate buffer (pH approximately 4.0, used during formulation) with HEPES-buffered sodium chloride solution (HBS, pH approximately 7.5) neutralizes the liposome surface and releases nucleic acid from the surface. The released nucleic acid can then be removed by chromatography using standard techniques, followed by a buffer with a pH value above the pKa of the lipid used.

Optionally, lipid vesicles (ie lipid particles) can be formed by hydration in an aqueous buffer solution and sorted by size using the methods described above before adding the nucleic acid. As described above, the aqueous buffer solution should have a pH value below the pKa value of the aminolipid. A solution of nucleic acids can be added to these size-sorted, preformed vesicles. In order for encapsulation of the nucleic acids to occur in these preformed vesicles, the mixture must contain an alcohol such as ethanol. In the case of ethanol, it should be present in a concentration of from about 20 to about 45 wt./vol.%. In addition, it may be necessary to heat the mixture of preformed vesicles and nucleic acid in a mixture of aqueous buffer solution and ethanol to a temperature of from about 25 to about 50°C, depending on the composition of the lipid vesicles and the nature of the nucleic acid. One skilled in the art will appreciate that optimizing the encapsulation process to achieve the desired level of nucleic acid content in the lipid

- 55 046694 vesicles will require changes in variables such as ethanol concentration and temperature. Examples of suitable conditions for nucleic acid encapsulation are presented in the Examples section. Once the nucleic acids are encapsulated within the preformed vesicles, the external pH can be increased to neutralize, at least partially, the surface charge. Unencapsulated and surface-absorbed nucleic acids can then be removed as described above.

Mode of application.

The lipid particles of the invention can be used to deliver a therapeutic agent into a cell, in vitro or in vivo. In some embodiments, the therapeutic agent is a nucleic acid that is delivered into a cell using lipid nucleic acid particles of the invention. Although the following description of various methods of using lipid particles and related pharmaceutical compositions according to the invention uses as examples the description relating to lipid-nucleic acid particles, it is understood that these methods and compositions can be readily adapted for the delivery of any therapeutic agent for the treatment of any disease or disorders that would benefit from such treatment.

In some embodiments, the present invention is directed to methods for introducing a nucleic acid into a cell. Preferred nucleic acids for introduction into cells are siRNAs, immunostimulatory oligonucleotides, plasmids, antisense oligonucleotides and ribozymes. These methods can be carried out by contacting the particles or compositions of the invention with cells for a period of time sufficient to allow intracellular delivery.

The compositions of the invention can be adsorbed to virtually any cell type, such as tumor cell lines, including, but not limited to, HeLa, HCT116, A375, MCF7, B16F10, Hep3b, HUH7, HepG2, Skov3, U87, and PC3 cell lines. After adsorption, lipid nucleic acid particles can either be taken up by endocytosis by a portion of the cells, exchange lipids with cell membranes, or fuse with cells. Transfer or insertion of the nucleic acid portion of the complex can occur through any of these pathways. Without limiting the scope of the present invention, it is believed that if particles enter a cell by endocytosis, the particles then interact with the endosomal membrane, which leads to destabilization of the endosomal membrane, possibly through the formation of non-bilayer phases, resulting in the introduction of encapsulated nucleic acid into the cell cytoplasm. Similarly, in the case of direct fusion of particles with the plasma membrane of a cell, if such fusion occurs, the liposome membrane integrates into the cell membrane and the contents of the liposome mix with the intracellular fluid. Contact between cells and lipid-nucleic acid compositions, if it occurs in vitro, will take place in a biologically compatible environment. The concentration of the compositions can vary widely depending on the particular application, but is generally from about 1 μM to about 10 mM. In some embodiments, treatment of cells with lipid nucleic acid compositions will generally be carried out at physiological temperatures (about 37° C.) for a period of time from about 1 to 24 hours, preferably from about 2 to 8 hours. For in vitro applications, delivery of nucleic acids can carried out in any cell grown in culture, both plant and animal origin, vertebrate or invertebrate cell, from any tissue and of any type. In preferred embodiments, the cells will be animal cells, more preferably mammalian cells, and most preferably human cells.

In one set of embodiments, a suspension of lipid-nucleic acid particles is added to plate-cultured cells at 60-80% confluency having a cell density of about 10 ³ to about 10 ⁵ cells/ml, more preferably about 2 x 10 ⁴ cells/ml. The concentration of the suspension added to the cells is preferably from about 0.01 to 20 μg/ml, more preferably about 1 μg/ml.

Typical applications include the use of well-known methods for providing intracellular delivery of siRNA to knock out or silence specific cellular targets. Another embodiment involves the delivery of DNA or mRNA sequences encoding therapeutically useful polypeptides. In this case, therapy for genetic diseases is provided by supplying deficient or missing gene products (ie, for Duchenne dystrophy, see Kunkel, et al., Brit. Med. Bull, 45(3):630-643 (1989), and for cystic fibrosis, see Goodfellow, Nature, 341:102-103 (1989)). Other uses of the compositions of the invention include the introduction of antisense oligonucleotides into cells (see Bennett, et al., Mol. Pharm. 41:1023-1033, 1992).

In another embodiment, the compositions of the invention can also be used to deliver nucleic acids to cells in vivo, using methods known to one skilled in the art. With respect to the use of the present invention for the delivery of DNA or mRNA sequences, the work of Zhu, et al., Science, 261:209-211 (1993), incorporated herein by reference, describes intravenous delivery of an expression plasmid carrying a cytomegalo insert

- 56 046694 viral promoter (CMV) - chloramphenicol acetyltransferase (CAT) gene, using DOTMA-DOPE complexes. Hyde, et al., Nature, 362:250-256 (1993), incorporated herein by reference, describes the delivery of the cystic fibrosis transmembrane regulatory protein (CFTR) gene to the airway epithelium and alveoli in mouse lungs using liposomes . In Brigham, et al., Am. J. Med. Sci. 298:278-281 (1989), incorporated herein by reference, describes in vivo transfection of mouse lung cells with a functionally active prokaryotic gene encoding an intracellular enzyme, chloramphenicol acetyltransferase (CAT). Thus, the compositions according to the invention can be used for the treatment of infectious diseases.

Thus, another feature of the compositions according to the invention is that they can be used to silence or modulate a target gene, non-limiting examples of which are FVII, Eg5, PCSK9, TPX2, apoB, SAA, TTR, RSV, PDGF β 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, cyclin D gene , VEGF gene, EGFR gene, cyclin A gene, cyclin E gene, WNT-1 gene, β-catenin gene, c-MET gene, PKC gene, NFKB gene, STAT3 gene, survivin gene, Her2/Neu gene, topoisomerase I gene , topoisomerase II a gene, p73 gene, p21(WAF1/CIP1) gene, p27(KIP1) gene, PPM1D gene, RAS gene, caveolin I gene, MIB I gene, MTAI gene, M68 gene, tumor suppressor genes, tumor suppressor gene p53, member of the p53 family DN-p63, tumor suppressor gene pRb, tumor suppressor gene APC1, tumor suppressor gene BRCA1, tumor suppressor gene PTEN, chimeric gene mLL, chimeric gene BCR/ABL, chimeric gene TEL/AML1, chimeric gene EWS/ FLI1, TLS/FUS1 chimeric gene, PAX3/FKHR chimeric gene, AML1/ETO chimeric gene, av-integrin gene, FIt-1 receptor gene, tubulin gene, human papillomavirus gene, gene required for human papillomavirus replication, human immunodeficiency virus gene , gene required for human immunodeficiency virus replication, hepatitis A virus gene, gene required for hepatitis A virus replication, hepatitis B virus gene, gene required for hepatitis B virus replication, hepatitis C virus gene, gene required for hepatitis virus replication C, hepatitis D virus gene, gene required for hepatitis D virus replication, hepatitis E virus gene, gene required for hepatitis E virus replication, hepatitis F virus gene, gene required for hepatitis F virus replication, hepatitis G virus gene, gene , required for hepatitis G virus replication, hepatitis H virus gene, gene required for hepatitis H virus replication, respiratory syncytial virus gene, gene required for respiratory syncytial virus replication, herpes simplex virus gene, gene required for herpes simplex virus replication, herpes cytomegalovirus gene, gene required for herpes cytomegalovirus replication, Epstein-Barr herpes virus gene, gene required for Epstein-Barr herpes virus replication, Kaposi's sarcoma-associated herpes virus gene, gene required for sarcoma-associated herpes virus replication Kaposi, JC virus gene, human gene required for JC virus replication, myxovirus gene, gene required for myxovirus gene replication, rhinovirus gene, gene required for rhinovirus replication, coronavirus gene, gene required for coronavirus replication, West Nile virus gene , gene required for West Nile virus replication, St. Louis encephalitis virus gene, gene required for St. Louis encephalitis virus replication, tick-borne encephalitis virus gene, gene required for tick-borne encephalitis virus replication, Murray Valley encephalitis virus gene, gene, Murray Valley encephalitis virus gene required for replication, dengue virus gene, gene required for dengue virus gene replication, simian vacuolating virus 40 gene, simian vacuolating virus 40 gene required for replication, human lymphotropic T-cell virus gene, gene required for replication human lymphotropic T-cell virus, Moloney murine leukemia virus gene, gene required for replication of Moloney murine leukemia virus, encephalomyocarditis virus gene, gene required for encephalomyocarditis virus replication, measles virus gene, gene required for measles virus replication, varicella virus gene smallpox, gene required for varicella virus replication, adenovirus gene, gene required for adenovirus replication, yellow fever virus gene, gene required for yellow fever virus replication, poliovirus gene, gene required for poliovirus replication, poxvirus gene, gene, required for poxvirus replication, Plasmodium gene, gene required for Plasmodium replication, Mycobacterium ulcerans gene, gene required for Mycobacterium ulcerans replication, Mycobacterium tuberculosis gene, gene required for Mycobacterium tuberculosis replication, Mycobacterium leprae gene, gene required for Mycobacterium leprae replication , Staphylococcus aureus gene, gene required for replication of Staphylococcus aureus, gene of Streptococcus pneumoniae, gene required for replication of Streptococcus pneumoniae, gene of Streptococcus pyogenes, gene required for replication of Streptococcus pyogenes, gene of Chlamydia pneumoniae, gene required for replication of Chlamydia pneumoniae, gene Mycoplasma pneumoniae, gene required for replication of Mycoplasma pneumoniae, integrin gene, selectin gene, complement gene, chemokine gene, chemokine receptor gene, GCSF gene, Gro1 gene, Gro2 gene, Gro3 gene, PF4 gene, MIG gene, pro-platelet gene basic protein, MIP-1I gene, MIP-1J gene, RANTES gene, MCP-1 gene, MCP-2 gene, MCP-3 gene, CMBKR1 gene, CMBKR2 gene, CMBKR3 gene, CMBKR5v, AIF-1 gene, 1- gene 309, ion channel component gene,

- 57 046694 neurotransmitter receptor gene, neurotransmitter ligand gene, amyloid family gene, presenilin gene, HD gene, DRPLA gene, SCA1 gene, SCA2 gene, MJD1 gene, CACNL1A4 gene, SCA7 gene, SCA8 gene, allelic gene found in LOH cells or one allelic gene from a family of polymorphic genes.

In the case of in vivo administration, the pharmaceutical compositions are administered parenterally, i.e. intra-articular, intravenous, intraperitoneal, subcutaneous or intramuscular. In some embodiments, the pharmaceutical compositions are administered intravenously or intraperitoneally via bolus injection. For one example, see Stadler, et al., US Pat. No. 5,286,634, the contents of which are incorporated by reference herein. Intracellular delivery of nucleic acids is also discussed in Straubringer, et al., Methods in Enzymology, Academic Press, New York. 101:512-527 (1983); Mannino, et al., Biotechniques, 6:682-690 (1988); Nicolau, et al., Crit. Rev. Ther. Drug Carrier Syst. 6:239-271 (1989) and Behr, Acc. Chem. Res. 26:274-278 (1993). Additional methods of administering lipid-based therapeutics are described, for example, in Rahman et al., US Pat. No. 3,993,754; Sears, US Patent No. 4145410; Papahadjopoulos et al., US Patent No. 4235871; Schneider, US Patent No. 4224179; Lenk et al., US Patent No. 4,522,803 and Fountain et al., US Pat. No. 4,588,578.

In other methods, pharmaceuticals can be brought into contact with the target tissue by direct application of the drug to the tissue. The application can be made using a local, open or closed method. By topical is meant the direct application of a pharmaceutical preparation to tissue exposed to the environment, such as skin, pharynx, external auditory canal, etc.

Open methods are methods that involve cutting the patient's skin and directly visually observing the underlying tissue to which pharmaceuticals will be applied. This is usually achieved by surgical procedures such as thoracotomy to access the lungs, abdominal laparotomy to access the abdominal viscera, or other direct surgical means to access the target tissue.

Closed techniques are invasive techniques in which the internal target tissues are not directly observed visually, but are accessed by penetrating instruments through small wounds in the skin. For example, drugs can be injected into the peritoneum by lavage using a needle. Similarly, pharmaceuticals can be injected into the meninges or into the spinal cord by infusion during a lumbar puncture, after which the patient is positioned as is standard practice during spinal anesthesia or spinal cord examination using metrazamide. In another embodiment, the drugs can be administered through endoscopic devices.

Lipid nucleic acid compositions can also be administered as an aerosol inhaled into the lungs (see Brigham, et al., Am. J. Sci. 298(4):278-281 (1989)), or by direct injection into the site of disease ( Culver, Human Gene Therapy, Mary Ann Liebert, Inc., Publishers, New York, p. 70-71 (1994).

The methods according to the invention can be carried out in relation to various host organisms. Preferred host organisms include mammalian species such as humans, non-human primates, dogs, cats, cattle, horses, sheep, and the like.

