HK1126785B - Modified sirna molecules and uses thereof - Google Patents

Description

Modified siRNA molecules and uses thereof

CROSS-REFERENCE TO RELATED APPLICATIONS

This application claims priority from U.S. provisional application No. 60/732,964, filed on day 11, 2005 and U.S. provisional application No. 60/817,933, filed on day 6, 2006, 30, which are hereby fully incorporated by reference herein for all purposes.

Background

RNA interference (RNAi) is an evolutionarily conserved, sequence-specific mechanism triggered by double-stranded RNA (dsRNA) that induces degradation of complementary target single-stranded mRNA and "silences" the corresponding translated sequences (McManus et al, Nature Rev. Genet. (reviewed in Nature genetics), 3: 737 (2002)). RNAi works by enzymatically cleaving longer dsRNA strands into biologically active "short interfering RNA" (siRNA) sequences of about 21-23 nucleotides in length (Elbashir et al, GenesDev., 15: 188 (2001)). sirnas can be used to down-regulate or silence the transcription and translation of a gene product of interest, i.e., a target sequence.

As part of the innate defense mechanism against invading pathogens, the mammalian immune system is activated by a number of foreign RNAs (Alexopoullu et al, Nature (Nature), 413: 732-738 (2001); Heil et al, Science, 303: 1526-1529 (2004); Diebold et al, Science, 303: 1529-1531(2004)) and DNA species (Krieg, Ann. Rev. Immunol. (Ann. Rev. Ann. Annu. Rev., 20: 709-760(2002)), which result in the release of interferons and inflammatory cytokines. The consequences of activating this response can be severe, with local and systemic inflammatory responses potentially leading to toxic shock-like syndrome. These immunotoxicities can be triggered by very low doses of immunostimulatory agents, particularly in more sensitive species, including humans (Michie et al, N.Engl. J.Med. (journal of New England medical science), 318: 1481-1486 (1988); Krown et al, Semin. Oncol. (symposium of Oncology), 13: 207-217 (1986)). It has recently been demonstrated that synthetic siRNA can be potent activators of the innate immune response when administered with vectors that facilitate intracellular delivery (Judge et al, Nat. Biotechnol. (Nature Biotechnology), 23: 457-. Although not well established, immune recognition of siRNA is sequence dependent and may activate innate immune cells via the Toll-like receptor-7 (TLR7) pathway, leading to potent induction of interferon-alpha (IFN- α)) and inflammatory cytokines. Toxicity associated with administration of siRNA in vivo has been attributed to such a response (Morrissey et al, nat. Biotechnol. (Nature Biotechnology), 23: 1002-1007 (2005); Judge et al, supra).

Stability of synthetic sirnas against rapid nuclease degradation is generally seen as a prerequisite for in vivo and therapeutic applications. This can be achieved using a variety of stable chemical techniques previously developed for other nucleic acid drugs, such as ribozymes and antisense molecules (Manoharan, curr. opin. chem. biol. (a modern chemical biology point of view), 8: 570- & 579 (2004)). These include chemical modifications to the native 2 '-OH group in the ribose sugar backbone, such as 2' -O-methyl (2 'OMe) and 2' -fluoro (2 'F) substitutions as 2' -modified nucleotides can be easily introduced into siRNA during RNA synthesis. Although numerous reports have shown that a design containing 2' OMe (Czauderna et al, Nucl. acids Res. (nucleic acids research), 31: 2705-, or "locked nucleic acid" (LNA) (Hornung et al, supra; Elmen et al, Nucl. acids Res. (nucleic acids research), 33: 439-, it retains functional RNAi activity, but such modifications appear to tolerate only certain disease-defined location or sequence-related situations. In fact, in many cases, the introduction of chemical modifications into the natural siRNA duplex may have an adverse effect on RNAi activity (Hornung et al, supra; Czauderna et al, supra; Prakash et al, supra, Chiu et al, supra; Elmen et al, supra). Consequently, the design of chemically modified sirnas requires random screening methods to identify duplexes that retain potent gene silencing activity.

The weak uptake of foreign nucleic acids by cells represents an additional barrier to the development of siRNA based drugs. The siRNA may be encapsulated within liposomes, known as stable nucleic acid-lipid particles (SNALP), which enhance intracellular uptake of nucleic acids and are suitable for systemic administration. These systems are effective in modulating RNAi in vitro (Judge et al, supra) and have been shown to inhibit viral replication in a mouse model of hepatitis B at therapeutically useful siRNA doses (Morrissey et al, supra). However, these studies were performed using synthetic sirnas comprising greater than 90% modified nucleotides, which can compromise the efficacy of RNAi-mediated gene silencing.

Therefore, there is a strong need in the art for minimally modified siRNA molecules that abrogate the immunostimulatory activity of siRNA without adverse effects on RNAi activity. This invention addresses this and other needs.

Summary of The Invention

The present invention provides chemically modified siRNA molecules and methods of silencing expression of a target gene using such siRNA molecules.

The present invention is based, in part, on the surprising discovery that minimal chemical modifications, such as 2 '-O-methyl (2' OMe) modifications, at selected positions within one or both strands of an siRNA duplex are sufficient to reduce or completely eliminate the immunostimulatory activity of the siRNA. In some cases, by chemically modifying the non-target sense strand of the siRNA duplex, the immunostimulatory activity of the siRNA can be abolished, while retaining full RNAi activity. Alternatively, a minimal degree of chemical modification, such as a 2' OMe modification, at selected positions within the sense and antisense strands of the siRNA duplex is sufficient to reduce the immunostimulatory properties of the siRNA while retaining RNAi activity. Using apolipoprotein b (apob) and mitotic kinesin Eg5 as non-limiting examples of endogenous gene targets, potent gene silencing can be achieved in vivo using the modified siRNA molecules of the invention without cytokine-induced, immunotoxic or non-target effects associated with immune activation triggered by the corresponding unmodified siRNA sequences. As a result, the patient will experience the full benefit of siRNA treatment without suffering any of the immunostimulatory side effects associated with such treatment.

In one aspect, the invention provides a modified siRNA comprising a double-stranded region of about 15 to about 60 nucleotides in length (e.g., about 15 to 60, 15 to 50, 15 to 40, 15 to 30, 15 to 25, or 19 to 25 nucleotides in length), wherein the modified siRNA is less immunostimulatory than a corresponding unmodified siRNA sequence and is capable of silencing expression of a target sequence.

Typically, the modified siRNA comprises about 1% to about 100% (e.g., about 1%, 5%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, or 100%) of modified nucleotides in the duplex region of the siRNA. In preferred embodiments, less than about 20% (e.g., less than about 20%, 15%, 10%, or 5%) of the nucleotides in the double-stranded region, or from about 1% to about 20% (e.g., from about 1% -20%, 5% -20%, 10% -20%, or 15% -20%) of the nucleotides comprise modified nucleotides. As a non-limiting example, the modified siRNA may include as few as two 2 'OMe-modified nucleotides, which represent about 5% of the native 2' -OH positions in the double-stranded region of the siRNA duplex. This minimal chemical modification, when incorporated into the highly immunostimulatory siRNA duplex, can reduce or completely eliminate siRNA-mediated induction of interferon and inflammatory cytokines in vitro and in vivo (see, example 1).

In some embodiments, the modified siRNA includes modified nucleotides including, but not limited to, 2 'OMe nucleotides, 2' -deoxy-2 '-fluoro (2' F) nucleotides, 2 '-deoxy nucleotides, 2' -O- (2-Methoxyethyl) (MOE) nucleotides, Locked Nucleic Acid (LNA) nucleotides, and mixtures thereof. In preferred embodiments, the modified siRNA includes 2 'OMe nucleotides (e.g., 2' OMe purine and/or pyrimidine nucleotides), such as, for example, 2 'OMe-guanosine nucleotides, 2' OMe-uridine nucleotides, 2 'OMe-adenosine nucleotides, 2' OMe-cytosine nucleotides, and mixtures thereof. In certain instances, the modified siRNA does not include a 2' OMe-cytosine nucleotide. In other embodiments, the modified siRNA comprises a hairpin loop structure.

Modified sirnas can include modified nucleotides in one strand (i.e., sense or antisense) or both strands of the siRNA duplex region. Preferably, uridine and/or guanosine nucleotides are modified at selected positions in the double-stranded region of the siRNA duplex. With respect to uridine nucleotide modifications, at least one, two, three, four, five, six, or more uridine nucleotides in the sense and/or antisense strand may be modified uridine nucleotides, such as 2' OMe-uridine nucleotides. In some embodiments, each uridine nucleotide in the sense and/or antisense strand is a 2' OMe-uridine nucleotide. With respect to guanosine nucleotide modifications, at least one, two, three, four, five, six, or more guanosine nucleotides in the sense and/or antisense strand may be modified guanosine nucleotides, such as 2' OMe-guanosine nucleotides. In some embodiments, each guanosine nucleotide in the sense and/or antisense strand is a 2' OMe-guanosine nucleotide.

In certain embodiments, the modified siRNA is at least about 5%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% less immunostimulatory than a corresponding unmodified siRNA sequence. Preferably, the modified siRNA is at least about 80% (e.g., 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100%) less immunostimulatory than a corresponding unmodified siRNA sequence. It should be readily apparent to those skilled in the art that the immunostimulatory properties of the modified siRNA molecule and the corresponding unmodified siRNA molecule can be determined, for example, by measuring INF- α and/or IL-6 levels 2-12 hours after systemic administration in a mammal using a suitable lipid-based delivery system, such as the SNALP delivery system or other lipid nucleic acid complex (lipoplex) systems disclosed herein.

In certain embodiments, the modified siRNA has an IC₅₀IC less than or equal to corresponding unmodified siRNA ₅₀10 fold (i.e., the modified siRNA has an IC less than or equal to the corresponding unmodified siRNA₅₀10 ofMultiple IC₅₀). In other embodiments, the modified siRNA has an IC₅₀IC less than or equal to corresponding unmodified siRNA₅₀3 times of the total weight of the product. In other embodiments, the modified siRNA preferably has an IC₅₀IC less than or equal to corresponding unmodified siRNA₅₀2 times of the total weight of the powder. It should be readily apparent to those skilled in the art that dose response curves can be generated and the IC of modified sirnas and corresponding unmodified sirnas can be readily determined using methods known to those skilled in the art₅₀The value is obtained.

Preferably, the modified siRNA is at least about 80% (e.g., 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100%) less immunostimulatory than a corresponding unmodified siRNA sequence, and the modified siRNA has an IC₅₀IC less than or equal to corresponding unmodified siRNA₅₀10 times (preferably, 3 times, and more preferably, 2 times).

In another embodiment, the modified siRNA is capable of silencing expression of at least about 5%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, 100%, 105%, 110%, 115%, 120%, 125%, or more of the target sequence relative to the corresponding unmodified siRNA sequence.

In some embodiments, the modified siRNA does not include phosphate backbone modifications, e.g., in the sense and/or antisense strand of the double-stranded region. In other embodiments, the modified siRNA does not include 2' -deoxynucleotides, e.g., in the sense and/or antisense strand of the double-stranded region. In some cases, the nucleotides at the 3' -end of the double-stranded region of the sense and/or antisense strand are not modified nucleotides. In certain other cases, the nucleotides at the 3 '-end of the double-stranded region (e.g., within one, two, three, or four nucleotides of the 3' -end) proximal to the sense and/or antisense strand are not modified nucleotides.

The modified siRNA of the invention may have a 3' overhang of one, two, three, four or more nucleotides at one or both ends of the double-stranded region, or may lack an overhang (i.e., have blunt ends). Preferably, the modified siRNA has a 3' overhang of two nucleotides at each end of the double-stranded region. In certain instances, the 3 'overhang on the antisense strand has complementarity to the target sequence, and the 3' overhang on the sense strand has complementarity to the complementary strand of the target sequence (see, e.g., the ApoB siRNA duplex in table 3). Alternatively, the 3' overhang does not have complementarity to its target sequence or complementary strand. In some embodiments, the 3 ' overhang includes one, two, three, four, or more nucleotides, such as 2 ' -deoxy (2 ' H) nucleotides. Preferably, the 3' overhang includes deoxythymidine (dT) nucleotides.

In some embodiments, the corresponding unmodified siRNA sequence comprises at least one, two, three, four, five, six, seven, or more 5 '-GU-3' motifs. The 5 '-GU-3' motif can be in the sense strand, the antisense strand, or both strands of the unmodified siRNA sequence.

In certain embodiments, the modified siRNA further comprises a carrier system, e.g., to deliver the modified siRNA to a cell of a mammal. Non-limiting examples of vector systems suitable for use in the present invention include nucleic acid-lipid particles, liposomes, micelles, viral particles, nucleic acid complexes, and mixtures thereof. In certain instances, the modified siRNA molecule is complexed with a lipid, such as a cationic lipid, to form a lipid nucleic acid complex. In certain other cases, the modified siRNA molecule is complexed with a polymer, such as a cationic polymer (e.g., Polyethylenimine (PEI)), to form a polymer-nucleic acid complex (polyplex). The modified siRNA molecule may also be complexed with a cyclodextrin or a polymer thereof. Preferably, the modified siRNA molecule is encapsulated in a nucleic acid-lipid particle.

The invention also provides a pharmaceutical composition comprising a modified siRNA as described herein and a pharmaceutically acceptable carrier.

In a related aspect, the invention provides a modified siRNA comprising a double-stranded region of about 15 to about 60 nucleotides in length (e.g., about 15 to 60, 15 to 50, 15 to 40, 15 to 30, 15 to 25, or 19 to 25 nucleotides in length), wherein at least one, two, three, four, five, six, seven, eight, nine, ten, or more nucleotides in a sense strand of the siRNA comprise modified nucleotides, and no nucleotides in an antisense strand of the siRNA are modified nucleotides.