Dosages of particles of the invention containing lipid and therapeutic agent will depend on the ratio of therapeutic agent to lipid and the judgment of the administering physician based on the age, weight and condition of the patient.

In one embodiment, the present invention is directed to a method for modulating the expression of a target polynucleotide or polypeptide. These methods generally involve contacting a cell with a lipid particle of the invention that is associated with a nucleic acid capable of modulating the expression of a target polynucleotide or polypeptide. As used herein, the term modulation means a change in the expression of a target polynucleotide or polypeptide. In various embodiments, modulation may mean increasing or increasing, or may mean decreasing or decreasing. Methods for measuring the expression level of a target polynucleotide or polypeptide are known and available to those skilled in the art and include, for example, methods using reverse transcription-polymerase chain reaction (RT-PCR) and immunohistochemical techniques. In some embodiments, the expression level of the target polynucleotide or polypeptide is increased or decreased by at least 10%, 20%, 30%, 40%, 50%, or greater than 50% compared to the corresponding control value.

For example, if increased expression of a polypeptide is desired, the nucleic acid may be contained in an expression vector that contains a polynucleotide encoding the desired polypeptide. On the other hand, if reduced expression of a polynucleotide or polypeptide is desired, the nucleic acid may be, for example, an antisense oligonucleotide, siRNA, or microRNA comprising a polynucleotide sequence that specifically hybridizes to a polynucleotide encoding the target polypeptide, thereby disrupting expression of the target polynucleotide or polypeptide. In another embodiment, the nucleic acid may be a plasmid expressing such an antisense oligonucleotide, siRNA or microRNA.

- 58 046694

In one embodiment, the present invention is directed to a method of modulating the expression of a polypeptide by a cell, comprising delivering into the cell a lipid particle that consists or substantially consists of a cationic lipid of formula I, a neutral lipid, a sterol, a PEG or a PEG-modified lipid, e.g. in a molar ratio of about 20-65% cationic lipid of formula I, 3-25% neutral lipid, 15-55% sterol and 0.5-15% PEG or PEG-modified lipid, wherein the lipid particle is associated with a nucleic acid capable of modulating polypeptide expression. In some embodiments, the lipid molar ratio is about 60/7.5/31/1.5, 57.5/7.5/31.5/3.5, 57.2/7.1/34.3/1 ,4, 52/13/30/5, 50/10/38.5/1.5, 50/10/35/5, 40/10/40/10, 40/15/40/5 or 35/15 /40/10 (mol.% cationic lipid of formula I/DSPC or DPPC/cholesterol/PEG-DMG or PEG-cDMA). In some embodiments, the lipid particle also includes a targeting moiety, such as a targeting lipid described herein (e.g., the lipid particle is substantially composed of a cationic lipid of formula I, a neutral lipid, a sterol, a PEG or PEG-modified lipid, and a targeting moiety ). In another group of embodiments, the neutral lipid in these compositions is replaced by DPPC, POPC, DOPE or SM. In another group of embodiments, the PEG or PEG-modified lipid is replaced by PEG-DSG, PEG-DMG, or PEG-DPG.

In some embodiments, the therapeutic agent is selected from siRNA, microRNA, antisense oligonucleotide, plasmid capable of expressing siRNA, microRNA, or antisense oligonucleotide, and wherein the siRNA, microRNA, or antisense oligonucleotide comprises a polynucleotide that specifically binds to a polynucleotide encoding the polypeptide or to a complementary it with a polynucleotide, such that the expression of the polypeptide is reduced.

In other embodiments, the nucleic acid is a plasmid that encodes a polypeptide or a functional variant or fragment thereof such that expression of the polypeptide or a functional variant or fragment thereof is increased.

In related embodiments, the present invention is directed to the creation of reagents useful for transfecting cells in culture. For example, the lipid compositions described herein can be used to deliver nucleic acids to cultured cells (e.g., adherent cells, suspension cells, etc.)

In related embodiments, the present invention is directed to a method of treating a disease or disorder characterized by overexpression of a polypeptide in a subject, comprising administering to the subject a pharmaceutical composition of the invention, wherein the therapeutic agent is selected from siRNA, micro RNA, an antisense oligonucleotide, and a plasmid capable of expressing siRNA, micro RNA or antisense oligonucleotide, and wherein siRNA, micro RNA or antisense RNA comprises a polynucleotide that specifically binds to a polynucleotide encoding the polypeptide or a polynucleotide complementary thereto.

In one embodiment, the pharmaceutical composition contains a lipid particle that consists or substantially consists of a cationic lipid of formula I, DSPC, cholesterol and PEG-DMG, PEG-C-DOMG or PEG-cDMA, for example, in a molar ratio of about 20- 65% cationic lipid of formula I, 3-25% neutral lipid, 15-55% sterol and 0.5-15% PEG or PEG-modified lipid PEG-DMG, PEG-C-DOMG or PEG-cDMA, wherein the lipid particle associated with a therapeutic nucleic acid. In some embodiments, the lipid molar ratio is about 60/7.5/31/1.5, 57.5/7.5/31.5/3.5, 57.2/7.1/34.3/1 ,4, 52/13/30/5, 50/10/38.5/1.5, 50/10/35/5, 40/10/40/10, 35/15/40/10 or 40/15 /40/5 (mol.% cationic lipid of formula I/DSPC/cholesterol/PEG-DMG or PEG-cDMA). In some embodiments, the lipid particle also includes a targeting lipid described herein (e.g., the lipid particle is substantially composed of a cationic lipid of formula I, a neutral lipid, a sterol, a PEG or PEG-modified lipid, and a targeting moiety (e.g., GalNAc3-PEG -DSG)). In some embodiments, when the targeting lipid contains a PEG moiety and is added to an existing liposomal composition, the amount of PEG-modified lipid is reduced such that the total amount of PEG-modified lipid (i.e., PEG-modified lipid, e.g., PEG-DMG, and PEG-containing targeting lipid), expressed as mole percent, was kept constant (eg, 0.3, 1.5, or 3.5 mole%). In another group of embodiments, the neutral lipid in these compositions is replaced by DPPC, POPC, DOPE or SM. In another group of embodiments, the PEG or PEG-modified lipid is replaced by PEG-DSG or PEG-DPG. In another related embodiment, the present invention includes a method of treating a disease or disorder characterized by decreased expression of a polypeptide in a subject, comprising administering to the subject a pharmaceutical composition of the invention, wherein the therapeutic agent is a plasmid that encodes the polypeptide or a functional variant or fragment thereof.

The present invention is further directed to a method of inducing an immune response in a subject, comprising administering to the subject a pharmaceutical composition of the invention, wherein the therapeutic agent is an immunostimulatory oligonucleotide. In some

- 59 046694 embodiments, the immune response is a humoral or mucosal immune response that consists or substantially consists of a cationic lipid of formula I, DSPC, cholesterol and PEG-DMG, PEG-C-DOMG or PEG-cDMA, for example, in a molar ratio of about 20-65% cationic lipid of formula I, 3-25% neutral lipid, 15-55% sterol and 0.5-15% PEG or PEG-modified lipid PEG-DMG, PEG-C-DOMG or PEG-cDMA, wherein the lipid particle is associated with a therapeutic nucleic acid. In some embodiments, the lipid molar ratio is about 60/7.5/31/1.5, 57.5/7.5/31.5/3.5, 57.2/7.1/34.3/1 ,4, 52/13/30/5, 50/10/38.5/1.5, 50/10/35/5, 40/10/40/10, 35/15/40/10 or 40/15 /40/5 (mol.% cationic lipid of formula I/DSPC/cholesterol/PEG-DMG or PEG-cDMA). In some embodiments, the lipid particle also includes a targeting lipid described herein (eg, the lipid particle is substantially composed of a cationic lipid of Formula I, a neutral lipid, a sterol, a PEG or PEG-modified lipid, and a targeting moiety). In some embodiments, when the targeting lipid contains a PEG moiety and is added to an existing liposomal composition, the amount of PEG-modified lipid is reduced such that the total amount of PEG-modified lipid (i.e., PEG-modified lipid, e.g., PEG-DMG, and PEG-containing targeting lipid), expressed as mole percent, was kept constant (eg, 0.3, 1.5, or 3.5 mole%). In another group of embodiments, the neutral lipid in these compositions is replaced by DPPC, POPC, DOPE or SM. In another group of embodiments, the PEG or PEG-modified lipid is replaced by PEG-DSG or PEG-DPG. In additional embodiments, the pharmaceutical composition is administered to a subject in combination with a vaccine or antigen. Thus, the invention itself is directed to the creation of vaccines containing a lipid particle according to the invention, which includes an immunostimulating oligonucleotide, and is also associated with an antigen to which an immune response is desired. In some embodiments, the antigen is a tumor antigen or is associated with an infectious agent, such as, for example, a virus, bacterium, or parasitic organism.

Many tumor antigens, infectious agent antigens, and antigens associated with other diseases are known to those skilled in the art, and examples thereof are described in the references cited herein. Antigens suitable for use in the invention include, but are not limited to, polypeptide antigens and DNA antigens. Specific examples of antigens are hepatitis A, hepatitis B, smallpox, polio, anthrax, influenza, typhus, tetanus, measles, rotavirus, diphtheria, whooping cough, tuberculosis and rubella antigens. In one embodiment, the antigen is a recombinant hepatitis B antigen. In another embodiment, the antigen is a recombinant hepatitis A antigen. In yet another embodiment, the antigen is a tumor antigen. Examples of such tumor-associated antigens are MUC-1, EBV antigen and Burkitt's lymphoma-associated antigens. An additional feature is that the antigen is a recombinant antigen of a tumor antigen related to the protein tyrosinase. One skilled in the art will be aware of other antigens suitable for use in the invention.

Tumor-associated antigens useful in the present invention include both mutated and non-mutated molecules that may be a feature of one tumor type, be common to multiple tumor types, and/or be expressed or overexpressed exclusively in tumor cells versus with normal cells. In addition to proteins and glycoproteins, tumor-specific expression patterns of carbohydrates, gangliosides, glycolipids and mucins have been documented. Examples of tumor-associated antigens for use in contemplated cancer vaccines include protein products of oncogenes, tumor suppressor genes and other genes with mutations or rearrangements characteristic of tumor cells, reactivated embryonic gene products, oncofetal antigens, tissue-specific (but not tumor-specific) differentiation antigens, receptors growth factors, cell surface carbohydrate residues, foreign viral proteins and a number of other self-proteins.

Specific embodiments of tumor-associated antigens include, for example, mutated antigens such as the protein products of the Ras p21 proto-oncogenes, the p53 and BCR-abl tumor suppressor oncogenes, as well as CDK4, MUM1, caspase 8, and β catenin; overexpressed antigens such as galectin 4, galectin 9, carbonic anhydrase, aldolase A, PRAME, Her2/neu, ErbB-2 and KSA, oncofetal antigens such as α-fetoprotein (AFP), human chorionic gonadotropin (hCG); autoantigens such as carcinomaembryonic antigen (CEA) and melanocyte differentiation antigens such as Mart 1/Melan A, gp100, gp75, tyrosinase, TRP1 and TRP2; prostate-specific antigens such as PSA, PAP, PSMA, PSM-P1 and PSM-P2; reactivated embryonic gene products such as MAGE I, MAGE 3, MAGE 4, GAGE 1, GAGE 2, BAGE, RAGE, and other testicular cancer-specific antigens such as NY-ESO1, SSX2 and SCP1; mucins such as Muc-1 and Muc-2; gangliosides such as GM2, GD2 and GD3, neutral glycolipids and glycoproteins such as Lewis (y) and globo-H; and glycoproteins such as _Tn , Thompsen-Friedenreich (TF) antigen and sTn. In this document I will also include

- 60 046694 chens as tumor-associated antigens, whole cell and tumor cell lysates, as well as their immunogenic parts, as well as idiotypes of immunoglobulins expressed by monoclonal proliferating B lymphocytes for use against B-cell lymphomas.

Pathogens include, but are not limited to, infectious agents such as viruses that infect mammals, particularly humans. Examples of infectious viruses include, but are not limited to: Retroviridae (e.g., human immunodeficiency viruses such as HIV-1 (also called HTLV-III, LAV, or HTLV-III/LAV or HIV-III); and other isolates such as HIV- LP; Picornaviridae (eg, polioviruses, hepatitis A virus; enteroviruses, human coxsackieviruses, rhinoviruses, ECHO viruses); Calciviridae (eg, strains that cause gastroenteritis); for example, dengue viruses, encephalitis viruses, yellow fever viruses); Coronoviridae (for example, coronoviruses); Rhabdoviradae (for example, vesicular stomatitis viruses, rabies viruses); ; Filoviridae (eg Ebola viruses); Paramyxoviridae (eg parainfluenza viruses, mumps virus, measles virus, respiratory syncytial virus); Bungaviridae (eg Hantaan viruses, bungaviruses, phleboviruses and nairoviruses); Arenaviridae (hemorrhagic fever viruses); Reoviridae (eg reoviruses, orbiviruses and rotaviruses); Birnaviridae; Hepadnaviridae (hepatitis B virus); Parvovirida (parvoviruses); Papovaviridae (papilloma viruses, polyoma viruses); Adenoviridae (most adenoviruses); Herpesviridae herpes simplex virus (HSV) 1 and 2, varicella zoster virus, cytomegalovirus (CMV), herpes virus; Poxviridae (variola viruses, vaccinia viruses, poxviruses); and Iridoviridae (eg, African swine fever virus); and unclassified viruses (for example, etiological agents of spongiform encephalopathies, delta hepatitis agent (considered a defective satellite of the hepatitis B virus), agents of non-A, non-B hepatitis (class 1 = with internal transmission; class 2 = with parenteral transmission (i.e. e. hepatitis C); Norwalk virus and related viruses and astroviruses).