In another aspect, the present invention provides a modified siRNA comprising a double-stranded region of about 15 to about 60 nucleotides in length (e.g., about 15 to 60, 15 to 50, 15 to 40, 15 to 30, 15 to 25, or 19 to 25 nucleotides in length), wherein at least two nucleotides in the double-stranded region comprise a modified nucleotide selected from the group consisting of a modified guanosine nucleotide, a modified uridine nucleotide, and mixtures thereof. The modified siRNA is significantly less immunostimulatory than the corresponding unmodified siRNA sequence and is capable of silencing expression of the target sequence.

Typically, the modified siRNA comprises about 1% to about 100% (e.g., about 1%, 5%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, or 100%) of modified nucleotides in the double-stranded region of the siRNA duplex. In preferred embodiments, less than about 30% (e.g., less than about 30%, 25%, 20%, 15%, 10%, or 5%) or from about 1% to about 30% (e.g., about 1% -30%, 5% -30%, 10% -30%, 15% -30%, 20% -30%, or 25% -30%) of the nucleotides in the double-stranded region comprise modified nucleotides. By way of non-limiting example, the modified siRNA may comprise 10 2 ' OMe-guanosine and/or 2 ' OMe-uridine nucleotides, which represent less than about 30% of the native 2 ' -OH positions in the duplex region of the siRNA duplex. This minimal chemical modification, when incorporated into the highly immunostimulatory siRNA sequence, can reduce or completely eliminate siRNA-mediated induction of interferon and inflammatory cytokines in vitro and in vivo (see, examples 2-4).

In some embodiments, the modified siRNA includes modified guanosine and/or uridine nucleotides including, but not limited to, 2 'OMe-guanosine nucleotides, 2' OMe-uridine nucleotides, 2 'F-guanosine nucleotides, 2' F-uridine nucleotides, 2 '-deoxyguanosine nucleotides, 2' -deoxyuridine nucleotides, 2 'OMe-guanosine nucleotides, 2' OMe-uridine nucleotides, LNA guanosine nucleotides, LNA uridine nucleotides, and mixtures thereof. In preferred embodiments, the modified siRNA includes 2 'OMe-guanosine nucleotides, 2' OMe-uridine nucleotides, and mixtures thereof. In other embodiments, the modified siRNA comprises a hairpin loop structure.

The modified siRNA may include modified nucleotides in one strand (i.e., sense or antisense) or both strands of the double-stranded region of the siRNA. Preferably, at least two, three, four, five, six, seven, eight, nine, ten or more uridine and/or guanosine nucleotides are modified at selected positions in the double-stranded region of the siRNA duplex.

In certain embodiments, the modified siRNA is at least about 5%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% less immunostimulatory than a corresponding unmodified siRNA sequence. Preferably, the modified siRNA is at least about 80% (e.g., 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100%) less immunostimulatory than a corresponding unmodified siRNA sequence, and has an IC less than or equal to that of a corresponding unmodified siRNA ₅₀10 times of IC₅₀. In other embodiments, the modified siRNA has an IC less than or equal to the corresponding unmodified siRNA₅₀3 times of IC₅₀. In thatIn other embodiments, the modified siRNA preferably has an IC less than or equal to the corresponding unmodified siRNA₅₀2 times of IC₅₀。

In other embodiments, the modified siRNA is capable of silencing expression of at least about 5%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, 100%, 105%, 110%, 115%, 120%, 125% or more of the target sequence relative to the corresponding unmodified siRNA sequence.

In some embodiments, the modified siRNA does not include phosphate backbone modifications, e.g., in the sense and/or antisense strand of the double-stranded region. In other embodiments, the modified siRNA does not include 2' -deoxynucleotides, e.g., in the sense and/or antisense strand of the double-stranded region. In some cases, the nucleotides at the 3' -end of the double-stranded region of the sense and/or antisense strand are not modified nucleotides. In certain other cases, the nucleotides at the 3 '-end of the double-stranded region (e.g., within one, two, three, or four nucleotides of the 3' -end) proximal to the sense and/or antisense strand are not modified nucleotides.

The modified siRNA of the invention may have a 3' overhang of one, two, three, four or more nucleotides at one or both ends of the double-stranded region, or may lack an overhang (i.e., have blunt ends). Preferably, the modified siRNA has a 3' overhang of two nucleotides at each end of the double-stranded region. In some embodiments, the 3 ' overhang includes one, two, three, four, or more nucleotides, such as 2 ' -deoxy (2 ' H) nucleotides. Preferably, the 3' overhang includes deoxythymidine (dT) nucleotides.

In some embodiments, the corresponding unmodified siRNA sequence comprises at least one, two, three, four, five, six, seven, or more 5 '-GU-3' motifs. The 5 '-GU-3' motif can be in the sense strand, the antisense strand, or both strands of the unmodified siRNA sequence.

In certain embodiments, the modified siRNA further comprises a carrier system, e.g., to deliver the modified siRNA to a cell of a mammal. Non-limiting examples of vector systems include nucleic acid-lipid particles, liposomes, micelles, viral particles, nucleic acid complexes, and mixtures thereof.

In certain instances, the modified siRNA molecule is complexed with a lipid, such as a cationic lipid, to form a lipid nucleic acid complex. In certain other cases, the modified siRNA molecule is complexed with a polymer, such as a cationic polymer (e.g., PEI), to form a polymer-nucleic acid complex. The modified siRNA molecule may also be complexed with a cyclodextrin or a polymer thereof. Preferably, the modified siRNA molecule is encapsulated in a nucleic acid-lipid particle.

The invention also provides a pharmaceutical composition comprising a modified siRNA as described herein and a pharmaceutically acceptable carrier.

In another aspect, the invention provides a nucleic acid-lipid particle comprising a modified siRNA described herein, a cationic lipid, and a non-cationic lipid. In certain instances, the nucleic acid-lipid particle further comprises a conjugated lipid that inhibits aggregation of the particle. Preferably, the nucleic acid-lipid particle comprises a modified siRNA described herein, a cationic lipid, a non-cationic lipid, and a conjugated lipid that inhibits aggregation of the particle.

The cationic lipid may be, for example, N, N-dioleyl-N, N-dimethylammonium chloride (DODAC), N, N-distearoyl-N, N-dimethylammonium bromide (DDAB), N- (1- (2, 3-dioleoyloxy) propyl) -N, N, N-trimethylammonium chloride (DOTAP), N- (1- (2, 3-dioleyloxy) propyl) -N, N, N-trimethylammonium chloride (DOTMA), N, N-dimethyl-2, 3-dioleyloxypropylamine (DODMA), 1, 2-dioleyl (linoleyl) oxy-N, N-dimethylaminopropane (DLInDMA), 1, 2-dilinolyl (linolenyl) oxy-N, N-dimethylaminopropane (DLMA end), or mixtures thereof. The cationic lipid may comprise from about 20 mol% to about 50 mol% or about 40 mol% of the total lipid present in the particle.

The non-cationic lipid may be an anionic lipid or a neutral lipid, including, but not limited to, Distearoylphosphatidylcholine (DSPC), Dioleoylphosphatidylethanolamine (DOPE), palmitoyloleoylphosphatidylcholine (POPC), palmitoyloleoylphosphatidylethanolamine (POPE), palmitoyloleoylphosphatidylglycerol (POPG), Dipalmitoylphosphatidylethanolamine (DPPC), Dipalmitoylphosphatidylethanolamine (DPPE), Dimyristoylphosphatidylethanolamine (DMPE), Distearoylphosphatidylethanolamine (DSPE), monomethyl phosphatidylethanolamine, dimethyl phosphatidylethanolamine, Dioleoylphosphatidylethanolamine (DEPE), Stearoyloleoylphosphatidylethanolamine (SOPE), Egg Phosphatidylcholine (EPC), cholesterol, or mixtures thereof. The non-cationic lipid may comprise from about 5 mol% to about 90 mol% or about 20 mol% of the total lipid present in the particle.

The conjugated lipid that inhibits aggregation of particles may be a polyethylene glycol (PEG) -lipid conjugate, a polyamide (ATTA) -lipid conjugate, a cationic polymer-lipid Conjugate (CPLs), or a mixture thereof. In a preferred embodiment, the nucleic acid-lipid particle comprises a PEG-lipid conjugate or an ATTA-lipid conjugate. In certain embodiments, the PEG-lipid conjugate or ATTA-lipid conjugate is used with CPL. Conjugated lipids that inhibit particle aggregation may include PEG-lipids including, for example, PEG-Diacylglycerol (DAG), PEG-Dialkyloxypropyl (DAA), PEG-phospholipid, PEG-ceramide (Cer), or mixtures thereof. The PEG-DAA conjugate may be PEG-dilauryloxypropyl (C12), PEG-dimyristoyloxypropyl (C14), PEG-dipalmitoyloxypropyl (C16), or PEG-distearoyloxypropyl (C18). In some embodiments, the conjugated lipid that inhibits aggregation of particles is CPL having the formula: A-W-Y, wherein A is a lipid moiety, W is a hydrophilic polymer, and Y is a polycationic moiety. W may be a polymer selected from the group consisting of: PEG, polyamide, polylactic acid, polyglycolic acid, polylactic acid/polyglycolic acid copolymers, or combinations thereof, said polymers having a molecular weight of from about 250 to about 7000 daltons. In some embodiments, Y has at least 4 positive charges at the selected pH. In some embodiments, Y may be lysine, arginine, asparagine, glutamine, derivatives thereof, or combinations thereof. The conjugated lipid that prevents aggregation of the particles may be 0 mol% to about 20 mol% or about 2 mol% of the total lipid present in the particles.

In some embodiments, the nucleic acid-lipid particle further comprises cholesterol, e.g., from about 10 mol% to about 60 mol%, from about 30 mol% to about 50 mol%, or about 48 mol% of the total lipid present in the particle.

In certain embodiments, the particle is exposed to the nuclease at 37 ℃ for at least 20, 30, 45, or 60 minutes; or the modified siRNA in the nucleic acid-lipid particle is not substantially degraded after incubating the particle in serum at 37 ℃ for at least 30, 45, or 60 minutes.

In some embodiments, the modified siRNA is completely encapsulated within the nucleic acid-lipid particle. In other embodiments, the modified siRNA is complexed with a lipid moiety of the particle.

The invention also provides pharmaceutical compositions comprising the nucleic acid-lipid particles described herein and a pharmaceutically acceptable carrier.

In another aspect, the modified siRNA described herein is used in a method of silencing expression of a target sequence. In particular, it is an object of the present invention to provide in vitro and in vivo methods for treating a disease or disorder in a mammal by down-regulating or silencing the transcription and/or translation of a target gene of interest. In one embodiment, the invention provides a method of introducing an siRNA that silences expression (e.g., mRNA and/or protein levels) of a target sequence into a cell by contacting the cell with a modified siRNA described herein. In another embodiment, the invention provides a method of delivering an siRNA that silences expression of a target sequence in vivo by administering a modified siRNA described herein to a mammal. Administration of the modified siRNA may be by any route known in the art, such as, for example, oral, intranasal, intravenous, intraperitoneal, intramuscular, intraarticular, intralesional, intratracheal, subcutaneous, or intradermal.

In these methods, the modified siRNA is typically formulated with a carrier system, and the carrier system comprising the modified siRNA is administered to a mammal in need of such treatment. Examples of vector systems suitable for use in the present invention include, but are not limited to, nucleic acid-lipid particles, liposomes, micelles, viral particles, nucleic acid complexes (e.g., lipid nucleic acid complexes, polymer-nucleic acid complexes, etc.), and mixtures thereof. The vector system can include at least one, two, three, four, five, six, seven, eight, nine, ten, or more modified siRNA molecules described herein. Alternatively, cells are removed from a mammal, such as a human, the modified siRNA is delivered in vitro, and the cells are then administered to the mammal, such as by injection.

In some embodiments, the modified siRNA is in a nucleic acid-lipid particle comprising the modified siRNA, a cationic lipid, and a non-cationic lipid. Preferably, the modified siRNA is in a nucleic acid-lipid particle comprising the modified siRNA, a cationic lipid, a non-cationic lipid, and a conjugated lipid that inhibits aggregation of particles. A therapeutically effective amount of the nucleic acid-lipid particle can be administered to a mammalian subject (e.g., a rodent, such as a mouse, or a primate, such as a human, chimpanzee, or monkey).

In another embodiment, at least about 1%, 2%, 4%, 6%, 8%, or 10% of the total administered dose of the nucleic acid-lipid particle is present in the plasma at about 1, 2, 4, 6, 8, 12, 16, 18, or 24 hours after administration. In another embodiment, greater than about 20%, 30%, or 40% or as much as about 60%, 70%, or 80% of the total administered dose of the nucleic acid-lipid particle is present in the plasma at about 1, 4, 6, 8, 10, 12, 20, or 24 hours after administration. In one embodiment, the effect of the modified siRNA at a site proximal or distal to the site of administration (e.g., down-regulation of the target sequence) is detectable at about 12, 24, 48, 72, or 96 hours, or at about 6, 8, 10, 12, 14, 16, 18, 19, 20, 22, 24, 26, or 28 days after administration of the nucleic acid-lipid particle. In another embodiment, downregulation of target sequence expression is detectable at about 12, 24, 48, 72, or 96 hours after administration, or at about 6, 8, 10, 12, 14, 16, 18, 19, 20, 22, 24, 26, or 28 days after administration. In certain instances, the down-regulation of gene sequence expression is detected by measuring mRNA or protein levels in a biological sample from the mammal.

The nucleic acid-lipid particles are suitable for intravenous nucleic acid delivery because they are stable in the circulation, are of a size required to result in pharmacokinetic behavior into extravascular sites, and target cell populations. The invention also provides pharmaceutical compositions comprising the nucleic acid-lipid particles.

In another aspect, the present invention provides a method for modifying an siRNA with immunostimulatory properties, comprising: (a) providing an unmodified siRNA sequence that is capable of silencing expression of a target sequence and that comprises a double-stranded sequence of about 15 to about 60 nucleotides in length (e.g., about 15 to 60, 15 to 50, 15 to 40, 15 to 30, 15 to 25, or 19 to 25 nucleotides in length); and (b) modifying the siRNA by replacing at least one nucleotide in the sense or antisense strand with a modified nucleotide, thereby producing a modified siRNA that is less immunostimulatory than the unmodified siRNA sequence and capable of silencing expression of the target sequence.

In some embodiments, the modified nucleotides include, but are not limited to, 2 'OMe nucleotides, 2' F nucleotides, 2 '-deoxynucleotides, 2' OMe nucleotides, LNA nucleotides, and mixtures thereof. In preferred embodiments, the modified nucleotides include 2 'OMe nucleotides (e.g., 2' OMe purine and/or pyrimidine nucleotides), such as, for example, 2 'OMe-guanosine nucleotides, 2' OMe-uridine nucleotides, 2 'OMe-adenosine nucleotides, 2' OMe-cytosine nucleotides, and mixtures thereof. In some cases, the modified nucleotide is not a 2' OMe-cytosine nucleotide.

In certain instances, the unmodified siRNA sequence includes at least one, two, three, four, five, six, seven, or more 5 '-GU-3' motifs. The 5 '-GU-3' motif can be in the sense strand, the antisense strand, or both strands of the unmodified siRNA sequence. Preferably, at least one nucleotide in the 5 '-GU-3' motif is substituted with a modified nucleotide. As a non-limiting example, both nucleotides in the 5 '-GU-3' motif may be substituted with modified nucleotides.

In some embodiments, the method further comprises: (c) the modified siRNA is demonstrated to be less immunostimulatory by contacting the modified siRNA with a mammalian effector cell under conditions suitable for the effector cell to produce a detectable immune response. The mammalian effector cell may be from a first immunized mammal (i.e., a mammal that has not previously been contacted with the gene product of the siRNA sequence). The mammalian effector cells can be, for example, Peripheral Blood Mononuclear Cells (PBMCs), macrophages, and the like. The detectable immune response may include production of cytokines or growth factors, such as, for example, TNF- α, IFN- α, IFN- β, IFN- γ, IL-6, IL-12, or combinations thereof.

In a related aspect, the invention provides methods for identifying and modifying sirnas having immunostimulatory properties. The method comprises the following steps: (a) contacting the unmodified siRNA sequence with a mammalian effector cell under conditions suitable for the effector cell to produce a detectable immune response; (b) identifying the unmodified siRNA sequence as an immunostimulatory siRNA by the presence of a detectable immune response in the effector cell; and (c) modifying the immunostimulatory siRNA by substituting at least one nucleotide with a modified nucleotide, thereby producing a modified siRNA sequence that is less immunostimulatory than the unmodified siRNA sequence.

In certain embodiments, the modified nucleotides include, but are not limited to, 2 'OMe nucleotides, 2' F nucleotides, 2 '-deoxynucleotides, 2' OMe nucleotides, LNA nucleotides, and mixtures thereof. In preferred embodiments, the modified nucleotides include 2 'OMe nucleotides (e.g., 2' OMe purine and/or pyrimidine nucleotides), such as, for example, 2 'OMe-guanosine nucleotides, 2' OMe-uridine nucleotides, 2 'OMe-adenosine nucleotides, 2' OMe-cytosine nucleotides, and mixtures thereof. In some cases, the modified nucleotide is not a 2' OMe-cytosine nucleotide.

In certain instances, the unmodified siRNA sequence includes at least one, two, three, four, five, six, seven, or more 5 '-GU-3' motifs. The 5 '-GU-3' motif can be in the sense strand, the antisense strand, or both strands of the unmodified siRNA sequence. Preferably, at least one nucleotide in the 5 '-GU-3' motif is substituted with a modified nucleotide. As a non-limiting example, both nucleotides in the 5 '-GU-3' motif are substituted with modified nucleotides.

In one embodiment, the mammalian effector cells are Peripheral Blood Mononuclear Cells (PBMCs), macrophages, or the like. In another embodiment, the detectable immune response includes the production of cytokines or growth factors, such as, for example, TNF- α, IFN- α, IFN- β, IFN- γ, IL-6, IL-12, or a combination thereof.

In another aspect, the invention provides an isolated nucleic acid molecule comprising the modified sequences listed in tables 1 and 2. The modified sequence may also include its complementary strand, thereby producing a modified siRNA duplex. In a related aspect, the invention provides an isolated nucleic acid molecule comprising a modified siRNA duplex set forth in tables 3, 5 and 6.

Other features, objects, and advantages of the invention, as well as preferred embodiments thereof, will be apparent from the description, examples, and claims that follow.

Brief Description of Drawings

FIG. 1 illustrates data showing that 2' OMe modification in human PBMC abrogates immunostimulatory ssRNA-mediated interferon induction. Liposomal encapsulated, unmodified (native) and 2' OMeU-, G-, or GU-modified ssRNA representing sense (S) or Antisense (AS) strands of (A) β -gal and (B) ApoB siRNA were incubated with PBMC at increasing concentrations (5-135 nM). The sequences are shown in table 1. IFN- α was assayed in culture supernatants at 24 hours. Values are the mean + SD of triplicate cultures.

FIG. 2 illustrates data showing that immunostimulatory of ssRNA and siRNA complexed with Polyethylenimine (PEI) was abolished by 2' OMe modification. Interferon- α reactions from human PBMC cultures treated with PEI complexed (a) native β -gal sense (S), Antisense (AS), and 2 'OMe-modified sense ssRNAs with (B) native, 2' OMe GU (S), and Ac (AS) modified β -gal control siRNA duplexes. RNA was added at a final concentration of 225nM and IFN- α was determined in culture supernatants after 16 hours. The RNA sequences are shown in table 1.

Figure 3 illustrates data demonstrating that selected 2' OMe modifications to siRNA duplexes in human PBMC abolished cytokine induction. (A, B) IFN- α and (C) TNF- α induction from human PBMC cultured with increasing concentrations (25-675nM) of encapsulated (A) β -gal or (B, C) ApoB or ApoB mismatched siRNA. Cytokine responses to unmodified (native) siRNAs were compared to duplexes containing 2' OMe U, G, C, or a residues in the sense (S) or Antisense (AS) strand AS indicated (see, siRNA sequences of table 1). Secreted cytokines were measured after 24 hours of culture. Values are the mean + SD of triplicate cultures.

Figure 4 illustrates data showing that selective 2' OMe modification of siRNA duplexes abolished cytokine induction in vivo. (A, B) beta-gal, (C, D) ApoB, (E) ApoB mismatch, or (F) vFLIP siRNA 6 hours after intravenous administration, (A, C, E, F) serum IFN-alpha and (B, D) TNF-alpha and IL-6 levels. Responses to unmodified (native) siRNAs were compared to duplexes containing 2' OMe U, G, or C residues in the sense (S) or Antisense (AS) strands AS indicated (see, siRNA sequences of table 1). All mice received 40 μ g of encapsulated siRNA. Values are mean + SD from 3-4 animals. Lower quantitation levels were IFN-. alpha.75 pg/mL, TNF-. alpha.30 pg/mL, and IL-660 pg/mL.

FIG. 5 illustrates data showing that cytokine response to siRNA in vivo was abolished by selective binding of 2' OMe-uridine to the sense strand of siRNA. Serum interferon- α levels in mice were determined 6 hours after intravenous administration of SNALPs containing siRNA-targeted (a) influenza nucleoprotein (NP 1496) and polymerase (PA2087) or (B) cyclophilin B (cypb). The response to native, unmodified siRNA was compared to the response to 2' OMe U (S) modified duplexes. The sequences are provided in table 2.

FIG. 6 illustrates data showing the generation of non-inflammatory β -gal siRNAs that retain full RNAi activity. Immunostimulatory β -gal 728 sirnas are chemically modified by incorporating 2' OMe-uridine (U), guanosine (G), cytidine (C), or adenosine (a) into the sense strand (S) of the siRNA duplex. (A) Interferon-alpha induction in mice 6 hours after administration of 20 μ g siRNA encapsulated in SNALPs. (B) The same β -gal 728SNALP preparation of RNA interference activity in vitro. RNAi assays were performed in Neuro2A cells stably transfected with the LacZ gene of e.coli (e.coil). β -gal activity was assessed 48 hours after exposure to SNALPs and mean values were expressed relative to PBS-treated cells. The SNALPs used in these studies included lipid cholesterol: DSPC: DLinDMA: PEG-C-DMA in a molar ratio of 48: 10: 40: 2, and had particle sizes in the 80-90nm range in diameter. The RNA sequences are provided in table 2.

FIG. 7 illustrates data showing the generation of non-inflammatory luciferase siRNA that retained full RNAi activity. Immunostimulatory luciferase (Luc) siRNAs were chemically modified by incorporating 2' OMe-uridine (U) into the sense strand (S) of the siRNA duplex. (A) Interferon-alpha induction in mice 6 hours after administration of 20 μ g siRNA encapsulated in SNALPs. (B) In vitro RNA interference activity of the LucSNALP formulation. RNAi assays were performed in Neuro2A cells stably transfected with firefly luciferase. Luciferase activity was assessed 48 hours after exposure to SNALPs and mean values were expressed relative to PBS-treated cells. The SNALPs used in these studies included lipid cholesterol: DSPC: DLinDMA: PEG-C-DMA in a molar ratio of 48: 10: 40: 2, and had particle sizes in the 75-85nm range in diameter. The RNA sequences are provided in table 2.

FIG. 8 illustrates data showing in vitro silencing of ApoB expression by 2' OMe-modified siRNA. HepG2 cells were treated with encapsulated ApoB or mismatched siRNA at the indicated concentrations (0-45 nM). Unmodified (native) ApoB sirnas were compared to ApoB duplexes containing 2' OMe U, G, or C residues in the sense strand (S) or GU motifs, U or C residues in the Antisense (AS) strand AS indicated (see, modified siRNA sequences of table 1). Unmodified and 2' OMe-U (S) ApoB mismatched siRNA served as control duplexes. After 48 hours, ApoB protein in the culture supernatant was measured by ELISA. ApoB levels are expressed as% PBS-treated control cultures. Each value is derived from the average of duplicate cultures and represents 3 independent experiments.

Figure 9 illustrates data showing that encapsulation of siRNA in lipid particles protects against serum nuclease degradation. Unmodified naked (upper) or SNALP-encapsulated (middle) ApoB siRNA was incubated in 50% mouse serum at 37 ℃. Duplex integrity was assessed by non-denaturing PAGE analysis at the indicated time points. Addition of Triton-X to disrupt the integrity of the lipid particles (bottom) restored the sensitivity of siRNA nucleases.

Figure 10 illustrates data showing silencing of ApoB expression in vivo without inactivating the innate immune response. (A) - (E) in vivo effects following intravenous administration of encapsulated ApoB or mismatched siRNA in mice. On days 0, 1 and 2, mice were treated with encapsulated unmodified, 2 'OMe U (S), or gu (as) -modified ApoB, and unmodified or 2' OMe U (S) -modified mismatched siRNA at 5 mg/kg/day. (A) Daily change in body weight (% body weight on day 0) of mice treated with ApoB (solid labeled) and mismatch (open labeled) siRNA over a 4 day study period. (B) Serum IFN- α from the test blood samples 6 hours after the initial treatment. ND is not detected; lower quantitation level 75 pg/ml. (C) ApoB mRNA levels in the liver. (D) ApoB protein in serum. (E) Serum cholesterol (mM) 2 days after the last siRNA treatment. ApoB levels were expressed as% ApoB mRNA or ApoB protein compared to PBS-treated animals. All values are mean + SD of 5 animals. All data are representative of 2 independent experiments.

Fig. 11 illustrates data showing silencing activity of various unmodified and chemically modified ApoB sirnas. SNALP-formulated siRNA silencing efficacy was measured 7 days after the end of IV treatment at a daily siRNA dose of 2mg/kg for three consecutive days. ApoB silencing activity was measured with respect to the reduction in plasma ApoB-100 protein levels compared to PBS-treated controls. Each bar represents the mean (n ═ 5) ± Standard Deviation (SD) of the group.

Fig. 12 illustrates data showing immunostimulatory properties of various unmodified and chemically modified ApoB sirnas. The immunostimulatory properties of each siRNA, characterized by cytokine release, were measured 6 hours after the initial IV administration of SNALP-formulated siRNA. Plasma concentrations of the cytokine interferon-alpha were measured using ELISA. For treatments that elicited significant responses (values in excess of 200pg/mL), plasma samples were diluted 10-fold and each animal was analyzed independently, so that the bars in the figure represent group mean (n-5) ± Standard Deviation (SD). For treatments that caused very small reactions (values less than 200pg/mL), samples were pooled together and assayed in 4-fold dilutions.

FIG. 13 illustrates data showing that selective 2' OMe modification of Eg 52263 siRNA retains RNAi activity in human HeLa cells.

Figure 14 illustrates data showing that selective 2' OMe modification to Eg 52263 siRNA retained RNAi activity in mouse Neuro2A cells.

Figure 15 illustrates data showing that selective 2' OMe modification to Eg 52263 siRNA abolished interferon induction associated with systemic administration of natural duplexes.