Gram-positive and gram-negative bacteria also serve as antigens in vertebrates. Such gram-positive bacteria include, but are not limited to, Pasteurella species, Staphylococci species, and Streptococcus species. Gram-negative bacteria include, but are not limited to, Escherichia coli, Pseudomonas species and Salmonella species. Specific examples of infectious bacteria include, but are not limited to, Helicobacterpyloris, Borelia burgdorferi, Legionella pneumophilia, Mycobacteria sps (e.g., M. tuberculosis, M. avium, M. intracellulare, M. kansaii, M. gordonae), Staphylococcus aureus, Neisseria gonorrhoeae, Neisseria meningitidis, Listeria monocytogenes, Streptococcus pyogenes (group A streptococcus), Streptococcus agalactiae (group B streptococcus), Streptococcus (viridans group), Streptococcus faecalis, Streptococcus bovis, Streptococcus (anaerobic sps.), Streptococcus pneumoniae, pathogenic Campylobacter sp., Enterococcus sp . , Haemophilus infuenzae, Bacillus antracis, corynebacterium diphtheriae, corynebacterium sp., Erysipelothrix rhusiopathiae, Clostridium perfringers, Clostridium tetani, Enterobacter aerogenes, Klebsiella pneumoniae, Pasturella multocida, Bacteroides sp., Fusobacterium nucleatum, Streptobacillus moniliformis , Treponema pallidium, Treponema pertenue, Leptospira, Rickettsia and Actinomyces israelli.

Additional examples of pathogens include, but are not limited to, infectious fungi that infect mammals, and in particular humans. Examples of infectious fungi include, but are not limited to: Cryptococcus neoformans, Histoplasma capsulatum, Coccidioides immitis, Blastomyces dermatitidis, Chlamydia trachomatis, Candida albicans. Examples of infectious parasitic organisms include Plasmodium, such as Plasmodium falciparum, Plasmodium malariae, Plasmodium ovale and Plasmodium vivax. Other infectious organisms (ie protozoa) include Toxoplasma gondii.

Pharmaceutical compositions.

In one embodiment, the present invention is directed to the creation of pharmaceutical compositions containing a nucleic acid agent identified using the liver cell model screen described herein. The composition includes an agent, for example dsRNA, and a pharmaceutically acceptable carrier. The pharmaceutical composition is useful for treating a disease or disorder associated with gene expression or activity. The composition of such pharmaceutical compositions is developed based on the method of delivery. One example is compositions that are formulated for systemic administration by parenteral delivery. Pharmaceutical compositions containing the identified agent are administered in dosages sufficient to inhibit expression of the target gene, for example the factor VII gene. In general, a suitable dose of dsRNA agent will be in the range of 0.01 to 5.0 mg per kg of recipient body weight per day, typically in the range of 1 μg to 1 mg per kg of body weight per day. The pharmaceutical composition may be administered once daily, or the dsRNA may be administered in two, three, or multiple sub-doses at suitable intervals throughout the day, or even by continuous infusion or controlled release formulation delivery. In this case, the content

- 61 046694 dsRNA in each sub-dose should be appropriately reduced to achieve the total daily dose. The dosage unit can be formulated for delivery over several days, for example, using a conventional sustained release formulation that provides sustained release of dsRNA over a period of several days. Sustained release compositions are well known to those skilled in the art and are particularly useful for vaginal delivery of agents such as may be used with the agents of the invention. In this embodiment, the dosage unit contains a corresponding number of daily doses.

One skilled in the art will appreciate that several factors may influence the dosage and timing of effective treatment of a subject, including, but not limited to, the severity of the disease or disorder, prior treatments, the general health and/or age of the subject, and the presence of other diseases. Additionally, treating a subject with a therapeutically effective amount of the composition may comprise a single course of treatment or a series of courses of treatment. Calculations of effective dosages and in vivo half-lives for individual dsRNAs of the invention can be made using standard techniques or based on in vivo testing using an appropriate animal model, as described elsewhere in this document.

In some embodiments, pharmaceutical compositions containing lipid nucleic acid particles of the invention are prepared in accordance with standard methods and further contain a pharmaceutically acceptable carrier. In general, isotonic sodium chloride solution may be used as a pharmaceutically acceptable carrier. Other suitable carriers include, for example, water, buffered water, 0.9% sodium chloride solution, 0.3% glycine, etc., containing glycoproteins to enhance stability such as albumin, lipoprotein, globulin, etc. B compositions comprising sodium chloride solution or other salt-containing carrier, the carrier is preferably added after formation of the lipid particles. Thus, once the lipid-nucleic acid particle compositions have been formed, the compositions can be diluted with pharmaceutically acceptable carriers such as isotonic sodium chloride solution.

The resulting pharmaceutical preparations can be sterilized using standard, well-known sterilization techniques. Aqueous solutions can then be packaged for use or sterile filtered and lyophilized, the lyophilized preparation being mixed with a sterile aqueous solution before administration. The compositions may contain pharmaceutically acceptable auxiliary substances necessary to create conditions close to physiological, such as pH-correcting and buffering agents, tonicity-correcting agents, etc., for example, sodium acetate, sodium lactate, sodium chloride, potassium chloride, calcium chloride, etc. Additionally, the lipid suspension may include lipid protecting agents that protect the lipids during storage from damage caused by free radicals and peroxidation. Suitable lipophilic free radical scavengers such as α-tocopherol and water-soluble iron-specific chelators such as ferrioxamine.

The concentration of the lipid particle or lipid-nucleic acid particle in pharmaceutical compositions can vary widely, i.e. from less than about 0.01%, typically equal to or at least about 0.05-5%, to 10-30% by weight, and will be selected based primarily on fluid volumes, viscosities, etc., in according to the specific route of administration chosen. For example, the concentration may be increased to reduce treatment-related fluid load. This may be particularly necessary for patients with atherosclerosis-associated congestive heart failure or severe hypertension. In another embodiment, complexes consisting of inflammatory lipids may be diluted to lower concentrations to reduce inflammation at the site of administration. In one group of embodiments, the nucleic acid will carry an attached label and will be used for diagnosis (indicating the presence of a complementary nucleic acid). In this case, the amount of complexes administered will depend on the specific label used, the disease state being diagnosed, and the judgment of the clinician, but will generally be from about 0.01 to about 50 mg/kg body weight (eg, for a nucleic acid agent). preferably from about 0.1 to about 5 mg/kg body weight. In some embodiments, the complex administered contains from about 0.004 to about 50 mg/kg body weight of the nucleic acid agent (eg, from about 0.006 to about 0.2 mg/kg).

As noted above, the therapeutic agent (eg, nucleic acid) lipid particles of the invention may contain polyethylene glycol (PEG)-modified phospholipids, PEG ceramides or gangliosides, G _M1 -modified lipids, or other lipids effective to prevent or limit aggregation. The addition of such components not only prevents the aggregation of complexes. Rather, it may further provide a method of increasing the residence time in the circulatory system and increasing the delivery of the lipid and nucleic acid composition to target tissues.

- 62 046694

The present invention is directed to the creation of compositions of a lipid and a therapeutic agent in the form of a kit. Typically the kit will consist of a container having compartments for storing the various elements of the kit. The kit will contain the particles or pharmaceutical compositions of the invention, preferably in dehydrated or concentrated form, with instructions for their rehydration or dilution and administration. In some embodiments, the particles contain the active agent, while in other embodiments they do not.

Pharmaceutical compositions containing an agent identified through a liver cell model screen can be administered by a number of routes, depending on whether local or systemic treatment is required and the area to which treatment is applied. Administration can be local, pulmonary, for example, by inhalation or insufflation of powders or aerosols, including using a nebulizer; intratracheal, intranasal, epidermal or transdermal, oral or parenteral. The route of administration can be designed to result in preferential localization in specific tissues using local delivery, for example, direct intra-articular injection into the joint, rectal administration for direct delivery to the colon and intestines, intravaginal administration for delivery to the cervix and vagina, intravitreal injection for delivery to the eye. Parenteral administration includes intravenous, intra-arterial, intra-articular, subcutaneous, intraperitoneal or intramuscular injection or infusion; or intracranial, eg intrathecal or intraventricular administration.

Pharmaceutical and topical compositions may include transdermal patches, ointments, lotions, creams, gels, drops, suppositories, aerosols, solutions and powders. Standard pharmaceutical carriers, liquid, powder or oil bases, thickeners, etc. may be necessary or desirable. Condoms, gloves, etc. may also be helpful. coated. Preferred topical compositions include those wherein the dsRNA of the invention is in admixture with a topical delivery component such as a lipid, liposome, fatty acid, fatty acid ester, steroid, chelating agent or surfactant. Preferred lipids and liposomes include neutral (eg, dioleoylphosphatidylethanolamine (DOPE), dimyristoylphosphatidylcholine (DMPC), distearoylphosphatidylcholine), negatively charged (eg, dimyristoylphosphatidylglycerol, or DMPG), and cationic (eg, dioleoyltetramethylaminopropyl DOTAP and dioleoylphosphatidylethanolamine DOTMA). The dsRNAs according to the invention can be encapsulated within liposomes or can form complexes with them, in particular with cationic liposomes. In another embodiment, dsRNAs may be complexed with lipids, in particular cationic lipids. Preferred fatty acids and esters include, but are not limited to, arachidonic acid, oleic acid, arachidic acid, lauric acid, caprylic acid, capric acid, myristic acid, palmitic acid, stearic acid, linoleic acid, linolenic acid, capric acid diester, capric acid triester acids, monoolein, dilaurin, glyceryl-1monocaprate, 1-dodecyl acycloheptan-2-one, acylcarnitine, acylcholine or C _1-10 alkyl ester (eg isopropyl myristate, IPM), monoglyceride, diglyceride or pharmaceutically acceptable salts thereof. Compositions for topical use are described in detail in US patent application Ser. No. 09/315298, filed May 20, 1999, the contents of which are incorporated by reference in their entirety.

Compositions and formulations for oral administration include powders or granules, microgranules, nanogranules, suspensions or solutions in water or non-aqueous media, capsules, gelatin capsules, sachets, tablets or minitablets. Thickeners, flavoring agents, diluents, emulsifying agents, dispersing agents or binders may be required. Preferred compositions for oral administration are compositions in which the dsRNA of the invention is administered in combination with one or more penetration agents, surfactants and chelating agents. Preferred surfactants include fatty acids and/or their esters and salts, bile acids and/or their salts. Preferred bile acids/salts include chenodeoxycholic acid (CDCA) and ursodeoxycholic acid (UDCA), cholic acid, dehydrocholic acid, deoxycholic acid, glucholic acid, glycolic acid, glycodeoxycholic acid, taurocholic acid, taurodeoxycholic acid, sodium tauro-24,25-dihydro -fusidate and sodium glycodihydrofusidate. Preferred fatty acids include arachidonic acid, undecanoic acid, oleic acid, lauric acid, caprylic acid, capric acid, myristic acid, palmitic acid, stearic acid, linoleic acid, linolenic acid, capric acid diester, capric acid triester, monoolein, dilaurin, glyceryl -1-monocaprate, 1-dodecyl azacycloheptan-2-one, acylcarnitine, acylcholine or monoglyceride, diglyceride or pharmaceutically acceptable salts thereof (eg sodium). Also preferred are combinations of penetration promoting substances, for example, combinations of fatty acids/salts with bile acids/salts. In particular, a particularly preferred combination is the combination of sodium lauric acid, capric acid and UDCA. Additional substances that promote

- 63 046694 penetration, include polyoxyethylene-9-lauryl ether, polyoxyethylene-20-cetyl ether.

Delivery of dsRNA according to the invention can be carried out orally, in the form of granules, including in the form of an aerosol of dry particles, or in combination with the formation of micro- and nanogranules. Complexing agents for dsRNA include polyamino acids; polyimines; polyacrylates; polyalkyl acrylates, polyoxitanes, polyalkyl cyanoacrylates; cationized gelatins, albumins, starches, acrylates, polyethylene glycols (PEGs) and starches; polyalkylcyanoacrylates; DEAE derivatives of polyimines, pollulans, celluloses and starches. In particular, preferred complexing agents include chitosan, N-trimethylchitosan, poly-L-lysine, polyhistidine, polyornithine, polyspermine, protamine, polyvinylpyridine, polythiodiethylaminomethylethylene P(TDAE), polyaminostyrene (eg, p-amino), poly(methyl cyanoacrylate), poly (ethyl cyanoacrylate), poly(butyl cyanoacrylate), poly(isobutyl cyanoacrylate), poly(isohexyl cyanoacrylate), DEAE methacrylate, DEAE hexyl acrylate, DEAE acrylamide, DEAE albumin and DEAE dextran, polymethyl acrylate, polyhexyl acrylate, poly(P,Llactic acid) , poly(D,L-lactic acid/glycolic acid copolymer) (PLGA), alginate and polyethylene glycol (PEG). Oral compositions for dsRNA and their preparation are described in detail in US Patent Application Ser. No. 08/886829 (filed July 1, 1997), ser. No. 09/108,673 (filed July 1998), ser. No. 09/256515 (filed Feb. 23, 1999), ser. No. 09/082624 (filed May 21, 1998) and ser. No. 09/315298 (filed May 20, 1999), each of which is incorporated herein by reference in its entirety.

Compositions and formulations for parenteral, intrathecal or intraventricular administration may include sterile aqueous solutions, which may further contain buffers, diluents and other suitable additives such as, but not limited to, penetration agents, carrier compounds and other pharmaceutically acceptable carriers or Excipients.

Pharmaceutical compositions include, but are not limited to, solutions, emulsions, and compositions containing liposomes. These compositions can be prepared from a variety of components, including, but not limited to, preformed liquids, self-emulsifying solids, and self-emulsifying semi-solids.

Pharmaceutical compositions, which in yet another embodiment may be presented in dosage unit form, can be prepared in accordance with standard methods well known to those skilled in the pharmaceutical industry. Such methods include the step of combining the active substances with pharmaceutical carrier(s) or excipient(s). In general, the compositions are prepared by mixing until homogeneous and intimate the active substances with liquid carriers or finely divided solid carriers or both and then, if necessary, forming a product.

The compositions can be given in any of many possible dosage forms, including, but not limited to, tablets, capsules, gelatin capsules, liquid syrups, soft gels, suppositories and enemas. The compositions of the invention may also be prepared as suspensions in aqueous, non-aqueous or mixed media. Aqueous suspensions may additionally contain substances that increase the viscosity of the suspension, including, for example, sodium carboxymethylcellulose, sorbitol and/or dextran. The suspension may also contain stabilizers.

In one embodiment of the invention, the pharmaceutical compositions can be prepared for use in the form of a foam. Pharmaceutical foams include, but are not limited to, compositions such as emulsions, microemulsions, creams, jellies, and liposomes. Although they are generally similar in nature, these compositions differ in the components and stability of the finished product. Methods for preparing such compositions and compositions are generally known to those skilled in the pharmaceutical field and can be applied to the compositions and formulations of the invention.

The present invention is not limited in its applications to the preparation and assembly of the components set forth in the following description. The invention is suitable for other applications and can be practiced and carried out in various ways. Additionally, phraseology and terminology used in this document are intended to be descriptive and should not be construed as limiting. The use of the words including, containing or having, involving, and variations thereof in this document is intended to include all items listed thereafter and their equivalents, as well as additional items.

Examples

The following examples are provided for illustrative purposes and are not intended to limit the scope of the claimed invention.

When used in the examples given in this document, the term ApoE means ApoE3 unless otherwise specified.

Example 1. siRNA duplexes for targeting Luc and FVII.

In table 7 shows examples of sequences for targeting Luc and FVII.

- 64 046694

Table 7

Duplex Sense/antisense Sequence 5’-3’ Target 1000/2434 2433/1001 CUU ACG CUG AGU ACU UCG AdTdT U*CG AAG fUAC UCA GCG fUAA GdT*dT C*UfU ACG CUG AGfU ACU UCG AdT*dT UCG AAG UAC UCA GCG UAA GdTdT Luc Luc 2433/2434 C*UfU ACG CUG AGfU ACU UCG AdT*dT U*CG AAG fUAC UCA GCG fUAA GdT*dT Luc 1000/1001 CUU ACG CUG AGU ACU UCG AdTdT UCG AAG UAC UCA GCG UAA GdTdT Luc AD- 1596 GGAUCAUCUCAAGUCUUACdTdT GUAAGACUUGAGAUGAUCCdTdT FVII AD- 1661 GGAfUfCAfUfCfUfCAAGfUfCfUfUAfCdTsdT GfUA AGAfCfUfUGAGAflGAfUfCfCdT* dT FVII

I o γ-ο-ρ-ο LL

Note: L8: ^{0 2} · lowercase: 2'-O-methyl-modified nucleotide, *: backbone phosphorothioate bond, fN: 2'-fluoronucleotide, dN: 2'-deoxynucleotide.

Example 2 In vivo assessment of FVII using cationic lipid liposomes.

Experiments on in vivo silencing of factor VII and ApoB in rodents. C57BL/6 mice (Charles River Labs, Massachusetts, USA) and Sprague-Dawley rats (Charles River Labs, Massachusetts, USA) were injected with either sodium chloride solution or siRNA in the required compositions by injection into the tail vein in a volume of 0. 01 ml/g. At various time points after administration, animals were anesthetized by inhalation of fluorinated ether and blood samples were collected in serum separation tubes by collecting blood from the retro-orbital sinus. Serum protein factor VII levels were determined in samples using a chromogenic assay (Coaset Factor VII, DiaPharma Group, Ohio, USA or Biophen FVII, Aniara Corporation, Ohio, USA) according to the manufacturer's protocols. A standard curve was prepared using serum collected from animals treated with sodium chloride solution. In experiments where the mRNA content of liver cells was assessed at various time points after administration, the animals were sacrificed and the livers were removed and flash frozen in liquid nitrogen. Frozen liver tissue was ground into powder. Tissue lysates were prepared and factor VII and apoB mRNA levels were determined using the branched DNA assay (QuantiGene Assay, Panomics, California, USA).

Example 3. Liposomal compositions for targeting FVII.

Factor VII (FVII), a known protein involved in the coagulation cascade, is synthesized in the liver (in hepatocytes) and secreted into the plasma. Plasma FVII levels can be determined using a simple colorimetric plate assay. As such, FVII provides a useful model for determining siRNA-mediated suppression of hepatocyte protein production and for monitoring plasma concentrations and tissue distribution of lipid-nucleic acid particles and siRNA.

Factor VII knockdown in mice.

FVII activity was assessed in animals treated with FVII siRNA at 24 hours after intravenous (at a saturating dose) injection into C57BL/6 mice. FVII content was measured using a commercially available serum or tissue protein assay kit following the manufacturer's instructions on a microplate scale. The reduction in FVII was determined in comparison to untreated control mice and the results were expressed as % FVII residual. Four dose levels (2, 5, 12.5, 25 mg/kg FVII siRNA) were used in the initial screening of each new liposome formulation and the number of dosages was expanded in subsequent experiments based on the results obtained in the initial screening.

Determination of tolerability.

The tolerability of each new siRNA liposome formulation was assessed by monitoring weight changes, cellular behavior, clinical chemistry, and in some cases hematology assays. The weight of the animals was recorded before treatment and 24 hours after treatment. Data were recorded as % change in body weight. In addition to body weight measurements, a complete clinical chemistry panel, including liver function markers, was performed at each dose level (2, 5, 12.5, and 25 mg/kg siRNA) at 24 h postinjection using aliquots of serum collected for analysis. FVII. Samples were sent for analysis to the Central Veterinary Laboratory (Langley, BC). In some cases, additional mice were included in the treatment group to collect whole blood for hematological analysis.

- 65 046694

Determination of the therapeutic index.

The therapeutic index (TI) is a conditional parameter obtained by comparing toxicity and activity. In these studies, TI was defined as:

TI=MTD (maximum tolerated dose)/ED ₅₀ (50% FVII knockdown dose).

The MTD for the present studies was established as the lowest dose producing a >7% reduction in body weight and a >200-fold increase in alanine aminotransferase (ALT), a clinical biochemical marker with high specificity for liver injury in rodents. ED50 was determined from FVII dose-activity curves.

AD 1661 siRNA, which is presented in example 1, was introduced into compositions containing DLin-M-C3-DMA:DSPC:cholesterol:PEG-DMG in the following molar ratios, which were prepared and tested by the methods described in example 2: 60:7 ,5:31:1.5; 50:10:38.5:1.5 and 40:20:38.5:1.5. The results of these in vivo experiments are presented in FIG. 1, demonstrating the silencing ability of the tested compositions.

Example 4. Preparation of 1,2-di-O-alkyl-sn3-carbamoylglyceride (PEG-DMG).

^r c<\^oh

R' ⁰ la R = С^дН29 lb R = С-1вНзз lc R = CigHj?

DSC, TEA DCM D°C-RT

Ila R - C-| ₄ H ₂ g IlbR =C ₁₆ H ₃₃ lie R - Ci βΗ ₃ 7

_H2N

III mPEG _20D o-NH _{2 R} / _o \ _o 0 H ''x Ru/DCM R

0°C- ^RT lVa R = C ₁₄ N _g9

IVb R = C ₁₆ H ₃₃ IVc R = C _1S H ₃₇

Receiving IVa.

1,2-Di-O-tetradecyl-sn-glyceride Ia (30 g, 61.80 mmol) and N-succinimidyl carbonate (DSC, 23.76 g, 1.5 eq.) were placed in dichloromethane (DCM, 500 ml ) and stirred, keeping it in a mixture of ice and water. Triethylamine (TEA, 25.30 ml, 3 eq.) was added to the stirred solution and the reaction mixture was then allowed to stir overnight at ambient temperature. The progress of the reaction was monitored by TLC. The reaction mixture was diluted with DCM (400 ml) and the organic layer was washed with water (2 times 500 ml), aqueous NaHCO ₃ solution (500 ml), followed by standard examination. The resulting precipitate was dried at ambient temperature under high vacuum overnight. After drying, the crude carbonate IIa thus obtained was dissolved in dichloromethane (500 ml) and stirred in an ice bath. mPEG ₂₀₀₀ -NH ₂ (III, 103.00 g, 47.20 mmol, purchased from NOF Corporation, Japan) and anhydrous pyridine (Py, 80 ml, in excess) were added to the stirred solution under argon atmosphere. The reaction mixture was then left to stir overnight at ambient temperature. Solvents and volatile compounds were removed under vacuum and the residue was dissolved in DCM (200 ml) and applied to a column of silica gel packed in ethyl acetate. The column was first eluted with ethyl acetate and then with a gradient of 5-10% methanol in dichloromethane to obtain the desired PEG-lipid IVa as a white solid (105.30 g, 83%).

^1H NMR (CDCl3, 400 MHz) δ = 5.20-5.12 (m, 1H), 4.18-4.01 (m, 2H), 3.80-3.70 (m, 2H), 3.70-3.20 (m, -O-CH2-CH2-O-, PEG-CH2), 2.10-2.01 (m, 2H), 1.70-1.60 (m, 2H) , 1.56-1.45 (m, 4H), 1.31-1.15 (m, 48H), 0.84 (t, J=6.5 Hz, 6H).

Detected mass spectrum range: 2660-2836.

Receiving IVb.

1,2-Di-O-hexadecyl-sn-glyceride 1b (1.00 g, 1.848 mmol) and DSC (0.710 g, 1.5 eq.) were placed together in dichloromethane (20 ml) and cooled to 0°C in mixtures of ice and water. Triethylamine (1.00 ml, 3 eq.) was added and the reaction mixture was left to stir overnight. The reaction progress was monitored by TLC, diluted with DCM, washed with water (2 times), NaHCO3 solution and dried over sodium sulfate. The solvents were removed under reduced pressure and the resulting precipitate IIb was stored under high vacuum overnight. This compound was used directly for the next reaction without further purification. MPEG ₂₀₀₀ -MH ₂ III (1.50 g, 0.687 mmol, purchased from NOF Corporation, Japan) and IIb (0.702 g, 1.5 eq.) were dissolved in dichloromethane (20 ml) under argon. The reaction mixture was cooled to 0°C. Pyridine (1 ml, in excess) was added and the reaction mixture was stirred overnight. The progress of the reaction was monitored by TLC. Solvents and volatile compounds were removed under vacuum and the residue was purified by chromatography (eluted first with ethyl acetate, then

- 66 046694 gradient 5-10% MeOH/DCM) to obtain the desired compound IVb as a white solid (1.46 g, 76%).

^1H NMR (CDCi ₃ , 400 MHz) δ = 5.17 (t, J=5.5 Hz, 1H), 4.13 (dd, J=4.00 Hz, 11.00 Hz, 1H), 4 .05 (dd, J=5.00 Hz, 11.00 Hz, 1H), 3.82-3.75 (m, 2H), 3.70-3.20 (m, -O-CH ₂ -CH ₂ -O-, PEG-CH2), 2.05-1.90 (m, 2H), 1.80-1.70 (m, 2H), 1.61-1.45 (m, 6H), 1 .35-1.17 (m, 56H), 0.85 (t, J=6.5 Hz, 6H).

Detected mass spectrum range: 2716-2892.

Receiving IVc.

1,2-Di-O-octadecyl-n-glyceride Ic (4.00 g, 6.70 mmol) and DSC (2.58 g, 1.5 eq.) were placed together in dichloromethane (60 ml) and cooled to 0°C in a mixture of ice and water. Triethylamine (2.75 ml, 3 eq.) was added and the reaction mixture was left to stir overnight. The reaction progress was monitored by TLC, diluted with DCM, washed with water (2 times), NaHCO ₃ solution and dried over sodium sulfate. The solvents were removed under reduced pressure and the residue was stored under high vacuum overnight. This compound was used directly for the next reaction without further purification. MPEG ₂₀₀₀ ^H ₂ III (1.50 g, 0.687 mmol, purchased from NOF Corporation, Japan) and IIc (0.760 g, 1.5 eq.) were dissolved in dichloromethane (20 ml) under argon. The reaction mixture was cooled to 0°C. Pyridine (1 ml, in excess) was added and the reaction mixture was stirred overnight. The progress of the reaction was monitored by TLC. Solvents and volatile compounds were removed in vacuo and the residue was purified by chromatography (eluting with ethyl acetate followed by a 5-10% MeOH/DCM gradient) to obtain the desired compound IVc as a white solid (0.92 g, 48%).

^1H NMR (CDC13, 400 MHz) δ = 5.22-5.15 (m, 1H), 4.16 (dd, J=4.00 Hz, 11.00 Hz, 1H), 4.06 (dd , J=5.00 Hz, 11.00 Hz, 1H), 3.81-3.75 (m, 2H), 3.70-3.20 (m, -O-CH ₂ -CH ₂ -O- , PEG-CH ₂ ), 1.80-1.70 (m, 2H), 1.60-1.48 (m, 4H), 1.31-1.15 (m, 64H), 0.85 ( t, J=6.5 Hz, 6H).

Detected mass spectrum range: 2774-2948.

Example 5 Preparation of DLin-M-C3-DMA (ie (6Z,9Z,28Z,31Z)-heptatriaconta-6,9,28,31-tetraen-19yl-4-(dimethylamino)butanoate).