Fig. 16 illustrates data necessary to show that selective 2' OMe modification of both chains of Eg 52263 siRNA is a complete abrogation of antibody response to the PEG component of SNALP delivery vehicle.

Figure 17 illustrates data showing that NP 411, NP929, NP 1116, and NP 1496 sirnas comprising selective 2' OMe modifications to the sense strand maintain influenza knockdown activity in MDCK cells in vitro. Figure 17A shows influenza virus infection of MDCK cells at 48 hours after 5 hours pretreatment with modified or unmodified siRNA. Figure 17B shows the percentage of HA relative to control virus only at 48 hours after infection of MDCK cells with a 1: 800 dilution of influenza virus and transfection with 2 μ g/ml modified or unmodified siRNA.

Figure 18 illustrates data showing that selective 2' OMe modification of the sense strand of NP 1496 siRNA did not adversely affect influenza knockdown activity when compared to the unmodified negative or control sequences.

Figure 19 illustrates data showing that combinations of 2' OMe-modified sirnas provide enhanced influenza knockdown in MDCK cells in vitro. Fig. 19A shows influenza virus infection of 48 h MDCK cells after 5 h pretreatment with various combinations of modified sirnas. Figure 17B shows the percentage of HA relative to the control virus only, at 48 hours when MDCK cells were infected with a 1: 800 dilution of influenza virus and transfected with 2 μ g/ml modified siRNA.

FIG. 20 illustrates data showing that selective 2' OMe modification of NP 1496 siRNA eliminates interferon induction in an in vitro cell culture system.

Figure 21 illustrates data showing selective 2' OMe modification of NP 1496 siRNA eliminates interferon induction associated with systemic administration of natural duplexes complexed with cationic polymer Polyethylenimine (PEI).

Detailed Description

I. Introduction to

Targeted silencing of disease-associated genes by synthetic sirnas holds promise as a new therapeutic strategy. However, unmodified siRNA may be immunostimulatory, e.g., stimulating a potent inflammatory response from innate immune cells, particularly when associated with a delivery vehicle that promotes intracellular uptake. This represents a significant obstacle to the development of siRNA therapies due to the toxicity and off-target gene effects associated with the inflammatory response. The present invention overcomes these limitations by reducing or completely eliminating the immune response to synthetic siRNA by using selective incorporation of modified nucleotides such as 2 '-O-methyl (2' OMe) uridine and/or guanosine nucleotides into one or both strands of the siRNA duplex. In particular, immunostimulatory sirnas that retain complete gene silencing activity can be readily produced by incorporating selective 2' OMe modifications in the duplex region of the siRNA duplex. As a non-limiting example, 2' OMe-modified siRNA targeting apolipoprotein b (ApoB) can modulate the vigorous silencing of its target mRNA, causing significant reductions in serum ApoB and cholesterol, when encapsulated in an effective systemic delivery vehicle such as a nucleic acid-lipid particle. This is achieved at therapeutically feasible siRNA doses without cytokine induction, toxicity and off-target effects associated with the use of unmodified siRNA. Advantageously, the methods of siRNA design and delivery described herein are broadly applicable and improve synthetic siRNA to a wide range of therapeutic areas.

Accordingly, the present invention provides chemically synthesized modified siRNA molecules, and methods of silencing target gene expression using such siRNA molecules. The invention also provides nucleic acid-lipid particles comprising the modified siRNA molecules described herein, a cationic lipid, and a non-cationic lipid, which may further comprise a conjugated lipid that inhibits aggregation of the particles. The invention further provides methods of silencing gene expression by administering to a mammalian subject a modified siRNA molecule described herein. The invention also provides methods for identifying and/or modifying sirnas having immunostimulatory properties.

Definition of

As used herein, the following terms have the meanings ascribed to them unless otherwise indicated.

The term "interfering RNA" or "RNAi" or "interfering RNA sequence" refers to a double-stranded RNA (i.e., a duplex RNA) that is capable of reducing or inhibiting target gene expression (i.e., by modulating degradation of mRNAs complementary to the sequence of the interfering RNA) when the interfering RNA is in the same cell as the target gene. Thus, interfering RNA refers to double-stranded RNA formed by two complementary strands or by a single, self-complementary strand. The interfering RNA may have substantial or complete identity to the target gene, or may include a mismatch region (i.e., a mismatch motif). The sequence of the interfering RNA may be identical to the full-length target gene or a subsequence thereof.

Interfering RNAs include "small-interfering RNAs" or "siRNAs", e.g., interfering RNAs of about 15-60, 15-50, or 15-40 (duplex) nucleotides in length, more typically about 15-30, 15-25, or 19-25 (duplex) nucleotides in length, and preferably about 20-24, 21-22, or 21-23 (duplex) nucleotides in length (e.g., each complementary sequence of a double-stranded siRNA is 15-60, 15-50, 15-40, 15-30, 15-25, or 19-25 nucleotides in length, preferably about 20-24, 21-22, or 21-23 nucleotides in length, and the double-stranded siRNA is about 15-60, 15-50, 15-40, 15-30, 15-25, or 19-25 base pairs in length, preferably about 20-24, 21-22 or 21-23 base pairs in length). The siRNA duplex may comprise a 3 'overhang of about 1 to about 4 nucleotides, or about 2 to about 3 nucleotides, and a 5' -phosphate end. Examples of sirnas include, but are not limited to, double-stranded polynucleotide molecules assembled from two single-stranded molecules, wherein one strand is the sense strand and the other strand is the complementary antisense strand; a double-stranded polynucleotide molecule assembled from single-stranded molecules, wherein the sense and antisense regions are linked by a nucleic acid-based or non-nucleic acid-based linker; a double-stranded polynucleotide molecule with a hairpin secondary structure having self-complementary sense and antisense regions; and a circular single stranded polynucleotide molecule with two or more loop structures and a stem with self-complementary sense and antisense regions, wherein the circular polynucleotide can be processed in vivo or in vitro to produce an active double stranded siRNA molecule.

Preferably, the siRNA is chemically synthesized. siRNA can also be generated by cleaving longer dsRNA (e.g., dsRNA longer than about 25 nucleotides) with E.coli RNase III or dicer. These enzymes treat dsRNA to biologically active siRNA (see, e.g., Yang et al, Proc. Natl. Acad. Sci. USA (Proc. Natl. Acad. Sci. USA) 99: 9942-. Preferably, the dsRNA is at least 50 nucleotides to about 100, 200, 300, 400 or 500 nucleotides in length. The dsRNA may be as long as 1000, 1500, 2000, 5000 nucleotides or longer in length. The dsRNA may encode a complete gene transcript or a partial gene transcript. In some cases, the siRNA may be encoded by a plasmid (e.g., transcribed as a sequence that automatically folds into a duplex with a hairpin loop).

As used herein, the term "mismatch motif or" mismatch region "refers to a portion of an siRNA sequence that does not have 100% complementarity to its target sequence. The siRNA can have at least one, two, three, four, five, six, or more mismatch regions. The mismatch regions may be contiguous or may be separated by 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12 or more nucleotides. The mismatch motif or region may comprise a single nucleotide or may comprise two, three, four, five or more nucleotides.

An "effective amount" or "therapeutically effective amount" of an siRNA is an amount sufficient to produce a desired effect, e.g., inhibition of expression of the target sequence as compared to the normal expression level detected in the absence of the siRNA. Inhibition of expression of the target gene or target sequence is obtained when the value obtained using siRNA relative to a control is about 90%, 85%, 80%, 75%, 70%, 65%, 60%, 55%, 50%, 45%, 40%, 35%, 30%, 25%, 20%, 15%, 10%, 5%, or 0%. Suitable assays for measuring expression of a target gene or target sequence include, for example, assaying protein or mRNA levels using techniques known to those skilled in the art such as dot blotting, northern blotting, in situ hybridization, ELISA, immunoprecipitation, enzyme function, and phenotypic assays known to those skilled in the art.

By siRNA "reduce", "reducing", or "reducing" an immune response is intended to refer to a detectable reduction in an immune response to an siRNA (e.g., a modified siRNA). The amount of reduced immune response by the modified siRNA can be determined relative to the level of immune response in the presence of the unmodified siRNA. The detectable reduction may be about 5%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, 100%, or more less than the immune response detected in the presence of the unmodified siRNA. Reduction in an immune response to an siRNA is typically measured by a reduction in cytokine production (e.g., IFN γ, IFN α, TNF α, IL-6, or IL-12) in effector cells in vitro or in the serum of a mammalian subject following administration of the siRNA.

As used herein, the term "effector cell" refers to a cell, preferably a mammalian cell, that produces a detectable immune response when contacted with an immunostimulatory siRNA. Exemplary effector cells include, for example, dendritic cells, macrophages, Peripheral Blood Mononuclear Cells (PBMCs), spleen cells, and the like. Detectable immune responses include, for example, the production of cytokines or growth factors such as TNF- α, IFN- α, IFN- β, IFN- γ, IL-1, IL-2, IL-3, IL-4, IL-5, IL-6, IL-10, IL-12, IL-13, TGF, and combinations thereof.

"substantial identity" refers to a sequence that hybridizes to a reference sequence under stringent conditions, or a sequence that has a specified percentage of identity over a specified region of the reference sequence.

In the context of two or more nucleic acids, the term "substantially identical" or "substantial identity" refers to two or more sequences or subsequences that are the same or have a specified percentage of nucleotides that are the same (i.e., at least about 60%, preferably at least about 65%, 70%, 75%, 80%, 85%, 90%, or 95% identity over a particular region) when compared and aligned for maximum correspondence over a comparison window, or designated region, as measured using one of the following sequence comparison algorithms or by manual comparison and visual inspection. This definition also similarly refers to the complement of the sequence, when the context indicates. Preferably, the substantial identity exists over a region that is at least about 5, 10, 15, 20, 25, 30, 35, 40, 45, 50, 55, or 60 nucleotides in length.

For sequence comparison, typically, one sequence serves as a reference sequence to which test sequences are compared. When using a sequence comparison algorithm, test and reference sequences are input into a computer, subsequence coordinates are designated, if necessary, and sequence algorithm program parameters are designated. Default program parameters may be used, or alternative parameters may be specified. The sequence comparison algorithm then calculates the percent sequence identity of the test sequence relative to the reference sequence based on the program parameters.

As used herein, a "comparison window" includes a reference to a fragment of any one of a number of consecutive positions selected from the group consisting of about 5 to about 60, typically about 10 to about 45, more typically about 15 to about 30, wherein, after two sequences are optimally aligned, a sequence can be compared to a reference sequence of the same number of consecutive positions. Methods of sequence alignment for comparison are well known in the art. Optimal alignment of sequences for comparison can be determined by Smith and Waterman, adv.appl.math. (advanced application mathematics) 2: 482(1981) by Needleman and Wunsch, j.mol.blol. (journal of molecular biology) 48: 443(1970) by Pearson and Lipman, proc.natl.acad.sci.usa (proceedings of the national academy of sciences usa) 85: 2444(1988), by computerized execution of these algorithms (GAP, BESTFIT, FASTA, and TFASTA in the Wisconsin Genetics Software Package, genetic computing group, 575Science Dr., Madison, Wis.), or by manual alignment and visual observation (see, e.g., Current Protocols in Molecular Biology, Austebel et al, eds. (1995 supplement)).

Preferred examples of algorithms suitable for determining percent sequence identity and sequence similarity are the BLAST and BLAST 2.0 algorithms described in Altschul et al, nuc.acids Res. (nucleic acid research) 25: 3389-3402(1977) and Altschul et al, J.mol.biol. (J.Mol. mol. biol.) 215: 403-410(1990). Percent sequence identity for nucleic acids of the invention is determined using BLAST and BLAST 2.0, with the parameters described herein. Software for performing BLAST analysis is publicly available through the National Center for Biotechnology Information, http:// www.ncbi.nlm.nih.gov /).

The BLAST algorithm also performs a statistical analysis of the similarity between two sequences (see, e.g., Karlin and Altschul, Proc. Natl. Acad. Sci. USA (Proc. Natl. Acad. Sci.) 90: 5873-5787 (1993)). One measure of similarity provided by the BLAST algorithm is the smallest sum likelihood (P (N)), which provides an indication of likelihood by which a match between two nucleotide sequences would occur by chance. For example, a nucleic acid is considered similar to a reference sequence if the minimum sum likelihood in a comparison of the test nucleic acid to the reference nucleic acid is less than about 0.2, more preferably less than about 0.01, and most preferably less than about 0.001.

As used herein, the term "nucleic acid" refers to a polymer comprising at least two deoxyribonucleotides or ribonucleotides in either single or double stranded form, and includes DNA and RNA. The DNA may be, for example, in the form of antisense molecules, plasmid DNA, pre-condensed DNA, PCR products, vectors (P1, PAC, BAC, YAC, artificial chromosomes), expression cassettes, chimeric sequences, chromosomal DNA, or derivatives and combinations of these groups. The RNA can be in the form of siRNA, mRNA, tRNA, rRNA, tRNA, vRNA, and combinations thereof. Nucleic acids include nucleic acids containing known nucleotide analogs or modified backbone residues or linkages, which are synthetic, naturally occurring, and not naturally occurring, and which have similar binding properties as the reference nucleic acid. Examples of such analogs include, but are not limited to, phosphorothioate, phosphoramidate, methylphosphonate, chiral-methylphosphonate, 2' -O-methyl ribonucleotides, and peptide-nucleic acids (PNAs). Unless specifically limited, the term includes nucleic acids containing known analogs of natural nucleotides that have similar binding properties to the reference nucleic acid. Unless otherwise indicated, a particular nucleic acid sequence also potentially includes conservatively modified variants thereof (e.g., degenerate codon substitutions), alleles, orthologs, SNPs and complementary sequences and the sequences specifically indicated. In particular, degenerate codon substitutions may be obtained by generating sequences in which the third position of one or more selected (or all) codons is substituted with mixed base and/or deoxyinosine residues (Batzer et al, Nucleic Acid Res., 19: 5081 (1991); Ohtsuka et al, J biol. chem. (J. chem., 260.2605-2608 (1985); Rossolini et al, mol. cell. probes, 8: 91-98 (1994)). "nucleotide" includes sugar Deoxyribose (DNA) or Ribose (RNA), base, and phosphate groups. Nucleotides are linked together by phosphate groups. "base" includes purines and pyrimidines, which further include the natural compounds adenine, thymine, guanine, cytosine, uracil, inosine, and natural analogs of purines and pyrimidines as well as synthetic derivatives including, but not limited to, modifications to place new reactive groups such as, but not limited to, amines, alcohols, thiols, carboxylates, and alkyl halides.