Solution of (6Z,9Z,28Z,31Z)-heptriaconte-6,9,28,31-tetraen-19-ol (0.53 g), 4-\,\-dimethy.l1minobutyric acid hydrochloride (0.51 g) , 4-\,\-dimethy.l1minopyridine(0.61 g) and 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochloride (0.53 g) in dichloromethane (5 ml) were stirred at room temperature overnight. The solution was washed with dilute hydrochloric acid and then with dilute aqueous sodium bicarbonate solution. The organic fractions were dried over anhydrous magnesium sulfate, filtered and the solvent was removed on a rotary evaporator. The residue was passed through a silica gel column (20 g) using a 1-5% methanol/dichloromethane gradient elution. The fractions containing the purified product were combined and the solvent was removed to obtain a colorless oily liquid (0.54 g). The compounds of the present invention can be synthesized by the methods described in the following articles, which are fully included in the description

1. Schlueter, Urs; Lu, Jun; Fraser-Reid, Bert. Synthetic Approaches To Heavily Lipidated Phosphoglyceroinositides. Organic Letters (2003), 5(3), 255-257.

2. King, J. F.; Allbutt, A. D. Can. J. Chem. 1970, 48, 1754-1769

3. Mach, Mateusz; Schlueter, Urs; Mathew, Felix; Fraser-Reid, Bert; Hazen, Kevin C. Comparing n-pentenyl orthoesters and n-pentenyl glycosides as alternative glycosyl donors. Tetrahedron (2002), 58(36), 7345-7354.

Example 6 Efficacy in rats of MC3 liposomes having different liposome compositions.

To test the dose-dependent effect of MC3-containing liposome compositions in rats, the following liposome compositions were prepared essentially as described in Example 2. In the following table. 8 The included components are listed as follows: MC3-DSPC-cholesterol-PEGS14. In table 8 shows examples of tested compositions.

Animals: Sprague-Dawley line.

Total: 27.

Enter. about. (µl): injection 5 µl/g.

_______________________________________________________ Table 8

Group Band size Target siRNA Conc, (mg/ml) Input, vol. (µl/g) Dose (mg/kg) Carrier 1 3 5 PBS 2 3 FVII 1661 0.06 5 0.30 50-10-38.5-1.5 3 3 FVII 1661 0.02 5 0.10 50-10-38.5-1.5 4 3 FVII 1661 0.006 5 0.03 50-10-38.5-1.5 5 3 FVII 1661 0.002 5 0.01 50-10-38.5-1.5 6 3 FVII 1661 0.06 5 0.30 40-15-40-5 7 3 FVII 1661 0.02 5 0.10 40-15-40-5 8 3 FVII 1661 0.006 5 0.03 40-15-40-5 9 3 FVII 1661 0.002 5 0.01 40-15-40-5

- 67 046694

As shown in FIG. 2, liposome formulations having 50 mol% MC3 exhibited a dose-response curve with efficacy at slightly lower siRNA concentrations than those of liposome formulations having 40 mol% MC3.

Example 7: The effectiveness of MCZ liposomes demonstrates ApoE dependence in mice.

To further test the role of ApoE in the effectiveness of various liposome formulations, wild-type and ApoE knockout mice were injected with MCZ liposomes containing AD-1661 siRNA compositions at a dose of 0.1, 0.03 and 0.01 mg/kg, essentially so as described in Example 2. Half of the liposome formulations were premixed with recombinant ApoE protein to determine whether exogenous addition of ApoE could compensate for the protein deficiency in mice. In table 9 shows examples of tested compositions.

Experimental plan.

Animals: C57BL/6 and ApoE knockouts.

Total: 42.

Enter. about. (µl): varies, depending on mass.

Table 9

Group Band size Mouse type Target siRNA Conc. (mg/ml) Dose (mg/kg) Carrier 1 3 C57BL/6 0.00 PBS 2 3 C57BL/6 FVII 1661 0.0100 0.100 MSZ 50-10-38.5-1.5 s ApoE 3 3 C57BL/6 FVII 1661 0.0030 0.030 MSZ 50-10-38.5-1.5 s ApoE 4 3 C57BL/6 FVII 1661 0.0010 0.010 MSZ 50-10-38.5-1.5 s ApoE 5 3 C57BL/6 FVII 1661 0.0100 0.100 MSZ 50-10-38.5-1.5 without ApoE 6 3 C57BL/6 FVII 1661 0.0030 0.030 MSZ 50-10-38.5-1.5 without ApoE 7 3 C57BL/6 FVII 1661 0.0010 0.010 MSZ 50-10-38.5-1.5 without ApoE 8 3 Knockout by ApoE 0.00 PBS 9 3 Knockout by ApoE FVII 1661 0.0100 0.100 MSZ 50-10-38.5-1.5 s ApoE 10 3 Knockout by ApoE FVII 1661 0.0030 0.030 MSZ 50-10-38.5-1.5 s ApoE eleven 3 Knockout by ApoE FVII 1661 0.0010 0.010 MSZ 50-10-38.5-1.5 s ApoE 12 3 Knockout by ApoE FVII 1661 0.0100 0.100 MSZ 50-10-38.5-1.5 without ApoE 13 3 Knockout by ApoE FVII 1661 0.0030 0.030 MSZ 50-10-38.5-1.5 without ApoE 14 3 Knockout by ApoE FVII 1661 0.0010 0.010 MSZ 50-10-38.5-1.5 without ApoE

In fig. Figure 3 shows a dose-dependent decrease in FVII protein levels in wild-type mice (right columns), but not in ApoE-deficient knockout mice (left columns), upon administration of MCZ-containing liposomes, suggesting a role for ApoE in cellular uptake and/or delivery to the liver. . MS3 liposomes, prepared as described above, containing 1661 siRNAs, were administered at concentrations of 0.1, 0.03 and 0.01 mg/kg separately or in a pre-mixed mixture with ApoE lipoprotein. However, at much higher doses (eg, about 1.0 mg/kg or higher), compositions containing MC3 were found to mediate FVII silencing at the mRNA and protein levels (data not shown). As shown in FIG. 3, liposome formulations containing MC3 tested were unable to mediate FVII silencing in ApoE knockout mice unless premixed with recombinant ApoE. Thus, activity can be restored in ApoE knockout mice by premixing MC3 (MC3-containing liposomes) with ApoE.

- 68 046694

Example 8. The effectiveness of MSZ-containing liposome compositions varies depending on the mole percentage and tail length of phosphocholines.

To test the effect of varying the mole percentage and tail length of phosphocholines on the efficacy of various liposomal formulations, various formulations containing DSPC, DMPC and DLPC were tested for efficacy in silencing FVII at 0.01 and 0.03 mg/kg. In table 10 shows examples of tested compositions.

Experimental plan.

Animals: C57BL/6.

Total: 45.

Enter. about. (µl): varies, depending on mass.

Table 10

Group Band size Target siRNA Conc. (mg/ml) Dose (mg/kg) Carrier 1 3 0.00 PBS 2 3 FVII 1661 0.0010 0.010 MSZ 50-10-38.5-1.5 1661 DSFKh 3 3 FVII 1661 0.0003 0.003 MSZ 50-10-38.5-1.5 1661 DSFKh 4 3 FVII 1661 0.0010 0.010 MSZ 50-10-38.5-1.5 1661 DMFC 5 3 FVII 1661 0.0003 0.003 MSZ 50-10-38.5-1.5 1661 DMFC 6 3 FVII 1661 0.0010 0.010 MSZ 50-10-38.5-1.5 1661 DLFX 7 3 FVII 1661 0.0003 0.003 MSZ 50-10-38.5-1.5 1661 DLFX 8 3 FVII 1661 0.0010 0.010 MSZ 40-20-38.5-1.5 1661 DSFKh 9 3 FVII 1661 0.0003 0.003 MSZ 40-20-38.5-1.5 1661 DSFKh 10 3 FVII 1661 0.0010 0.010 MSZ 40-20-38.5-1.5 1661 DMFC eleven 3 FVII 1661 0.0003 0.003 MSZ 40-20-38.5-1.5 1661 DMFC 12 3 FVII 1661 0.0010 0.010 MSZ 40-20-38.5-1.5 1661 DLFX 13 3 FVII 1661 0.0003 0.003 MSZ 40-20-38.5-1.5 1661 DLFX 14 3 FVII 1661 0.0010 0.010 MSZ 30-30-38.5-1.5 1661 DMFC 15 3 FVII 1661 0.0003 0.003 MSZ 30-30-38.5-1.5 1661 DMFC

In fig. 4 shows the effects of changes in MC3 mole percentage, for example, comparing 50 and 40 mol% and in the case of a DMPC-containing composition 50, 40 and 30 mol%. In fig. 4 also demonstrates the effect of neutral lipid substitution and the difference in results for MC3 liposome formulations containing DSPC, DMPC and DLPC can be seen.

Example 9. Incorporation of GalNAc lipids into liposome compositions.

To explore the possibility of alternative delivery methods, in vivo experiments were carried out using liposome formulations containing lipids conjugated to N-acetylgalactosamine (GalNAc). GalNAc was chosen as a possible targeting ligand because the GalNAc receptor is known to be highly expressed in the liver. For this reason, studies were conducted in mice and rats to test the effectiveness of MC3-containing liposome compositions additionally containing the GalNAc3-PEG-DSG lipid of formula III, essentially as described in Example 2. In all experiments, the total amount of PEG-conjugated lipids was kept constant ( for example, where 0.5 mol% GalNAc3-PEG was added, the corresponding amount of PEG-DSG was reduced by 0.5 mol%. Four animals were used in the experiment in each of nine groups per genotype.

In table 11 shows the experimental parameters of the methods, which included the use of MC3-containing liposomes with a 5% concentration of PEG lipids, where the compositions were tested on C57BL6 mice. Liposomes contained the following molar ratios: 50/10/35/5 MC3/DSPC/cholesterol/PEG-DSG. Where 0.5% GalNAc3-PEG was added, the corresponding amount of PEG-DSG was reduced to 4.5%.

Experimental plan.

Animals: C57BL6.

Total: 36.

Enter. about. (µl): varies, depending on mass.

- 69 046694

Table 11

Group Band size Target siRNA Conc. (mg/ml) Input, vol. (µl/g) Dose (mg/kg) Carrier 1 4 10 PBS 2 4 FVII 1661 0.1 10 1.00 50/10/35/5 3 4 FVII 1661 0.05 10 0.50 50/10/35/5 4 4 FVII 1661 0.025 10 0.25 50/10/35/5 5 4 FVII 1661 0.0125 10 0.125 50/10/35/5 6 4 FVII 1661 ο,ι 10 1.00 50/10/35/4.5 with 0.5% GalNAc- lipidome 7 4 FVII 1661 0.05 10 0.50 50/10/35/4.5 with 0.5% GalNAc- lipidome 8 4 FVII 1661 0.025 10 0.25 50/10/35/4.5 with 0.5% GalNAc- lipidome 9 4 FVII 1661 0.0125 10 0.125 50/10/35/4.5 with 0.5% GalNAc- lipidome

In table 12 shows the experimental parameters of the methods, which included the use of MSZ-containing liposomes with a concentration of PEG-DSG lipids of 10 mol.%, where the compositions were tested on C57BL6 mice. Liposomes contained the following molar ratios: 50/10/30/10 MC3/DSPC/cholesterol/PEG-DSG. Where 0.5% GalNAc3-PEG was added, the corresponding amount of PEG-DSG was reduced to 9.5%.

Experimental plan.

Animals: C57BL6.

Total: 36.

Enter. about. (µl): varies, depending on mass.

Table 12

Group Band size Target siRNA Conc. (mg/ml) Enter. about. (µl/g) Dose (mg/kg) Carrier 1 4 10 PBS 2 4 FVII 1661 0.5 10 5.00 50/10/30/10 3 4 FVII 1661 0.25 10 2.50 50/10/30/10 4 4 FVII 1661 0.125 10 1.25 50/10/30/10 5 4 FVII 1661 0.0625 10 0.625 50/10/30/10 6 4 FVII 1661 0.5 10 5 50/10/30/9.5 with 0.5% GalNAc lipidome 7 4 FVII 1661 0.25 10 2.50 50/10/30/9.5 with 0.5% GalNAc 8 4 FVII 1661 0.125 10 1.25 50/10/30/9.5 with 0.5% GalNAc 9 4 FVII 1661 0.0625 10 0.63 50/10/30/9.5 with 0.5% GalNAc

In fig. Figure 5 shows the effects in which increased PEG shielding suppresses non-GalNAc-mediated silencing in C57BL6 mice. This was demonstrated in C57BL6 mice for both 5% and 10% PEG concentrations. Incorporation of C^-PEG (i.e., PEG-DSG) at a concentration of 10 mol.% effectively inhibits silencing, and this effect can be overcome by replacing 0.5 mol.% PEG lipid with an equimolar amount of GalNAc lipid (t i.e. GalNAc3-PEG-DSG of formula III). Therefore, increasing PEG shielding (eg from 5 to 10 mol%) appears to suppress non-GalNAc-mediated silencing as well as overall effectiveness.

Similar experiments were also carried out on rats, with PEG lipid (also PEG-DSG) included in liposomes in two concentrations - 5 and 10 mol.%. In table 13 shows experimental parameters of methods involving the use of MC3-containing liposomes with a 5% concentration of PEGlipids, where the compositions were tested in rats. Liposomes contained the following molar ratios: 50/10/35/5 MC3/DSPC/cholesterol/PEG-DSG. Where 0.5% GalNAc3-PEG was added, the corresponding amount of PEG-DSG was reduced to 4.5%.

Experimental design.

Animals: Sprague-Dawley rats.

Total: 36.

Enter. about. (µl): injection of a loading dose of a substance.