The term "gene" refers to a nucleic acid (e.g., DNA or RNA) sequence that includes a partial length or full length coding sequence necessary for the production of a polypeptide or precursor polypeptide.

As used herein, "gene product" refers to the product or polypeptide of a gene, such as an RNA transcript.

The term "lipid" refers to a group of organic compounds that include, but are not limited to, esters of fatty acids and are characterized as insoluble in water, but soluble in many organic solvents. They are generally divided into at least three categories: (1) "simple lipids" which include fats and oils as well as waxes; (2) "compound lipids" which include phospholipids and glycolipids; and (3) "derivatized lipids", such as steroids.

"lipid vesicle" refers to any lipid composition useful for delivery of compounds such as siRNA, including, but not limited to, liposomes in which a volume of water is encapsulated by an amphiphilic lipid bilayer; or wherein the lipid coat comprises an interior of macromolecular components, such as plasmids comprising interfering RNA sequences, having a reduced aqueous interior; or a lipid aggregate or micelle, wherein the encapsulated ingredient is contained in a relatively chaotic lipid mixture. The term lipid vesicle includes any of a variety of lipid-based carrier systems including, but not limited to, SPLPs, pSPLPs, SNALPs, liposomes, micelles, virosomes, lipid-nucleic acid complexes, and mixtures thereof.

As used herein, "encapsulated lipid" may refer to a lipid formulation that provides a compound, such as an siRNA, with sufficient encapsulation, partial encapsulation, or both. In a preferred embodiment, the nucleic acid is substantially encapsulated in the lipid formulation (e.g., to form SPLP, pSPLP, SNALP, or other nucleic acid-lipid particles).

As used herein, the term "SNALP" refers to a stable nucleic acid-lipid particle, including SPLP. SNALP represent vesicles that coat lipids of a reduced aqueous interior that includes nucleic acids (e.g., siRNA, ssDNA, dsDNA, ssRNA, microrna (miRNA), short hairpin RNA (shRNA), dsRNA, or plasmids, including plasmids from which interfering RNAs are transcribed). As used herein, the term "SPLP" refers to a nucleic acid-lipid particle that includes a nucleic acid (e.g., a plasmid) encapsulated within a lipid vesicle. SNALPs and SPLPs typically comprise a cationic lipid, a non-cationic lipid, and a lipid that prevents particle aggregation (e.g., a PEG-lipid conjugate). SNALPs and SPLPs are particularly useful for systemic applications because they exhibit extended circulatory life following intravenous (i.v.) injection, accumulate at remote sites (e.g., sites physically separate from the site of administration), and can regulate expression of transfected genes at these remote sites. SPLPs include "pSPLP" which includes an encapsulated flocculant-nucleic acid complex described in PCT publication No. WO 00/03683.

The nucleic acid-lipid particles of the invention typically have an average diameter of about 50nm to about 150nm, more typically about 60 to about 130nm, more typically about 70nm to about 110nm, most typically about 70nm to about 90nm, and are substantially non-toxic. In addition, when present in the nucleic acid-lipid particle of the present invention, the nucleic acid is resistant to degradation by nucleases in aqueous solution. For example, nucleic acid-lipid particles and methods for their preparation are described in U.S. patent nos. 5,976,567; 5,981,501, respectively; 6,534,484, respectively; 6,586,410, respectively; and 6,815,432; and PCT publication No. WO 96/40964.

The term "vesicle-forming lipid" is intended to include any amphiphilic lipid having a hydrophobic portion and a polar head group, and which can spontaneously form bilayer vesicles in water by itself, exemplified by a majority of phospholipids.

The term "lipid with vesicles" is intended to include any amphipathic lipid that is stable in combination with other amphipathic lipids incorporated into a lipid bilayer, with its hydrophobic portion in contact with the hydrophobic region of the inner, bilayer membrane, and its polar head group portion facing the outer, polar surface of the membrane. Lipids that employ vesicles include lipids that are capable of independently adopting a non-lamellar phase, which are also capable of adopting a bilayer structure in the presence of a bilayer stabilizing component. A typical example is Dioleoylphosphatidylethanolamine (DOPE). Bilayer stabilizing components include, but are not limited to, conjugated lipids that inhibit aggregation of nucleic acid-lipid particles, polyamide oligomers (e.g., ATTA-lipid derivatives), peptides, proteins, detergents, lipid-derivatives, PEG-lipid derivatives such as PEG coupled to dialkoxypropyl groups, PEG coupled to diacylglycerols, PEG coupled to phosphatidyl-ethanolamine, PEG conjugated to ceramides (see, e.g., U.S. Pat. No. 5,885,613), cationic PEG lipids, and mixtures thereof. PEG may be conjugated directly to the lipid, or may be linked to the lipid through a linker moiety. Any linker moiety suitable for coupling PEG to lipids can be used, including, for example, non-ester containing linker moieties, and ester containing linker moieties.

The term "amphipathic lipid" moiety refers to any suitable material wherein the hydrophobic portion of the lipid material faces the hydrophobic phase and the hydrophilic portion faces the aqueous phase. Amphiphilic lipids are generally the major component of lipid vesicles. The hydrophilic nature results from the presence of polar or charged groups such as carbohydrates, phosphates, carboxyl, sulfato, amino, mercapto, nitro, hydroxyl and other similar groups. Hydrophobicity may be imparted by the inclusion of non-polar groups including, but not limited to, long chain saturated and unsaturated aliphatic hydrocarbon groups and such groups substituted with one or more aromatic, alicyclic, or heterocyclic groups. Examples of amphiphilic compounds include, but are not limited to, phospholipids, aminolipids, and sphingolipids. Representative examples of phospholipids include, but are not limited to, phosphatidylcholine, phosphatidylethanolamine, phosphatidylserine, phosphatidylinosine, phosphatidic acid, palmitoyl oleoyl phosphatidylcholine, lysophosphatidylcholine, lysophosphatidylethanolamine, dipalmitoyl phosphatidylcholine, dioleoyl phosphatidylcholine, distearoylphosphatidylcholine or dilinoleoylphosphatidylcholine. Other compounds lacking phosphorus, such as sphingomyelin, the glycosphingolipid family, diacylglycerols, and β -acyloxyacids, are also in the group referred to as amphiphilic lipids. In addition, the amphiphilic lipids described above may be mixed with other lipids, including triglycerides and sterols.

The term "neutral lipid" refers to any of a number of lipid species that exist in the form of uncharged or neutral zwitterions at a selected pH. At physiological pH, such lipids include, for example, diacylphosphatidylcholine, diacylphosphatidylethanolamine, ceramide, sphingomyelin, cephalin, cholesterol, cerebrosides, and diacylglycerol.

The term "non-cationic lipid" refers to any neutral lipid as described above as well as anionic lipids.

The term "anionic lipid" refers to any lipid that is negatively charged at physiological pH. These lipids include, but are not limited to, phosphatidylglycerol, cardiolipin, diacylphosphatidylserine, diacylphosphatidic acid, N-lauroyl phosphatidylethanolamine, N-succinyl phosphatidylethanolamine, N-glutaryl phosphatidylethanolamine, lysyl phosphatidylglycerol, Palmitoyl Oleoyl Phosphatidylglycerol (POPG), and other anionic modifying groups attached to neutral lipids.

The term "cationic lipid" refers to any of a number of lipid species that carry a net positive charge at a selected pH, such as physiological pH (e.g., pH of about 7.0). It has surprisingly been found that cationic lipids comprising an alkyl chain having multiple sites of unsaturation, e.g. at least two or three sites of unsaturation, are particularly useful for forming nucleic acid-lipid particles with increased membrane fluidity. A number of cationic lipids and related analogs also useful in the present invention have been described in U.S. patent publication nos. 20060083780; U.S. patent nos. 5,208,036; 5,264,618, respectively; 5,279,833, respectively; 5,283,185, respectively; 5,753,613, respectively; and 5,785,992; and PCT publication No. WO 96/10390. Examples of cationic lipids include, but are not limited to, N-dioleyl-N, N-dimethylammonium chloride (DODAC); dioctadecyldimethylammonium (DODMA); distearyldimethylammonium (DSDMA); n- (1- (2, 3-dioleyloxy) propyl) -N, N, N-trimethylammonium chloride (DOTMA); n, N-distearyl-N, N-dimethylammonium bromide (DDAB); n- (1- (2, 3-dioleoyloxy) propyl) -N, N, N-trimethylammonium chloride (DOTAP); 3- (N- (N ', N' -dimethylaminoethane) -carbamoyl) cholesterol (DC-Chol), N- (1, 2 dimyristoyloxyprop-3-yl) -N, N-dimethyl-N-hydroxyethylammonium bromide (DMRIE), 1, 2-dioleyloxy-N, N-dimethylaminopropane (DLInDMA), 1, 2-di-linoyloxy-N, N-dimethylaminopropane (DLenDMA), and mixtures thereof. As non-limiting examples, cationic lipids having a positive charge below physiological pH include, but are not limited to, DODAP, DODMA, and DSDMA. In some cases, the cationic lipid includes a protonated tertiary amine head group, a C18 alkyl chain, an ether linkage between the head group and the alkyl chain, and 0 to 3 double bonds. Such lipids include, for example, DSDMA, DLinDMA, DLenDMA, and DODMA. The cationic lipid may also include ether linkages and pH titratable head groups. Such lipids include, for example, DODMA.

The term "hydrophobic lipid" refers to compounds having non-polar groups, including, but not limited to, long chain saturated and unsaturated aliphatic hydrocarbon groups, and these groups are optionally substituted with one or more aromatic, alicyclic or heterocyclic groups. Suitable examples include, but are not limited to, diacylglycerols, dialkylglycerols, N-N-dialkylamino, 1, 2-diacyloxy-3-aminopropanes and 1, 2-dialkyl-3-aminopropanes.

The term "fusogenic" refers to the ability of a liposome, SNALP, or other drug delivery system to fuse with a cell membrane. The membrane may be a plasma membrane or a membrane surrounding an organelle, such as an endosome, nucleus, etc.

As used herein, the term "aqueous solution" refers to a composition that contains water, in whole or in part.

As used herein, the term "organic lipid solution" refers to a composition comprising, in whole or in part, an organic solvent with lipids.

As used herein, "distal site" refers to a physically separate site that is not limited to an adjacent capillary bed, but includes sites that are widely distributed throughout an organism.

"serum-stability" in relation to nucleic acid-lipid particles refers to particles that are not significantly degraded after exposure to serum or nuclease assays that will significantly degrade free DNA or RNA. Suitable assays include, for example, standard serum assays, or DNase assays, or RNAse assays.

As used herein, "systemic delivery" refers to delivery that results in a broad biodistribution of a compound, such as siRNA, within an organism. Some administration techniques may result in the systemic delivery of certain compounds, but not others. Systemic delivery means that a useful, preferably therapeutic, amount of the compound is exposed to the majority of the body. In order to achieve broad biodistribution, a blood lifetime is generally required such that the compound is not rapidly degraded or cleared (such as by first-pass organs (liver, lung, etc.) or by rapid, non-specific cellular binding) before reaching the site of disease remote from the site of administration. Systemic delivery of the nucleic acid-lipid particles can be by any means known in the art, including, for example, intravenous, subcutaneous, and intraperitoneal. In a preferred embodiment, the systemic delivery of the nucleic acid-lipid particle is by intravenous delivery.

As used herein, "local delivery" refers to the delivery of a compound, such as an siRNA, directly to a target site within an organism. For example, the compounds may be delivered locally by direct injection to a disease site such as a tumor or other target site such as a site of inflammation or a target organ such as the liver, heart, pancreas, kidney, etc.

The term "mammal" refers to any mammalian species, such as humans, mice, rats, dogs, cats, hamsters, guinea pigs, rabbits, livestock, and the like.

III.siRNAs

The modified siRNA molecules of the invention are capable of silencing expression of a target sequence, are about 15-60 nucleotides in length, are less immunostimulatory than a corresponding unmodified siRNA sequence, and retain RNAi activity against the target sequence. In some embodiments, the modified siRNA comprises at least one 2' OMe purine orPyrimidine nucleotides, such as 2 'OMe-guanosine, 2' OMe-uridine, 2 'OMe-adenosine, and/or 2' OMe-cytosine nucleotides. In a preferred embodiment, one or more uridine and/or guanosine nucleotides are modified. The modified nucleotides may be present in one strand (i.e., sense or antisense) or both strands of the siRNA. The siRNA sequences may have overhangs (e.g., 3 'or 5' overhangs, as in Elbashir et al, Genes Dev. (Gene development), 15: 188(2001) orEtc., Cell (Cell), 107: 309(2001), or may have no overhang (i.e., have a blunt end).