Table 13

Group Band size Target siRNA Conc. (mg/ml) Enter. about. (µl/g) Dose (mg/kg) Carrier 1 4 5 PBS 2 4 FVII 1661 0.2 5 1.00 50/10/35/5 3 4 FVII 1661 ο,ι 5 0.50 50/10/35/5 4 4 FVII 1661 0.05 5 0.25 50/10/35/5 5 4 FVII 1661 0.025 5 0.125 50/10/35/5 6 4 FVII 1661 0.2 5 1.00 50/10/35/4.5 with 0.5% GalNAc lipidome 7 4 FVII 1661 ο,ι 5 0.50 50/10/35/4.5 with 0.5% GalNAc lipidome 8 4 FVII 1661 0.05 5 0.25 50/10/35/4.5 with 0.5% GalNAc lipidome 9 4 FVII 1661 0.025 5 0.125 50/10/35/4.5 with 0.5% GalNAc lipidome

In table 14 shows experimental parameters of methods involving the use of MC3-containing liposomes with a 10% concentration of PEG lipids, where the compositions were tested in rats. Liposomes contained the following molar ratios: 50/10/30/10 MSZ/DSPC/cholesterol/PEGDSG. Where 0.5% GalNAc3-PEG was added, the corresponding amount of PEG-DSG was reduced to 9.5%.

Experimental plan.

Animals: Sprague-Dawley.

Total: 36.

Enter. about. (µl): injection of a loading dose of a substance.

Table 14

Group Band size Target siRNA Conc. (mg/ml) Input, vol. (µl/g) Dose (mg/kg) Carrier 1 4 5 PBS 2 4 FVII 1661 1 5 5.00 50/10/30/10 3 4 FVII 1661 θ,5 5 2.50 50/10/30/10 4 4 FVII 1661 0.25 5 1.25 50/10/30/10 5 4 FVII 1661 0.125 5 0.625 50/10/30/10 6 4 FVII 1661 1 5 5.00 50/10/30/9.5 with 0.5% GalNAc lipidome 7 4 FVII 1661 0.5 5 2.50 50/10/30/9.5 with 0.5% GalNAc lipidome 8 4 FVII 1661 0.25 5 1.25 50/10/30/9.5 with 0.5% GalNAc lipidome 9 4 FVII 1661 0.125 5 0.625 50/10/30/9.5 with 0.5% GalNAc lipidome

In fig. 6 shows the results for MSZ compositions containing C ₁₈ PEG in amounts of 5 mol.% and 10 mol.%, administered to rats in the indicated dosages. Compositions containing 10 mol.% PEG-DSG showed negligible levels of silencing at tested concentrations (0.625-5 mg/kg) in rats. However, the inclusion of 0.5 mol.% GalNAc3-PEG-DSG of formula III (i.e., replacing 0.5 mol.% C18-PEG),

- 71 046694 restores the knockdown effect of FVII. Therefore, when compared with the mouse data, the higher shielding formulations generally retained efficacy better in rats, as evidenced by the difference in results for 5 and 10 mol% PEG concentrations.

Example 10. Evaluation of variations in the mole percentage of components of MC3-containing liposome compositions with or without the inclusion of 0.5 mol.% GalNAc3-PEG-DSG.

To determine the effectiveness of MC3-containing liposomes having different mole percentages of components, with or without the addition of GalNAc3-PEG-DSG, the following liposome compositions were prepared and tested in C57BL6 mice essentially as described in Example 2 above. The components listed in the table are presented in the following order: MC3/DSPC/cholesterol/PEG-DSG. Where 0.5% GalNAc3-PEG was added, the corresponding amount of PEG-DSG was reduced to 4.5%, as shown in Table 1. 15.

Animals: C57BL6.

Total: 33.

Enter. about. (µl): varies, depending on mass.

Table 15

Group Band size Target siRNA Conc. (mg/ml) Enter. about. (µl/g) Dose (mg/kg) Carrier 1 3 10 PBS 2 3 FVII 1661 0.1 10 1.00 50/10/35/5 3 3 FVII 1661 0.1 10 1.00 50/10/35/4.5 with 0.5% GalNAc lipidome 4 3 FVII 1661 0.1 10 1.00 40/15/40/5 5 3 FVII 1661 0.1 10 1.00 40/15/40/4.5 with 0.5% GalNAc lipidome 6 3 FVII 1661 0.1 10 1.00 30/25/40/5 7 3 FVII 1661 0.1 10 1.00 30/25/40/4.5 with 0.5% GalNAc lipidome 8 3 FVII 1661 0.1 10 1.00 20/35/40/5 9 3 FVII 1661 0.1 10 1.00 20/35/40/4.5 with 0.5% GalNAc lipidome

As shown in FIG. 7, the addition of GlaNAc to liposome formulations enhances FVII silencing of each formulation, i.e. where MC3 is present in amounts of 50, 40 and 30 mol.%.

Example 11. Efficacy of MC3- and GalNAc-containing liposomes in wild-type (WT) and ASGPR knockout mice.

To test the role of ASGPR in the effectiveness of various liposome compositions, wild-type and ASGPR knockout mice were administered MSZ liposomes containing the AD-1661 siRNA composition at a dose of 3, 1 and 0.3 mg/kg, as described in example 1. Components, indicated in the table are presented in the following order: MSZ/DSPC/cholesterol/PEG-DSG. Where 0.5% GalNAc3-PEG was added, the corresponding amount of PEG-DSG was reduced to 9.5%, as shown in Table 1. 16.

Experimental design.

Animals: C57BL6 and knockout for the ASGPr gene.

Total: 25 + 15.

Enter. about. (µl): varies, depending on mass.

- 72 046694

Table 16

Group Band size Target siRNA Conc. (mg/ml) Enter. about. (µl/g) Dose (mg/kg) Carrier 1 5 10 PBS 2 5 FVII 1661 0.3 10 3.00 50/10/30/10 3 5 FVII 1661 0.3 10 3.00 50/10/30/9.5 with 0.5% GalNAc lipidome 4 5 FVII 1661 0.1 10 1.00 50/10/30/9.5 with 0.5% GalNAc lipidome 5 5 FVII 1661 0.03 10 0.300 50/10/30/9.5 with 0.5% GalNAc lipidome 6 5 10 PBS 7 5 FVII 1661 0.3 10 3.00 50/10/30/10 8 5 FVII 1661 0.3 10 3.00 50/10/30/9.5 with 0.5% GalNAc lipidome

In fig. 8 shows the results of these experiments demonstrating that for compositions containing C ₁₈ PEG, restoration of FVII knockdown by inclusion of the GalNAc3-PEG-DSG lipid does not occur when administered to a strain of mice deficient in the asialoglycoprotein receptor (ASGPR), which is the putative receptor for targeting GalNAc residue.

Example 12. Synthesis of oligonucleotides.

Synthesis.

Oligonucleotides are synthesized using an AKTAoligopilot synthesizer. For the synthesis of oligonucleotides, we used a commercially available glass support with controlled pore size (dT-CPG, 500A, Prime Synthesis) and RNA phosphoramidites with standard protecting groups, 5'-O-dimethoxytrityl N ⁶ -benzoyl-2'-t-butyldimethylsilyl- adenosine-3'-ON,N'-diisopropyl-2-cyanoethylphosphoramidite, 5'-O-dimethoxytrityl-N4-acetyl-2'-t-butyldimethylsilyl-cytidine-3'-ON,N'-diisopropyl-2-cyanoethylphosphoramidite, 5'-O-dimethoxytrityl-^-isobutyryl-2H-butyldimethylsilylguanosine-3'-O-N\N-diisopropyl-2-cyanoethylphosphoramidite and 5'-O-dimethoxytrityl-2'-t-butyldimethylsilyl-uridine-3'-ON ,N'-diisopropyl-2-cyanoethylphosphoramidite (Pierce Nucleic Acids Technologies), 2'^-phosphoramidites, 5'-O-dimethoxytrityl-N ⁴ -acetyl-2'-fluoro-cytidine-3'-ON,N'-diisopropyl -2cyanoethylphosphoramidite and 5′-O-dimethoxytrityl-2′-fluoro-uridine-3′-ON,N′-diisopropyl-2-cyanoethylphosphoramidite were purchased from Promega. All phosphoramidites are used at a concentration of 0.2 M in acetonitrile (CH ₃ CN), with the exception of guanosine, which is used at a concentration of 0.2 M in 10% THF/ANC (vol%). Use an addition/recycle time of 16 minutes. The activator is 5-ethylthiotetrazole (0.75 M, American International Chemicals); iodine/water/pyridine is used to oxidize the PO group and PADS (2%) in 2,6-lutidine/acetonitrile (1:1 v/v) is used to oxidize the PS group.

The ligand-conjugated chains at the 3' end are synthesized using a solid support containing the appropriate ligand. For example, the inclusion of a cholesterol unit in the sequence is made from hydroxyprolinol-cholesterol-phosphoramidite. Cholesterol is combined with trans-4hydroxyprolinol through a 6-aminohexanoate bond to produce a hydroxyprolinolcholesterol moiety. 5′ end-labeled Cy-3 and Cy-5.5 (fluorophore) siRNAs were synthesized from the corresponding Quasar-570 (Cy-3) phosphoramidite purchased from Biosearch Technologies. Conjugation of ligands at the 5' end and at the internal position is carried out using appropriately protected ligand-phosphoramidite building blocks. Long-term, for 15 minutes, addition of a 0.1 M solution of phosphoramidite in anhydrous CH3CN in the presence of 5-(ethylthio)-III-tetrazole activator to the oligonucleotide bound to a solid support. Oxidation of the internucleotide phosphite to phosphate is carried out using a standard mixture of iodine with water, as indicated (1), or treatment with a mixture of tert-butyl hydroperoxide/acetonitrile/water (10:87:3) with a waiting time for oxidation of the conjugated oligonucleotide of 10 minutes. Phosphorothioate is introduced by oxidizing the phosphite to phosphorothioate using a sulfur transfer reagent such as DDTT (purchased from AM Chemicals), PADS and/or Becage reagent. Cholesterol phosphoramidite is synthesized independently and used at a concentration of 0.1 M in dichloromethane. The addition time for cholesterol phosphoramidite is 16 minutes.

Removal of protecting groups I (removal of protecting groups from nitrogenous bases).

Once the synthesis is complete, the carrier is transferred to a 100 mL glass flask (VWR). The oligonucleotide is cleaved from the carrier with simultaneous removal of protective groups from the bases and

- 73 046694 phosphate groups with 80 ml of a mixture of ethanol and ammonium [ammonium:ethanol (3:1)] for 6.5 hours at 55°C. The flask is cooled on ice for a short time and then the ethanol-ammonium mixture is filtered into a new 250 ml flask. The CPG is washed with 2 x 40 ml portions of ethanol/water (1:1 by volume). The volume of the mixture is then reduced to approximately 30 ml in a rotary evaporator. The mixture is then frozen on dry ice and dried under vacuum in a SpeedVac type concentrator.

Removal of protecting groups II (removal of 2'-TBDMS group).

The dried residue is resuspended in 26 ml of triethylamine, triethylamine trihydrofluoride (TEA-3HF) or pyridine-HF and DMSO (3:4:6) and heated at 60°C for 90 min to remove the tert-butyldimethylsilyl (TBDMS) groups at the 2' position. The reaction is then quenched by adding 50 ml of 20 mM sodium acetate and adjusted to pH 6.5. The oligonucleotide is stored in the freezer until purification.

Analysis.

Oligonucleotides are analyzed using high performance liquid chromatography (HPLC) before purification, and the choice of buffer and column depends on the nature of the sequence and/or conjugated ligand.

HPLC purification.

Ligand-conjugated oligonucleotides are purified using preparative reverse phase HPLC. Unconjugated oligonucleotides are purified using anion exchange HPLC on a self-packed TSKgel column. The buffer solutions used are 20 mM sodium phosphate (pH 8.5) in 10% CH3CN (buffer A) and 20 mM sodium phosphate (pH 8.5) in 10% CH3CN, 1 M NaBr (buffer B). Fractions containing full-length oligonucleotides are pooled, desalted and lyophilized. Approximately 0.15 OD of desalted oligonucleotides are diluted in water to 150 μl and pipetted into special tubes for capillary gel electrophoresis and liquid chromatography followed by mass spectrometry (LC/MS) analysis. The compounds are then analyzed by liquid chromatography followed by electrospray mass spectrometry and capillary gel electrophoresis.

Preparation of siRNA. To prepare siRNA, equimolar amounts of sense and antisense strands are heated in 1x PBS at 95°C for 5 min and slowly cooled to room temperature. The integrity of the duplex was confirmed by HPLC analysis.

Table 17

siRNA duplexes targeting Luc and FVII

Duplex Ate. ID Sequence 5'-3' Target 1000/1001 1 CUU ACG CUG AGU ACU UCG AdTdT Luc 2 UCG AAG UAC UCA GCG UAA GdTdT AD-1955 3 cuuAcGcuGAGuAcuucGAdT sdT Luc 4 UCGAAGuACUcAGCGuAAGdTsdT AD-1596 5 GGAUCAUCUCAAGUCUUACdTdT FVII 6 GUAAGACUUGAGAUGAUCCdTdT AD-1661 7 GGAfUfCAfUfCfUfCAAGfUfCfUfUAfCdTsdT FVII 8 GfUAAGAfCfUfUGAGAfUGAfUfCfCdTsdT

Lowercase letters indicate the 2'OMe modification and Nf represents the 2T-modified nitrogenous base, dT denotes deoxythymidine, s denotes phosphorothioate.

Example 13. Synthesis of tPEG ₂₀₀₀ -1,2-di-O-alkyl^n3-carbamoylglyceride.

PEG lipids, such as tPEG ₂₀₀₀ -1,2-di-O-alkyl^n3-carbamoylglyceride, were synthesized using

using the following methods: la R = G14H29 1bR = C ₁₆ H _M 1c R = C _1a H ₃₇ DSC, TEA DCM h ₂ | 0=C-RT 0 4 2a R = С-ЦНш 2c R = 3 ^{n mPEG} °° ^NH t η. _ο ^ _ο λ D-О H ^p Ru/DCM R ° ^ec - ^HT 4aR = C _1fl M _2S 4b R = C ₁₆ A ₃₃ 4c A = ·Οι _β Η ₃₇

- 74 046694 tPEG _2ooo -1,2-di-O-alkyl^n3-carbamoylglyceride.