The modified siRNA typically comprises about 1% to about 100% (e.g., about 1%, 5%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, or 100%) of modified nucleotides in the double-stranded region of the siRNA duplex. In a preferred embodiment, less than about 20% (e.g., less than about 20%, 19%, 18%, 17%, 16%, 15%, 14%, 13%, 12%, 11%, 10%, 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2%, or 1%) or from about 1% to about 20% (e.g., from about 1% -20%, 2% -20%, 3% -20%, 4% -20%, 5% -20%, 6% -20%, 7% -20%, 8% -20%, 9% -20%, 10% -20%, 11% -20%, 12% -20%, 13% -20%, 14% -20%, 15% -20%, 16% -20%, 17% -20%, 18% -20%, or 19% -20%) of the nucleotides in the double-stranded region include modified nucleotides. In another preferred embodiment, for example, when one or both strands of the siRNA is selectively modified at uridine and/or guanosine nucleotides, the resulting modified siRNA can include less than about 30% modified nucleotides (e.g., less than about 30%, 29%, 28%, 27%, 26%, 25%, 24%, 23%, 22%, 21%, 20%, 19%, 18%, 17%, 16%, 15%, 14%, 13%, 12%, 11%, 10%, 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2%, or 1% modified nucleotides), or from about 1% to about 30% modified nucleotides (e.g., about 1% -30%, 2% -30%, 3% -30%, 4% -30%, 5% -30%, 6% -30%, 7% -30%, 8% -30%, 9% -30%, 10% -30%, 11% -30%, 12% -30%, 13% -30%, 14% -30%, 15% -30%, 16% -30%, 17% -30%, 18% -30%, 19% -30%, 20% -30%, 21% -30%, 22% -30%, 23% -30%, 24% -30%, 25% -30%, 26% -30%, 27% -30%, 28% -30%, or 29% -30% of modified nucleotides.

Selection of siRNA sequences

Suitable siRNA sequences can be identified using any method known in the art. Typically, the ratio will be found in Elbashir et al, Nature (Nature), 411: 494-498(2001) and Elbashir et al, EMBO J. (J. embryology), 20: 6877-6888(2001) is compatible with the methods described in Reynolds et al, Nature Biotech, (Nature Biotech), 22 (3): 326-330 (2004).

Generally, the 3' nucleotide sequence of the AUG initiation codon of a transcript from a target gene of interest is examined for dinucleotide sequences (e.g., AA, NA, CC, GG, or UU, where N ═ C, G, or U) (see, e.g., Elbashir et al, EMBO J. (J. embryology), 20: 6877-6888 (2001)). The nucleotide immediately 3' of the dinucleotide sequence was identified as a potential siRNA target sequence. Typically, 19, 21, 23, 25, 27, 29, 31, 33, 35, or more nucleotides immediately 3' of the dinucleotide sequence are identified as potential siRNA target sites. In some embodiments, the dinucleotide is an AA or NA sequence, and 19 nucleotides immediately 3' of the AA or NA dinucleotide are identified as potential siRNA target sites. siRNA target sites are typically spaced at different positions along the length of the target gene. To further enhance the silencing efficiency of the siRNA sequence, potential siRNA target sites can be analyzed to identify sites that do not contain regions of homology to other coding sequences, e.g., within a target cell or organism. For example, a suitable siRNA target site of about 21 base pairs will typically not have more than 16-17 consecutive base pairs homologous to the coding sequence in the target cell or organism. If the siRNA sequence will be expressed from an RNA Pol III promoter, then an siRNA target sequence lacking more than 4 consecutive A or T is selected.

Once potential siRNA sequences have been identified, the sequences can be analyzed using various criteria known in the art. For example, to improve their silencing efficiency, the siRNA sequences can be analyzed by rational design algorithms to identify sequences with one or more of the following characteristics: (1) a G/C content of about 25% to about 60% G/C; (2) at least 3A/U at positions 15-19 of the sense strand; (3) no internal repeats; (4) a at position 19 of the sense strand; (5) a at position 3 of the sense strand; (6) u at position 10 of the sense strand; (7) is not G/C at position 19 of the sense strand; and (8) is not G at position 13 of the sense strand. siRNA design tools incorporating algorithms that specify appropriate values for each of these characteristics and are used to select siRNA can be found, for example, in http:// boz094.ust. hk/RNAi/siRNA. It will be appreciated by those skilled in the art that sequences having one or more of the foregoing characteristics may be selected for further analysis and detection as potential siRNA sequences.

In addition, potential siRNA target sequences with one or more of the following criteria may generally be excluded as sirnas: (1) a sequence of fragments comprising 4 or more identical bases in a row; (2) sequences including G homopolymers (i.e., to reduce possible non-specific effects due to the structural features of these polymers); (3) a sequence comprising a three base motif (e.g., GGG, CCC, AAA, or TTT); (4) a sequence comprising 7 or more fragments of G/C in a row; and (5) sequences comprising direct repeats of 4 or more bases within the candidate resulting in an internal foldback structure. However, it will be appreciated by those skilled in the art that sequences having one or more of the above-described characteristics may still be selected for further analysis and detection as potential siRNA sequences.

In some embodiments, potential siRNA target sequences may be further analyzed based on siRNA duplex asymmetry, as described in, for example, Khvorova et al, Cell, 115: 209-216 (2003); and Schwarz et al, Cell (Cell), 115: 199-208 (2003). In other embodiments, potential siRNA target sequences can be further analyzed based on secondary structure at the mRNA target site, as described, for example, in Luo et al, biophysis.res.cmmun. (biophysical research exchange), 318: 303-310 (2004). For example, mRNA secondary structure can be modeled using the MFold algorithm (available from http:// www.bioinfo.rpi.edu/applications/Mfold/rna/forml. cgi) to select siRNA sequences that facilitate accessibility at mRNA target sites where there is less secondary structure present in base-pairing and stem-loop forms.

Once potential siRNA sequences have been identified, the sequences can be analyzed for the presence of any immunostimulatory properties, e.g., using in vitro cytokine assays or in vivo animal model analysis. Motifs in the sense and/or antisense strand of the siRNA sequence, such as GU-rich motifs (e.g., 5 '-GU-3', 5 '-UGU-3', 5 '-GU-3', 5 '-ugugugu-3', etc.) can also provide an indication of whether the sequence can be immunostimulatory. Once the siRNA molecule is found to be immunostimulatory, it may then be modified to reduce its immunostimulatory properties, as described herein. As a non-limiting example, an siRNA sequence can be contacted with a mammalian effector cell under conditions such that the cell produces a detectable immune response to determine whether the siRNA is an immunostimulatory or non-immunostimulatory siRNA. The mammalian effector cells can be from a mammal that was first used in an experiment (i.e., a mammal that has not previously been contacted with the gene product of the siRNA sequence). The mammalian effector cells can be, for example, Peripheral Blood Mononuclear Cells (PBMCs), macrophages, and the like. The detectable immune response may include the production of cytokines or growth factors, such as, for example, TNF- α, IFN- α, IFN- β, IFN- γ, IL-6, IL-12, or a combination thereof. The siRNA molecule identified as being immunostimulatory may then be modified to reduce its immunostimulatory properties by substituting at least one nucleotide on the sense and/or antisense strand with a modified nucleotide. For example, less than about 30% (e.g., less than about 30%, 25%, 20%, 15%, 10%, or 5%) of the nucleotides in the double-stranded region of the siRNA duplex may be substituted with modified nucleotides, such as 2' OMe nucleotides. The modified siRNA can then be contacted with mammalian effector cells as described above to confirm that its immunostimulatory properties have been reduced or eliminated.

Suitable in vitro assays for detecting an immune response include, but are not limited to, the double monoclonal antibody sandwich immunoassay technique of David et al (U.S. Pat. No. 4,376,110); monoclonal-polyclonal antibody sandwich assays (Wide et al, in Kirkham and Hunter, eds., Radioimmunoassay Methods, e. and s. livingstone, Edinburgh (1970)); the "western blotting" method of Gordon et al (U.S. Pat. No. 4,452,901); immunoprecipitation of labeled ligands (Brown et al, J biol. chem. (J. chem. biol., 255: 4980-4983 (1980)); enzyme-linked immunosorbent assays (ELISAs), such as, for example, Raines et al, j. biol. chem. (journal of biology), 257: 5154-5160 (1982); immunocytochemistry techniques, including the use of fluorescent dyes (Brooks et al, Clin. Exp. Immunol. (clinical Experimental immunology), 39: 477 (1980)); and activity neutralization (Bowen-Pope et al, Proc. Natl. Acad. Sci. USA, proceedings of the national academy of sciences USA, 81: 2396-2400 (1984)). In addition to the immunoassays described above, many other immunoassays are available, including those described in U.S. patent nos. 3,817,827; 3,850,752, respectively; 3,901,654, respectively; 3,935,074, respectively; 3,984,533, respectively; 3,996,345; 4,034,074, respectively; and 4,098,876.

Non-limiting examples of in vivo models for detecting immune responses include in vivo mouse cytokine induction assays, which can be performed as follows: (1) siRNA can be administered into the lateral tail vein by standard intravenous injection; (2) at about 6 hours after administration, blood can be collected by cardiac puncture and processed as plasma for cytokine analysis; and (3) cytokines can be quantified using a sandwich ELISA kit according to the supplier's instructions (e.g., mouse and human IFN-alpha (PBL biomedicine; Piscataway, N.J.), human IL-6 and TNF-alpha (eBioscience; San Diego, Calif.), and mouse IL-6, TNF-alpha, and IFN-gamma (BD bioscience; San Diego, Calif.)).

Monoclonal ANTIBODIES that specifically bind cytokines and growth factors are commercially available from a variety of sources and can be produced using methods known in the art (see, e.g., Kohler and Milstein, Nature (Nature), 256: 495-497 (1975); and Harlow and Lane, ANTIBODIES, ALABORATORY MANUAL (ANTIBODIES, A laboratory Manual), Cold spring harbor Press (Cold spring harbor Publication), New York (1999)). The production of monoclonal antibodies has been previously described and can be achieved by any means known in the art (see, Buhring et al, in Hybridoma, Vol.10, No. 1, pp.77-78 (1991)). In some methods, monoclonal antibodies are labeled (e.g., with any combination detectable spectroscopically, photochemically, biochemically, electrically, optically, chemically, etc.) to facilitate detection.

B. Generation of siRNA molecules

The siRNA molecule may be provided in some form, including, for example, as one or more isolated small interfering RNA (siRNA) duplexes. The siRNA sequences may have overhangs (e.g., 3 'or 5' overhangs, as in Elbashir et al, Genes Dev. (Gene development), 15: 188(2001) orEtc., Cell (Cell), 107: 309(2001), or may lack an overhang (i.e., have a blunt end).

Preferably, the siRNA molecule is chemically synthesized. Single stranded molecules comprising the siRNA molecules can be synthesized using any of a number of techniques known in the art, such as those described in Usman et al, j.am.chem.soc. (journal of the american chemical society), 109: 7845 (1987); scaring et al, nucleic acids sres (nucleic acids research), 18: 5433 (1990); wincott et al, nucleic acids Res (nucleic acids research), 23: 2677-2684 (1995); and Wincott et al, Methods mol.bio. (Methods of molecular biology), 74: 59 (1997). Textbooks disclosing other bases for the conventional methods of the invention include Sambrook et al, Molecular Cloning, A Laboratory Manual (Molecular Cloning, A Laboratory Manual) (2 nd edition. 1989); kriegler, Gene Transfer and Expression: ALABORORT MANUAL (Gene transfer and expression: A laboratory Manual) (1990); and Current protocols in Molecular Biology (modern methods of Molecular Biology) (Ausubel et al, eds., 1994). The synthesis of single stranded molecules uses conventional nucleic acid protecting and coupling groups such as dimethoxytrityl at the 5 'end and phosphoramidite at the 3' end. As a non-limiting example, small scale synthesis can be performed on an Applied Biosystems synthesizer (Applied Biosystems synthesizer) using a 0.2. mu. mol scale method with a 2.5 minute coupling step for 2' -O-methylated nucleotides. Alternatively, synthesis on the 0.2 μmol scale may be performed on a 96-well plate synthesizer from Protogene (Palo Alto, Calif.). However, larger or smaller scale syntheses are also within the scope of the invention. Suitable reagents for synthesizing siRNA single-stranded molecules, methods of RNA deprotection, and methods of RNA purification are known to those skilled in the art.

The siRNA molecules can also be synthesized by tandem synthesis techniques in which the two strands are synthesized as a single contiguous fragment or strands separated by a lytic linker, which can then be cleaved to provide separate fragments or strands that hybridize to form an siRNA duplex. The linker may be a polynucleotide linker or a non-nucleotide linker. Tandem synthesis of siRNA can readily employ multi-well/multi-plate synthesis platforms, as well as large scale synthesis platforms using batch reactors, synthesis columns, and the like. Alternatively, the siRNA molecule may be assembled from two different single stranded molecules, wherein one strand comprises the sense strand and the other comprises the antisense strand of the siRNA. For example, each strand may be synthesized separately and joined together by hybridization or ligation after synthesis and/or deprotection. In certain other cases, the siRNA molecule may be synthesized as a single contiguous fragment in which self-complementary sense and antisense regions hybridize to form an siRNA duplex having a hairpin secondary structure.

C. Modified siRNA sequences

In certain aspects, the siRNA molecules of the invention comprise duplexes having two strands and at least one modified nucleotide in the duplex region, wherein each strand is about 15 to about 60 nucleotides in length. Advantageously, the modified siRNA is less immunostimulatory than a corresponding unmodified siRNA sequence, but retains the ability to silence expression of the target sequence.