Preparation of compound 4a (PEG-DMG): 1,2-Di-O-tetradecyl-n-glyceride 1a (30 g, 61.80 mmol) and N,N'-succinimidyl carbonate (DSC, 23.76 g, 1.5 eq.) were placed in dichloromethane (DCM, 500 ml) and stirred, keeping in a mixture of ice and water. Triethylamine (25.30 mL, 3 eq.) was added to the stirred solution and the reaction mixture was then allowed to stir overnight at ambient temperature. The progress of the reaction was monitored by TLC. The reaction mixture was diluted with DCM (400 ml) and the organic layer was washed with water (2 times 500 ml), aqueous NaHCO ₃ solution (500 ml), followed by standard examination.

The resulting precipitate was dried at ambient temperature under high vacuum overnight. After drying, the thus obtained crude carbonate 2a was dissolved in dichloromethane (500 ml) and stirred in an ice bath. mro^^-N^ (3, 103.00 g, 47.20 mmol, purchased from NOF Corporation, Japan) and anhydrous pyridine (80 mL, in excess) were added to the stirred solution under argon atmosphere. In some embodiments, the methoxy-(PEG) _k -amine has an x-value of 45 to 49, preferably 47-49, and more preferably 49. The reaction mixture is then allowed to stir overnight at ambient temperature. Solvents and volatile compounds were removed under vacuum and the residue was dissolved in DCM (200 ml) and applied to a column of silica gel packed in ethyl acetate. The column was first eluted with ethyl acetate followed by a gradient of 5-10% methanol in dichloromethane to obtain the desired PEG-lipid 4a as a white solid (105.30 g, 83%).

Ή NMR (CDCls, 400 MHz) δ = 5.20-5.12 (m, 1H), 4.18-4.01 (m, 2H), 3.80-3.70 (m, 2H), 3 ,70-3.20 (m, O-CH2-CH2-O-, PEG-CH2), 2.10-2.01 (m, 2H), 1.70-1.60 (m, 2H), 1 .56-1.45 (m, 4H), 1.31-1.15 (m, 48H), 0.84 (t, J=6.5 Hz, 6H).

Detected mass spectrum range: 2660-2836.

Preparation 4b: 1,2-Di-O-hexadecyl-sn-glyceride 1b (1.00 g, 1.848 mmol) and DSC (0.710 g, 1.5 eq) were placed together in dichloromethane (20 ml) and cooled to 0 °C in a mixture of ice and water. Triethylamine (1.00 mL, 3 eq) was added and stirred overnight. The reaction progress was monitored by TLC, diluted with DCM, washed with water (2 times), NaHCO3 solution and dried over sodium sulfate. Solvents were removed under reduced pressure and residue 2b was stored under high vacuum overnight. This compound was used directly for the next reaction without further purification. mTO^arNH 3 (1.50 g, 0.687 mmol, purchased from NOF Corporation, Japan) and compound 2b obtained in the previous step (0.702 g, 1.5 eq.) were dissolved in dichloromethane (20 ml) under argon. The reaction mixture was cooled to 0°C. Pyridine (1 ml, in excess) was added and stirred overnight. The progress of the reaction was monitored by TLC. Solvents and volatiles were removed in vacuo and the residue was purified by chromatography (eluted with ethyl acetate followed by a 5-10% MeOH/DCM gradient) to give the title 4b as a white solid (1.46 g, 76%).

^1H NMR ( _CDCl3 , 400 MHz) δ = 5.17 (t, J=5.5 Hz, 1H), 4.13 (dd, J=4.00 Hz, 11.00 Hz, 1H), 4 .05 (dd, J=5.00 Hz, 11.00 Hz, 1H), 3.82-3.75 (m, 2H), 3.70-3.20 (m, -O-CH2-CH2- O-, PEG-CH2), 2.05-1.90 (m, 2H), 1.80-1.70 (m, 2H), 1.61-1.45 (m, 6H), 1.35 -1.17 (m, 56H), 0.85 (t, J=6.5 Hz, 6H).

Detected mass spectrum range: 2716-2892.

Preparation 4c: 1,2-Di-O-octadecyl-sn-glyceride 1c (4.00 g, 6.70 mmol) and DSC (2.58 g, 1.5 eq.) were placed together in dichloromethane (60 ml) and cooled to 0°C in a mixture of ice and water. Triethylamine (2.75 mL, 3 eq.) was added and stirred overnight. The reaction progress was monitored by TLC, diluted with DCM, washed with water (2 times), NaHCO ₃ solution and dried over sodium sulfate. The solvents were removed under reduced pressure and the residue was stored under high vacuum overnight. This compound was used directly for the next reaction with further purification. mTOntw-NH 3 (1.50 g, 0.687 mmol, purchased from NOF Corporation, Japan) and compound 2c obtained in the previous step (0.760 g, 1.5 eq.) were dissolved in dichloromethane (20 ml) under argon. The reaction mixture was cooled to 0°C. Pyridine (1 ml, in excess) was added and stirred overnight. The progress of the reaction was monitored by TLC. Solvents and volatiles were removed in vacuo and the residue was purified by chromatography (eluted with ethyl acetate followed by a 5-10% MeOH/DCM gradient) to give the title compound 4c as a white solid (0.92 g, 48%).

^1H NMR ( _CDCl3 , 400 MHz) δ = 5.22-5.15 (m, 1H), 4.16 (dd, J=4.00 Hz, 11.00 Hz, 1H), 4.06 ( dd, J=5.00 Hz, 11.00 Hz, 1H), 3.81-3.75 (m, 2H), 3.70-3.20 (m, -O-CH2-CH2-O-, PEG-CH2), 1.80-1.70 (m, 2H), 1.60-1.48 (m, 4H), 1.31-1.15 (m, 64H), 0.85 (t, J=6.5 Hz, 6H).

Detected mass spectrum range: 2774-2948.

Example 14: General protocol for extrusion.

Lipids (cationic lipid of formula I, DSPC, cholesterol, DMG-PEG) are dissolved and mixed in ethanol according to the required molar ratio. Liposomes are formed using the ethanol injection method, in which mixed lipids are added to a sodium buffer solution

- 75 046694 acetate at pH 5.2. This leads to spontaneous formation of liposomes in 35% ethanol. Liposomes are extruded through a 0.08 μm polycarbonate membrane at least 2 times. The siRNA stock solution was prepared in sodium acetate solution with 35% ethanol and added for liposome loading. The solution of siRNA-liposome complexes was incubated at 37°C for 30 min and then diluted. Ethanol was removed and replaced with PBS buffer using dialysis or tangential flow filtration.

Example 15: General protocol for in-line mixing.

Two separate individual stock solutions are prepared, one containing the lipid and the other containing siRNA. A lipid stock solution containing the cationic lipid of formula I, DSPC, cholesterol and PEG lipid is prepared by dissolving in 90% ethanol. The remaining 10% is a low pH citrate buffer solution. The concentration of the initial lipid solution is 4 mg/ml. The pH range of this citrate buffer solution can be pH 3-5, depending on the type of fusogein lipid used. SiRNA is also dissolved in citrate buffer solution at a concentration of 4 mg/ml. For small scales, prepare 5 ml of each stock solution.

Stock solutions are completely clear and lipids must be completely dissolved before combining with siRNA. Therefore, the stock solutions can be heated to completely dissolve the lipids. The siRNAs used in this method can be unmodified or modified oligonucleotides and can be conjugated to lipophilic moieties such as cholesterol.

The individual stock solutions are combined by pumping each solution into a T-piece. To simultaneously control the start and stop of two flows, a pump with two Watson-Marlow channels is used. A polypropylene tube with a diameter of 1.6 mm is additionally replaced with a smaller diameter tube of 0.8 mm to increase the linear flow rate. Polypropylene tubing (inner diameter = 0.8 mm) is attached to each side of the T-piece. The polypropylene tube has a straight end with a diameter of 1.6 mm, with a resulting volume of 4.1 mm ³ . Each large end (1.6 mm) of polypropylene tubing is placed into tubes containing either a solubilized lipid stock solution or solubilized siRNA. After the T-piece, place one tube through which the combined flow will flow. After this, the tube is placed in a container with 2x volume of PBS. The PBS solution is stirred quickly. The flow rate for the pump is set at 300 rpm or 110 ml/min. Ethanol is removed and replaced with PBS by dialysis. The lipid compositions are then concentrated using centrifugation or diafiltration to the appropriate working concentration.

Example 16 Synthesis of [62,92,282,312]-heptatriaconte-6,9,28,31-tetraen-19-yl-4-(dimethylamino)butanoate] (cationic lipid of formula I or MC3)

(MSZ)

i) Vtride/tetrahydrofuran; ii) MsCl, Et ₃ N, dimethylaminopyridine/dichloromethane; iii) LiBr/dimethylformamide; iv) Mg, HCOOEt; v) NaOH, tetrahydrofuran, water; vi) EDCI, dimethylaminopyridine, diisopropylethylamine.

Preparation of alcohol compound 2.

A clean, dry glass reactor with a capacity of 200 L, equipped with an argon inlet and a thermowell, was charged with 60 L of tetrahydrofuran and 5.73 kg (20.4 mol) of linoleic acid. The reactor contents were cooled to below 0°C using an acetone/dry ice bath. To this cooled solution, 13.8 L of Vitride in toluene (60% w/v) was slowly added while maintaining the internal temperature of the reaction mixture below 0°C (note: the initial addition of vitride was an exothermic process and foaming was observed. Foaming ceased what

- 76 046694 res 15 minutes after the start of addition). Adding Vitride took 3 hours 45 minutes. Once addition was complete, the reaction mixture was stirred at ambient temperature for 2 hours. An aliquot was removed and the reaction was stopped using sat. Na ₂ SO ₄ , the crude product thus obtained was analyzed by TLC for the presence of the starting acid. TLC indicated completion of the reaction and the reaction mixture was cooled again to below 0°C in approximately 45 minutes. A saturated sodium sulfate solution (prepared by dissolving 1.1 kg sodium sulfate in 1.5 L of water) was slowly added to the reaction mixture over 45 minutes. Once this addition was complete, 25 L of ethyl acetate was added over 30 minutes with stirring. The resulting reaction mixture was filtered through a pad of celite for 45 minutes and the celite pad was washed with an additional volume of 17 L of ethyl acetate to completely remove the product from the precipitate. The combined organic fractions were concentrated under reduced pressure. The precipitate was dissolved in 15 L of ethyl acetate and the organic layer was washed with water (2 times 7 L each) and dried over sodium sulfate (1.1 kg). After filtration, the organic layer was concentrated under reduced pressure and dried under high vacuum to obtain the product, linoleyl alcohol, as an oily liquid. Approximate yield = 5.5 kg (theoretical yield = 5.43 kg). This product was used without further purification in the next step.

Method for producing linoleyl mesylate 3.

A clean, dry 200 L all glass reactor equipped with an argon inlet and thermowell was charged with 45 L of dichloromethane and 5.5 kg of the crude product obtained in step 1. To this solution was slowly added 11.5 L of triethylamine, followed by 0.252 kg (2.0 mol) dimethylaminopyridine. The solution was cooled to -10°C with a mixture of dry ice and acetone and to this cooled reaction mass was added dropwise over 3 hours a solution of mesyl chloride (3.2 L, 41.3 mol) in dichloromethane (10 L), while maintaining temperature below 0°C. Upon completion of the addition, the reaction mixture was stirred at 0°C for 1 hour, after which TLC (5% EtOAc in dichloromethane; phosphomolybdic acid stain) of the reaction mixture showed complete disappearance of the starting alcohol. 17 L of ice-cooled water was added to the reaction mixture and the layers were separated. The upper aqueous layer was washed again with 10 L of dichloromethane and the layers were separated. The combined organic layers were washed with 2 times 10 L of dilute hydrochloric acid (prepared by mixing 2 L of concentrated HCl with 18 L of reverse osmosis water), 2 times 7.5 L of water and 10 L of brine (prepared by dissolving 11 kg of NaCl in 10 liters of water purified by reverse osmosis). The organic layer was separated, dried over Na ₂ SO ₄ (2.75 kg) and filtered. The organic layer was evaporated under reduced pressure and dried under vacuum to obtain the crude mesylate as a light yellow oily liquid. Approximate yield = 7.1 kg (theoretical yield = 7.1 kg). This material was used without additional purification in the next step.

^1H NMR ( _CDCl3 , 400 MHz) δ = 5.42-5.21 (m, 4H), 4.20 (t, 2H), 3.06 (s, 3H), 2.79 (t, 2H ), 2.19-2.00 (m, 4H), 1.90-1.70 (m, 2H), 1.06-1.18 (m, 18H), 0.88 (t, 3H).

¹³ C NMR (CDCl ₃ ) δ = 130.76, 130.54, 128.6, 128.4, 70.67, 37.9, 32.05, 30.12, 29.87, 29.85, 29 .68, 29.65, 29.53, 27.72, 27.71, 26.15, 25.94, 23.09, 14.60.

Macc spectrometry. Calculated molecular weight for C ₁₉ H ₃₆ O ₃ S 344.53, experimentally determined 343.52 (M-H).

Preparation of linoleyl bromide 4.