Examples of modified nucleotides suitable for use in the present invention include, but are not limited to, ribonucleotides having a 2 '-O-methyl (2' OMe), 2 '-deoxy-2' -fluoro (2 'F), 2' -deoxy, 5-C-methyl, 2 '-O- (2-Methoxyethyl) (MOE), 4' -thio, 2 '-amino, or 2' -C-allyl group. Nucleotides with modifications confirmed by northern blotting, such as, for example, those described in Saenger, Principles of Nucleic acid Structure, eds. (1984) by Springer-Verlag, are also suitable for use in the siRNA molecules of the present invention. Some modified nucleotides include, but are not limited to, Locked Nucleic Acid (LNA) nucleotides (e.g., 2 ' -O, 4 ' -C-methylene- (D-ribofuranosyl) nucleotides), 2 ' -O- (2-Methoxyethyl) (MOE) nucleotides, 2 ' -methyl-thio-ethyl nucleotides, 2 ' -deoxy-2 ' -fluoro (2 ' F) nucleotides, 2 ' -deoxy-2 ' -chloro (2 ' Cl) nucleotides, and 2 ' -azido nucleotides. In certain instances, the siRNA molecules of the present invention comprise one or more G-clamp loop (clamp) nucleotides. A G-clamp nucleotide refers to a modified cytosine analog in which the modification confers the Watson-Crick and Hoogsteen-faced hydrogen bonding capabilities on complementary guanine nucleotides within the duplex (see, e.g., Lin et al, J.am.chem.Soc. (J.Am.Chem., 120: 8531-8532 (1998)). In addition, nucleotides having nucleotide base analogs, such as, for example, C-phenyl, C-naphthyl, other aromatic derivatives, inosine, pyrrole carboxamides, and nitropyrrole derivatives, such as 3-nitropyrrole, 4-nitroindole, 5-nitroindole, and 6-nitroindole (see, for example, Loakes, Nucl, Acids Res. (nucleic Acids research), 29: 2437-2447(2001)), can be incorporated into the siRNA molecules of the present invention.

In certain embodiments, the siRNA molecules of the invention further comprise one or more chemical modifications, such as end-capping moieties, phosphate backbone modifications, and the like. Examples of terminal end-capping moieties include, but are not limited to, an inverted deoxy non-base residue, glycerol modification, 4 ', 5 ' -methylene nucleotide, 1- (. beta. -D-erythrofuranosyl) nucleotide, 4 ' -thio nucleotide, carbocyclic nucleotide, 1, 5-anhydrohexitol nucleotide, L-nucleotide, alpha-nucleotide, modified base nucleotide, threo-pentofuranosyl nucleotide, acyclic 3 ', 4 ' -split nucleotide, acyclic 3, 4-dihydroxybutyryl nucleotide, acyclic 3, 5-dihydroxypentyl nucleotide, 3 ' -3 ' -inverted nucleotide moiety, 3 ' -3 ' -inverted non-base moiety, 3 ' -2 ' -inverted nucleotide moiety, 3 ' -2 ' -inverted non-base moiety, 5 '-5' reverse nucleotide moiety, 5 '-5' -reverse non-base moiety, 3 '-5' -reverse deoxy non-base moiety, 5 '-amino-alkyl phosphate, 1, 3-diamino-2-propyl phosphate, 3-aminopropyl phosphate, 6-aminohexyl phosphate, 1, 2-aminododecyl phosphate, hydroxypropyl phosphate, 1, 4-butanediol phosphate, 3' -phosphoramidate, 5 '-phosphoramidate, hexyl phosphate, aminohexyl phosphate, 3' -phosphate, 5 '-amino, 3' -phosphorothioate, 5 '-phosphorothioate, phosphorodithioate, and bridged or unbridged methyl phosphate or 5' -sulfhydryl moiety (see, for example, U.S. Pat. nos. 5,998,203; beaucage et al, Tetrahedron 49: 1925(1993)). Non-limiting examples of phosphate Backbone Modifications (i.e., resulting in linkage between modified nucleotides) include phosphorothioate, phosphorodithioate, methylphosphonate, phosphotriester, morpholino, amidate, carbamate, carboxymethyl, acetamidate, polyamide, sulfonate, sulfonamide, sulfamate, formacetal, thioformacetal, and alkylsilyl substitutions (see, e.g., Hunziker et al, Nucleic acid analogs: Synthesis and Properties, in modern synthetic Methods VCH, 331-417 (1995); Mesmaker et al, Novel Backbone latex for Oligonucleotides (new scaffold substitutions for Oligonucleotides), in Carbohydrate Modifications in antibiotics (Carbohydrate substitutions in Antisense Research), 24-39 (1994)). Such chemical modifications can occur at the 5 '-end and/or the 3' -end of the sense, antisense, or both strands of the siRNA.

In some embodiments, the sense and/or antisense strand can further comprise a 3 '-terminal overhang having from about 1 to about 4 (e.g., 1, 2,3, or 4) 2' -deoxyribonucleotides and/or any combination of modified and unmodified nucleotides. Other examples of the types of modified nucleotides and chemical modifications that may be incorporated into the modified siRNA molecules of the invention are described, for example, in british patent No. GB2,397,818B and U.S. patent publication nos. 20040192626 and 20050282188.

The modified siRNA molecules of the invention can optionally include one or more non-nucleotides in one or both strands of the siRNA. As used herein, the term "non-nucleotide" refers to any group or compound that can be introduced into a nucleic acid strand in place of one or more nucleotide units, including sugar and/or phosphate substitutions, and which allows the remaining bases to exhibit their activity. The group or compound is non-base in that it does not contain what are commonly referred to as nucleotide bases, such as adenosine, guanosine, cytosine, uracil or thymine, and therefore lacks a base at the 1' -position.

In other embodiments, the chemical modification of the siRNA comprises attaching a conjugate to the chemically modified siRNA molecule. The conjugate can be attached to the 5 'and/or 3' end of the sense and/or antisense strand of the chemically modified siRNA by covalent attachment, such as, for example, a biodegradable linker. The conjugates can also be attached to chemically modified sirnas, e.g., via a carbamate group or other linking group (see, e.g., U.S. patent publication nos. 20050074771, 20050043219, and 20050158727). In certain instances, the conjugate is a molecule that facilitates delivery of the chemically modified siRNA into a cell. Examples of conjugate molecules suitable for attachment to the chemically modified siRNA of the present invention include, but are not limited to, steroids such as cholesterol, glycols such as polyethylene glycol (PEG), Human Serum Albumin (HSA), fatty acids, carotenoids, terpenes, bile acids, folic acid (e.g., folic acid analogs and derivatives thereof), sugars (e.g., galactose, galactosamine, N-acetylgalactosamine, glucose, mannose, fructose, trehalose, and the like), phospholipids, peptides, ligands of cellular receptors capable of modulating cellular uptake, and combinations thereof (see, e.g., U.S. patent publication nos. 20030130130186, 20040110296, and 20040249178; U.S. patent No. 6,753,423). Other examples include lipophilic moieties, vitamins, polymers, peptides, proteins, nucleic acids, small molecules, oligosaccharides, carbohydrate clusters, intercalators, minor groove binders, disintegrators, and crosslinker conjugate molecules described in U.S. patent publication nos. 20050119470 and 20050107325. Other examples described in U.S. patent publication No. 20050153337 include 2 '-O-alkylamines, 2' -O-alkoxyalkylamines, polyamines, C5-cationic modified pyrimidines, cationic peptides, guanidine groups, ammonium groups, cationic amino acid conjugate molecules. Other examples include hydrophobic groups, membrane active compounds, cell penetrating compounds, cell targeting signals, interaction modifiers, and steric stabilizer conjugate molecules described in U.S. patent publication No. 20040167090. Other examples include the conjugate molecules described in U.S. patent publication No. 20050239739. The type of conjugate used and the degree of conjugation to the chemically modified siRNA molecule can be evaluated to improve the pharmacokinetic properties, bioavailability, and/or stability of the siRNA while retaining full RNAi activity. Likewise, using any of a number of well-known in vitro cell culture or in vivo animal models, one skilled in the art can screen chemically modified siRNA molecules having attached thereto some conjugates to identify those siRNA molecules with improved properties and full RNAi activity.

D. Target genes

The modified siRNA molecules described herein can be used to down-regulate or silence the translation (i.e., expression) of a gene of interest. Genes of interest include, but are not limited to, genes associated with viral infection and survival, genes associated with metabolic diseases and disorders (e.g., liver diseases and disorders), genes associated with tumorigenesis and cellular transformation, angiogenic genes, immune regulatory genes, such as those associated with inflammatory and autoimmune responses, ligand receptor genes, and genes associated with neurodegenerative disorders.

Genes associated with viral infection and survival include those expressed by viruses to bind, enter and replicate in cells. Of particular interest are viral sequences associated with chronic viral diseases. Viral sequences of particular interest include the sequences of the following viruses: filoviruses, such as ebola and marburg (see, e.g., U.S. patent application Ser. No. 11/584,341; and Geisbert et al, J.Infect.Dis. (J.Immunol., 193: 1650-1657 (2006)); arenaviruses, such as Lassa virus, Junin virus, Vivian hemorrhagic fever virus, Guinarito virus, and Sabia virus (Buchmeier et al, Arenaveridae: the viruses and the replication thereof in the Arenaviridae family: viruses and their replication), in FIELDS VIROLOGY, Knipe et al (ed.), 4 th edition, Lippincott-Raven, Philadelphia, (2001)); influenza viruses, such as influenza A, B and C (see, e.g., U.S. provisional patent application No. 60/737,945; Steinhauer et al, AnnuRev Genet. (Ann. J. GenVirol., 83: 2635-2662 (2002)); hepatitis virus (Hamasaki et al, FEBS Lett., 543: 51 (2003); Yokota et al, EMBO Rep. (embryology reports), 4: 602 (2003); Schlomai et al, Hepatology, 37: 764 (2003); Wilson et al, Proc.Natl.Acad.Sci.USA (Proc.Natl.Acad.Sci.USA), 100: 2783 (2003); Kapadia et al, Proc.Natl.Acad.Sci.USA (Proc.Natl.Acad.Acad.Sci.USA), 100: 2014 (2003); and FIELDS VIROLOGY (VIROLOGY field), Knipe et al (eds.), 4 th edition, Lippincott-Raven, Philadelphia (2001)); human Immunodeficiency Virus (HIV) (Banerjea et al, mol. ther. (molecular therapy), 8: 62 (2003); Song et al, J. Virol. (J. Virol., 77: 7174 (2003); Stephenson, JAMA, 289: 1494 (2003); Qin et al, Proc. Natl. Acad. Sci. USA (Proc. Natl. Acad. Sci., USA), 100: 183 (2003)); herpes virus (Jia et al, J.Virol. (J.Virol., 77: 3301 (2003)); and Human Papilloma Virus (HPV) (Hall et al, J.Virol. (J. Virol., 77: 6066 (2003); Jiang et al, Oncogene, 21: 6041 (2002)).

Exemplary filovirus nucleic acid sequences that can be silenced include, but are not limited to, nucleic acid sequences encoding structural proteins (e.g., VP30, VP35, Nucleoprotein (NP), polymerase protein (L-pol)) and membrane-associated proteins (e.g., VP40, Glycoprotein (GP), VP 24). The complete genomic sequence of ebola virus is described, for example, in Genbank accession No. NC — 002549; AY 769362; NC-006432; NC-004161; AY 729654; AY 354458; AY 142960; AB 050936; AF 522874; AF 499101; AF 272001; and AF 086833. The ebola virus VP24 sequences are listed, for example, in Genbank accession numbers U77385 and AY 058897. The Epbola virus L-pol sequence is listed, for example, in Genbank accession number X67110. The ebola virus VP40 sequence is listed, for example, in Genbank accession No. AY 058896. The Epbola virus NP sequence is listed, for example, in Genbank accession number AY 058895. Epulavirus GP sequences are described, for example, in Genbank accession number AY 058898; sanchez et al, Virus Res, (Virus research), 29: 215-240 (1993); wil et al, j.virol, (virology research), 67: 1203-1210 (1993); volchkov et al, FEBS lett, 305: 181-184 (1992); and in U.S. patent No. 6,713,069. Other ebola virus sequences are listed, for example, in Genbank accession numbers L11365 and X61274. The complete genomic sequence of marburg virus is described, for example, in Genbank accession No. NC _ 001608; AY 430365; AY 430366; and AY 358025. Marburg virus GP sequences are described, for example, in Genbank accession No. AF 005734; AF 005733; and AF 005732. The marburg virus VP35 sequences are listed, for example, in Genbank accession numbers AF005731 and AF 005730. Other marburg virus sequences are described, for example, in Genbank accession number X64406; z29337; AF 005735; and Z12132.

Exemplary influenza nucleic acid sequences that can be silenced include, but are not limited to, nucleic acid sequences encoding Nucleoprotein (NP), matrix protein (M1 and M2), nonstructural protein (NS1 and NS2), RNA polymerase (PA, PB1, PB2), Neuraminidase (NA), and Hemagglutinin (HA). Influenza a NP sequences are described, for example, in Genbank accession No. NC _ 004522; AY 818138; AB 166863; AB 188817; AB 189046; AB 189054; AB 189062; AY 646169; AY 646177; AY 651486; AY 651493; AY 651494; AY 651495; AY 651496; AY 651497; AY 651498; AY 651499; AY 651500; AY 651501; AY 651502; AY 651503; AY 651504; AY 651505; AY 651506; AY 651507; AY 651509; AY 651528; AY 770996; AY 790308; AY 818138; and AY 818140. Influenza a PA sequences are described, for example, in Genhank accession No. AY 818132; AY 790280; AY 646171; AY 818132; AY 818133; AY 646179; AY 818134; AY 551934; AY 651613; AY 651610; AY 651620; AY 651617; AY 651600; AY 651611; AY 651606; AY 651618; AY 651608; AY 651607; AY 651605; AY 651609; AY 651615; AY 651616; AY 651640; AY 651614; AY 651612; AY 651621; AY 651619; AY 770995; and AY 724786.