A clean, dry, 200 L all glass reactor equipped with an argon inlet and thermowell was charged with 25 L of dimethylformamide and 7.1 kg of the crude product obtained in step 2. This mixture was cooled to -10°C with a mixture of acetone and dry ice. To this stirred mixture was added a solution of lithium bromide (2.7 kg, 31.0 mol) in 25 L of dimethylformamide over 1.5 hours, while maintaining the reaction temperature below 0°C. Upon completion of the addition, the reaction mixture was stirred at 45°C for 18-20 hours until TLC (10% EtOAc in hexanes; phosphomolybdic acid stain) of an aliquot showed complete disappearance of the starting mesylate. The reaction mixture was diluted with 70 L of water and extracted with 57 L of hexane. The aqueous layer was further extracted with 2 times 10 L of hexane and the combined organic layers (approximately 120 L) were again washed with 2 times 10 L of water and 1 time of 10 L of brine (prepared by dissolving 14 kg of sodium chloride in 10 L of water). The resulting organic layer (120 L) was dried over sodium sulfate (4 kg) and concentrated under reduced pressure to obtain the crude product (6.5 kg). The crude product was purified by column chromatography using 60-120 mesh silica gel and hexane as eluent. Concentration of the purified product gave 5.5 kg (81%, three steps) of bromide 4 as a colorless liquid.

1H NMR (CDCl ₃ , 400 MHz) δ = 5.41-5.29 (m, 4H), 4.20 (d, 2H), 3.40 (t, J=7 Hz, 2H), 2.77 (t, J=6.6 Hz, 2H), 2.09-2.02 (m, 4H), 1.88-1.00 (m, 2H), 1.46-1.27 (m, 18H ), 0.88 (t, J=3.9 Hz, 3H).

¹³ C NMR (CDCl ₃ ) δ = 130.41, 130.25, 128.26, 128.12, 34.17, 33.05, 31.75, 29.82, 29.57, 29.54, 29 ,39, 28.95, 28.38, 27.42, 27.40, 25.84, 22.79, 14.28.

Preparation of dilinoleylmethanol 6.

A clean, dry, 20-L all-glass reactor equipped with an argon inlet, reflux condenser, and thermowell was degassed and purged with argon. Loaded into the reactor

- 77 046694

277 g (11.3 mol) of activated magnesium, and after that - 1.5 l of anhydrous ether. The reactor was again degassed three times and purged with argon. Bromide 4 (2.5 kg, 7.6 mol) was dissolved in 5 L of anhydrous diethyl ether under argon and 1 L of this solution was added to the reactor, followed by 25 ml (0.35 mol) of dibromomethane. The contents of the reactor were heated to 40°C using a water bath (boiling followed by reflux was observed, indicating the beginning of the formation of the Grignard reagent). Once the reaction had begun, the heating element was removed from the reactor and the remaining 4 L of bromide was added slowly over 2 hours 30 minutes, maintaining slight reflux of the mixture. Upon completion of the addition, the reaction mixture was again heated to achieve reflux (bath temperature 45°C) for 1 hour, after which an aliquot of the reaction mixture was quenched with water and analyzed by TLC (hexane, phosphomolybdic acid stain), which showed complete consumption of the starting bromide. The reaction mixture was cooled to below 10°C using an ice bath and a solution of ethyl formate (275 ml in 4 L diethyl ether) was added over 2 hours 30 minutes, and upon completion of addition the reaction mixture was warmed to room temperature and stirred for 1 hour. The reaction mixture was again cooled to 10°C and acetone (1.15 L) was slowly added to the mixture, after which 7 L of ice-cold water and a solution of 10% sulfuric acid (prepared by diluting 3.4 L of sulfuric acid with 34 L ice-chilled water). The product was extracted 3 times with 10 L of diethyl ether and the combined organic layers were washed with 10 L of brine and dried over sodium sulfate (2 kg). Concentration of the organic layer under reduced pressure gave the crude product (2 kg) as a mixture of the desired dilinoleyl alcohol with trace amounts of O-formylated product. The crude product was redissolved in tetrahydrofuran (4 L) and charged into a 20 L glass reactor. To this solution was added NaOH solution (0.934 kg, dissolved in 8 L of ice-cold water) and the contents were heated at 65°C for 18 hours, after which TLC (10% ether in hexane) showed complete conversion of the O-formylated product to the desired dilinoleylmethanol. The reaction mixture was cooled and extracted with diethyl ether (3 x 4 L), and the combined organic layers were washed with 5 L of brine and dried over sodium sulfate (4 kg). Filtration followed by concentration of the organic layer gave the crude product. The crude product thus obtained was purified by column chromatography using 60-120 mesh silica gel and 4% ether in hexane. Concentration of purified product fractions gave purified compound 6 (1.45 kg, 80%) as a colorless liquid.

1H NMR (400 MHz, CDC1 ₃ ) δ = 5.47-5.24 (m, 8H), 3.56 (dd, J=6.8, 4.2 Hz, 1H), 2.85-2, 66 (m, 4H), 2.121.91 (m, 9H), 1.50-1.17 (m, 46H), 0.98-0.76 (m, 6H).

¹³ C NMR (101 MHz, CDCls) δ 130.41, 130.37, 128.18, 128.15, 77.54, 77.22, 76.91, 72.25, 37.73, 31.75, 29.94, 29.89, 29.83, 29.73, 29.58, 29.53, 27.46, 27.43, 25.89, 25.86, 22.80, 14.30.

Preparation of [6Z,9Z,28Z,3^]-heptatriaconte-6,9,28,31-tetraen-19-yl-4-(dimethylamino)butanoate]

MC3 (8).

Dilinoleylmethanol 6 (144 g, 272 mmol) was dissolved in 1 L of dichloromethane and the hydrochloride salt of dimethylaminobutyric acid 7 (55 g, 328 mmol) was added to this solution, followed by diisopropylethylamine (70 ml) and dimethylaminopyridine (4 g). After stirring for 5 min at ambient temperature, EDCI (80 g, 417 mmol) was added and the reaction mixture was stirred at room temperature overnight, after which TLC analysis (silica gel, 5% MeOH in CH ₂ Cl ₂ ) showed complete disappearance of the starting material. alcohol compound. The reaction mixture was diluted with CH ₂ Cl ₂ (500 ml) and washed with saturated NaHCO ₃ solution (400 ml), water (400 ml) and brine (500 ml). The combined organic layers were dried over anhydrous. Na ₂ SO ₄ and the solvents were removed under vacuum. The crude product thus obtained (180 g) was purified by flash column chromatography [2.5 kg silica gel, using the following eluents: i) column packed with 6 L of 0.1% NEt ₃ in dichloromethane; after loading ii) 4 l 0.1% NEt ₃ in dichloromethane; iii) 16 L 2% MeOH - 98% 0.1% NEt ₃ in dichloromethane; iv) 4 l 2.5% MeOH - 97.5% 0.1% NEt ₃ in dichloromethane; v) 12 L 3% MeOH - 97% 0.1% NEt ₃ in dichloromethane] to obtain purified product 8 (MC3, 159 g, 91%) as a colorless oily liquid.

1H NMR (400 MHz, CDC ₁₃ ): δ 5.46-5.23 (m, 8H), 4.93-4.77 (m, 1H), 2.83-2.66 (m, 4H), 2.37-2.22 (m, 4H), 2.20 (s, 6H), 2.10-1.96 (m, 9H), 1.85-1.69 (m, 2H), 1. 49 (d, J=5.4 Hz, 4H), 1.39-1.15 (m, 39H), 0.95-0.75 (m, 6H).

¹³ C NMR (101 MHz, CDC13): δ 173.56, 130.38, 130.33, 128.17, 128.14, 77.54, 77.22, 76.90, 74.44, 59.17 , 45.64, 34.36, 32.69, 31.73, 29.87, 29.76, 29.74, 29.70, 29.56, 29.50, 27.44, 27.41, 25 .84, 25.55, 23.38, 22.78, 14.27.

EI-mass spectrometry (+ve): Molecular mass calculated for C ₄₃ H ₇₉ NO ₂ (M+ H) ⁺ : 642.6, determined experimentally: 642.6.

Example 17. Preparation of a siRNA composition using preformed vesicles.

Cationic lipid containing particles were prepared using the preformed vesicle method. Cationic lipid, DSPC, cholesterol and PEG lipid were solubilized into eta

-

Claims (21)

zero in a molar ratio of 40/10/40/10, respectively. The lipid mixture was added to an aqueous buffer solution (50 mM citrate, pH 4) with stirring to a final ethanol and lipid concentration of 30 vol% and 6.1 mg/mL, respectively, and allowed to equilibrate at room temperature for 2 min before extrusion. Hydrated lipids were extruded through two stacked 80 nm pore size filters (Nuclepore) at 22°C using a Lipex extruder (Northern Lipids, Vancouver, BC, Canada) to achieve a vesicle diameter of 70–90 nm, as determined by using Nicomp. Typically this required 1-3 runs. In the case of some mixtures of cationic lipids that did not form small vesicles, it was possible to form stable vesicles of 70–90 nm in size by hydrating the lipid mixture with a low-pH buffer solution (50 mM citrate, pH 3) to protonate the phosphate group of the DSPC head group. SiRNA for FVII (solubilized in an aqueous solution of 50 mM citrate, pH 4, containing 30% ethanol) was added to the vesicles, previously equilibrated to 35°C, at a rate of approximately 5 ml/min with stirring. Once a final siRNA/lipid ratio of 0.06 (w/w) was reached, the mixture was incubated for an additional 30 min at 35°C to rearrange the vesicles and encapsulate FVII siRNA. The ethanol was then removed and the external buffer was replaced with PBS (155 mM NaCl, 3 mM Na ₂ HPO ₄ , 1 mM KH ₂ PO ₄ , pH 7.5) using either dialysis or tangential flow diafiltration. The final ratio of encapsulated siRNA to lipid was determined after removal of unencapsulated siRNA using size exclusion chromatography spin columns or ion exchange chromatography spin columns. A dose response curve showing % residual FVII as a function of dose (mg/kg) is shown in FIG. 9. Example 18 Determination of pKa of cationic lipid of formula I. The pKa value for the cationic lipid of formula I was determined essentially as described in (Eastman et al., 1992, Biochemistry, 31:4262-4268), using the fluorescent probe 2-(p-toluidino)-6-naphthalenesulfonic acid ( TNS), which does not fluoresce in water, but acquires noticeable fluorescence when bound to membranes. Vesicles consisting of cationic lipid/DSPC/cholesterol/PEG-c-DOMG (in a molar ratio of 40:10:40:10) were diluted to 0.1 mM in buffer solutions (130 mM NaCl, 10 mM CH ₃ COONH ₄ , 10 mM MES, 10 mM HEPES) with different pH values ranging from 2 to 11. An aliquot of an aqueous TNS solution (to a final concentration of 1 μM) was added to the diluted vesicles and after a 30 s equilibration period, the fluorescence of the TNS-containing solution was measured at excitation wavelengths and emission 321 nm and 445 nm, respectively. The pKa value for cationic lipid-containing vesicles was determined by plotting the change in fluorescence value versus the pH of the solutions and fitting the data to a sigmoid curve using the commercial plotting program IgorPro. The pKa titration curve for the cationic lipid of formula I is shown in FIG. 10. CLAIM 1. Lipid composition for delivery of a therapeutic agent into a cell, containing a cationic lipid of formula I Formula I, or a pharmaceutically acceptable salt thereof, and a therapeutic agent, wherein said therapeutic agent is a nucleic acid, wherein the nucleic acid is mRNA. 2. Lipid composition according to claim 1, containing 40-65% cationic lipid of formula I, 5-10% neutral lipid, 25-40% sterol and 0.5-10% PEG or PEG-modified lipid. 3. The lipid composition according to claim 2, wherein the neutral lipid is selected from DSPC, DPPC, DMPC, POPC, DOPE and SM. 4. The lipid composition according to claim 2, wherein the sterol is cholesterol. 5. The lipid composition according to claim 2, in which the PEG lipid is PEG-C ₁₄ to PEG-C ₂₂ , PEG-Cst ₁₄ to PEG-Ce^ or PEG-DSPE. 6. Lipid composition according to any one of claims 1 to 5, obtained by in-line mixing. 7. The lipid composition according to claim 2, containing 57.5% cationic lipid of formula I, 7.5% neutral lipid, 31.5% sterol and 3.5% PEG or PEG-modified lipid. 8. Lipid composition according to claim 7, obtained by extrusion. 9. The lipid composition according to claim 2, in which the lipid:nucleic acid ratio is from 3 to 15. 10. The lipid composition according to claim 9, in which the lipid:nucleic acid ratio is from - 79 046694 5 to 13. 11. The lipid composition according to claim 1, further containing at least one apolipoprotein. 12. The lipid composition according to claim 11, wherein the apolipoprotein is ApoE, its active polymorphic forms, isoforms, variant and mutant forms and fragments or truncated forms. 13. The lipid composition according to claim 1, further comprising a targeting lipid. 14. The lipid composition according to claim 13, wherein the targeting lipid comprises N-acetylgalactosamine. 15. The lipid composition according to claim 14, wherein N-acetylgalactosamine includes a mono-, bi- or tri-antennary carbohydrate group. 16. The lipid composition according to claim 13, wherein said targeting lipid is present in the composition in a molar amount of from 0.001 to 5%. 17. The lipid composition of claim 13, wherein said targeting lipid is a compound selected from the group consisting of compounds of formula II, formula III, formula VI and formula VII: GalNAc3-flCr Formula II ia1MAc3-PEG-DSG Formula III Folate-PEG2000-DSG Formula VI Folate-PEG3400-DSG Formula VII. 18. The lipid composition according to claim 2, containing 50% cationic lipid of formula I, 10% neutral lipid, 38.5% sterol and 1.5% PEG or PEG-modified lipid. 19. The lipid composition according to claim 2, containing 50% cationic lipid of formula I, 10% neutral lipid, 35% sterol and 5% PEG or PEG-modified lipid. 20. The lipid composition according to claim 2, containing 57.2% cationic lipid of formula I, 7.1% neutral lipid, 34.3% sterol and 1.4% PEG or PEG-modified lipid. 21. A method of delivering a therapeutic agent into a cell, comprising administering to a subject a lipid composition according to any one of claims 1 to 20. -