Exemplary hepatitis virus nucleic acid sequences that can be silenced include, but are not limited to, nucleic acid sequences involved in transcription and translation (e.g., En1, En2, X, P) and nucleic acid sequences encoding structural proteins (e.g., core proteins, including C and C-related proteins, capsid and envelope proteins, including S, M and/or L proteins, or fragments thereof) (see, e.g., FIELDS virogy, supra). Exemplary hepatitis c nucleic acid sequences that can be silenced include, but are not limited to, serine proteases (e.g., NS3/NS4), helicases (e.g., NS3), polymerases (e.g., NS5B), and envelope proteins (e.g., E1, E2, and p 7). Hepatitis a nucleic acid sequences are listed, for example, in Genbank accession No. NC _ 001489; hepatitis b nucleic acid sequences are listed, for example, in Genbank accession No. NC _ 003977; hepatitis c nucleic acid sequences are listed, for example, in Genbank accession No. NC _ 004102; hepatitis delta nucleic acid sequences are listed, for example, in Genbank accession number NC-001653; hepatitis E nucleic acid sequences are listed, for example, in Genbank accession number NC-001434; and the hepatitis G nucleic acid sequence is listed, for example, in Genbank accession number NC-001710. Silencing sequences encoding genes associated with viral infection and survival may be conveniently used in combination with administration of conventional agents for treating the viral condition.

Genes associated with metabolic diseases and disorders (e.g., disorders in which the liver is a target and liver diseases and disorders) include, for example, genes expressed in dyslipidemia (e.g., liver X receptors such as LXR α and LXR β (Genbank accession No. NM _007121), Farnesoid X Receptor (FXR) (Genbank accession No. NM _005123), sterol-regulatory element binding protein (SREBP), site I protease (S1P), 3-hydroxy-3-methylglutaryl coenzyme-a reductase (HMG coenzyme-a reductase), B (ApoB), and apolipoprotein (ApoE)); and Genes that are aberrantly expressed in diabetes (e.g., glucose-6-phosphatase) (see, e.g., Forman et al, Cell 81: 687 (1995); Seol et al, mol. Endocrinol. (molecular Endocrinol.), 9: 72(1995), Zavacki et al, Proc. Natl. Acad. Sci. USA (Proc. Natl. Acad. Sci., USA), 94: 7909 (1997); Sakai et al, Cell (Cell), 85: 1037-1046 (1996); Duncan et al, J. biol. chem. (J. chem. biol.), 272: 12778-12785 (1997); Willy et al, Genes Dev. (Geneto., 9: 1033-1045 (1995); Lehmann et al. chem. (biochem. biol., 272: 3137-3140 (1997); Janoki et al, Natws. 731, Nature. (1998), Nature: 731-383, 1996); Cell (1996); Cell 93: 69383, Peh. biochem.)). It will be understood by those skilled in the art that genes associated with metabolic diseases and conditions (e.g., diseases and conditions in which the liver is the target and liver diseases and conditions) include genes expressed in the liver itself as well as genes expressed in other organs and tissues. Silencing of sequences encoding genes associated with metabolic diseases and disorders can be readily used in combination with administration of conventional agents used to treat the diseases or disorders.

Examples of gene sequences associated with tumorigenesis and cell transformation include mitotic kinesins, such as Eg 5; translocation sequences, such as MLL fusion genes, BCR-ABL (Wilda et al, Oncogene, 21: 5716 (2002); Scherr et al, Blood, 101: 1566(2003)), TEL-AML1, EWS-FLI1, TLS-FUS, PAX3-FKHR, BCL-2, AML1-ETO, and AML1-MTG8(Heidenreich et al, Blood, 101: 3157 (2003)); over-expressed sequences such as multidrug resistance Genes (Nieth et al, FEBS Lett., 545: 144 (2003); Wu et al, Cancer Res. (Cancer research) 63: 1515(2003)), cyclin (Li et al, Cancer Res. (Cancer research), 63: 3593 (2003); Zou et al, Genes Dev. (GeneDev., 16: 2923(2002)), β -catenin (Verma et al, Clin Cancer Res. (clinical Cancer research), 9: 1291(2003)), telomere end transferase Genes (Kosciolek et al, Mol Cancer Ther. (molecular Cancer therapy), 2: 209(2003)), c-MYC, N-MYC, BCL-2, ERBB 1, and ERBB2(Nagy et al, exp. cell Res. (cell research), 285: 39 (2003)); and mutated sequences, such as RAS (reviewed in Tuschl and Borkhardt, mol. interfaces, 2: 158 (2002)). Silencing of sequences encoding DNA repair enzymes is used in combination with administration of chemotherapeutic agents (Collis et al, cancer Res. (cancer research), 63: 1550 (2003)). Genes encoding proteins associated with tumor migration are also target sequences of interest, e.g., integrins, selectins, and metalloproteinases. The above examples are not exclusive. Any complete or partial gene sequence that contributes to or promotes tumorigenesis or cell transformation, tumor growth, or tumor metastasis may be included as a template sequence.

Angiogenic genes can promote the formation of new blood vessels. Of particular interest are Vascular Endothelial Growth Factor (VEGF) (Reich et al, mol. vis. (molecular point of view), 9: 210(2003)) or VEGFr. siRNA sequences targeting VEGFr are described, for example, in GB 2396864; U.S. patent publication numbers 20040142895; and CA 24576444.

The anti-angiogenic gene is capable of inhibiting neovascularization. These genes are particularly useful in the treatment of those cancers in which angiogenesis plays a role in the pathological development of the disease. Examples of anti-angiogenic genes include, but are not limited to, endostatin (see, e.g., U.S. patent No. 6,174,861), angiostatin (see, e.g., U.S. patent No. 5,639,725), and VEGF-R2 (see, e.g., Decaussin et al, j.pathol. (journal of pathology), 188: 369-377 (1999)).

An immune regulatory gene is a gene that regulates one or more immune responses. Examples of immunomodulatory genes include, but are not limited to, cytokines such as growth factors (e.g., TGF- α, TGF- β, EGF, FGF, IGF, NGF, PDGF, CGF, GM-CSF, SCF, etc.), interleukins (e.g., IL-2, IL-4, IL-12(Hill et al. J. Immunol., 171: 691(2003)), IL-15, IL-18, IL-20, etc.), interferons (e.g., IFN- α, IFN- β, IFN- γ, etc.), and TNF. Fas and Fas ligand genes are also immune regulatory target sequences of interest (Song et al, nat. Med. (Nature medicine), 9: 347 (2003)). Genes encoding secondary signaling molecules in hematopoietic and lymphoid cells are also encompassed by the invention, e.g., Tec family kinases such as Bruton's tyrosine kinase (Btk) (Heinonen et al, FEBS lett., 527: 274 (2002)).

Cellular receptor ligands include ligands that are capable of binding to cell surface receptors (e.g., insulin receptors, EPO receptors, G-protein coupled receptors, receptors with tyrosine kinase activity, cytokine receptors, growth factor receptors, etc.) to modulate (e.g., inhibit, activate, etc.) the physiological pathways in which the receptors are involved (e.g., glucose level regulation, blood cell development, mitogenesis, etc.). Examples of cellular receptor ligands include, but are not limited to, cytokines, growth factors, interleukins, interferons, Erythropoietin (EPO), insulin, glucagon, G-protein coupled receptor ligands, and the like. Expanded templates encoding trinucleotide repeats (e.g., CAG repeats) are used to silence pathological sequences in neurodegenerative disorders caused by expansion of trinucleotide repeats, such as spinal and bulbar muscular atrophy and huntington's disease (Caplen et al, hum. mol. genet. (human molecular genetics), 11: 175 (2002)).

In addition to its use to silence the expression of any of the above genes for therapeutic purposes, the siRNA described herein is also useful in research and development applications, as well as diagnostic, prophylactic, prognostic, clinical and other health care applications. As a non-limiting example, the modified siRNA molecules of the invention can be used in target validation studies directed at detecting whether a gene of interest has the potential to be a therapeutic target. The modified siRNA molecules of the invention may also be used in target identification studies aimed at the discovery of genes as potential therapeutic targets.

Vector systems containing siRNA

In one aspect, the invention provides a vector system comprising a modified siRNA molecule described herein. In some embodiments, the vector system is a lipid-based vector system, such as a stable nucleic acid-lipid particle (e.g., SNALP or SPLP), a cationic lipid or liposomal nucleic acid complex (i.e., a lipid nucleic acid complex), a liposome, a micelle, a viral particle, or a mixture thereof. In other embodiments, the vector system is a polymer-based vector system, such as a cationic polymer-nucleic acid complex (i.e., a polymer nucleic acid complex). In other embodiments, the carrier system is a cyclodextrin-based carrier system, such as a cyclodextrin polymer-nucleic acid complex. In other embodiments, the vector system is a protein-based vector system, such as a cationic peptide-nucleic acid complex. Preferably, the vector system is a stable nucleic acid-lipid particle, such as SNALP or SPLP. It will be appreciated by those skilled in the art that the modified siRNA molecules of the invention may also be delivered as naked siRNA.

A. Stable nucleic acid-lipid particles

The stabilized nucleic acid-lipid particles (SNALPs) of the invention typically comprise a modified siRNA molecule described herein, a cationic lipid (e.g., a cationic lipid of formula I or II), and a non-cationic lipid. The SNALPs may further comprise a bilayer stabilizing component (i.e., a conjugated lipid that inhibits particle aggregation). The SNALPs can include at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, or more modified siRNA molecules described herein, alone or in combination with at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, or more unmodified siRNA molecules.

The SNALPs of the invention typically have an average diameter of about 50nm to about 150nm, more typically about 60nm to about 130nm, more typically about 70nm to about 110nm, and most typically about 70nm to about 90nm, and are substantially non-toxic. Furthermore, when present in the nucleic acid-lipid particle, the nucleic acid is resistant to degradation with nucleases in aqueous solution. Nucleic acid-lipid particles and methods for their preparation are described, for example, in U.S. patent nos. 5,753,613; 5,785,992, respectively; 5,705,385, respectively; 5,976,567, respectively; 5,981,501, respectively; 6,110,745, respectively; and 6,320,017; and PCT publication No. WO 96/40964.

1. Cationic lipids

Various cationic lipids can be used in the stabilized nucleic acid-lipid particles of the invention, alone or in combination with one or more other cationic lipid species or non-cationic lipid species.

Cationic lipids useful in the present invention can be any of a number of lipid species that carry a net positive charge at physiological pH. Such lipids include, but are not limited to, DODAC, DODMA, DSDMA, DOTMA, DDAB, DOTAP, DOSPA, DOGS, DC-Chol, DMRIE, and combinations thereof. Many of these lipids and related analogs have been described in U.S. patent publication nos. 20060083780; U.S. Pat. nos. 5,208,036, 5,264,618, 5,279,833, 5,283,185, and 5,753,613 and 5,785,992; and PCT publication No. WO 96/10390. In addition, many commercially available cationic lipid formulations are available and can be used in the present invention. These include, for example, (commercial cationic liposomes including DOTMA and DOPE from GIBCO/BRL, Grand Island, N.Y., USA);(commercial cationic liposomes including DOSPA and DOPE, from GIBCO/BRL); and(commercial cationic liposomes including DOGS, from Promega Corp., Madison, Wisconsin, USA).

Also, the cationic lipid of formula I having the following structure is effectively used in the present invention.

Wherein R is¹And R²Is independently selected and is H or C₁-C₃Alkyl radical, R³And R⁴Is independently selected, and is an alkyl group having from about 10 to about 20 carbon atoms, and R³And R⁴Comprises at least two sites of unsaturation. In some cases, R³And R⁴Are all the same, i.e. R³And R⁴Are all linoleic (C18), and so on. In certain other cases, R³And R⁴Are different, i.e. R³Is tetradecatrienyl (C14), and R⁴Is a linoleyl group (C18). In a preferred embodiment, the cationic lipid of formula I is symmetrical, i.e. R³And R⁴Are all the same. In another preferred embodiment, R³And R⁴Both include at least two sites of unsaturation. In some embodiments, R³And R⁴Independently selected from the group consisting of dodecadienyl, tetradecadidienyl, hexadecadidienyl, linoleyl, and eicosadidienyl. In a preferred embodiment, R ³And R⁴Are all linoleic. In some embodiments, R³And R⁴Includes at least three sites of unsaturation and is independently selected from, for example, dodecatrienyl, tetradecatrienyl, hexadecatrienyl, linolenyl, and eicosatrienyl groups. In a particularly preferred embodiment, the cationic lipid of formula I is DLinDMA or DLenDMA.

In addition, the cationic lipid of formula II having the following structure is effectively used in the present invention.

Wherein R is¹And R²Is independently selected and is H or C₁-C₃Alkyl radical, R³And R⁴Is independently selected, and is an alkyl group having from about 10 to about 20 carbon atoms, and R³And R⁴Comprises at least two sites of unsaturation. In some cases, R³And R⁴Are all the same, i.e. R³And R⁴Are all linoleic (C18), and so on. In certain other cases, R³And R⁴Are different, i.e. R³Is tetradecatrienyl (C14), and R⁴Is a linoleyl group (C18). In a preferred embodiment, the cationic lipids of the invention are symmetrical, i.e. R³And R⁴Are all the same. In another preferred embodiment, R³And R⁴Both include at least two sites of unsaturation. In some embodiments, R³And R ⁴Independently selected from the group consisting of dodecadienyl, tetradecadidienyl, hexadecadidienyl, linoleyl, and eicosadidienyl. In a preferred embodiment, R³And R⁴Are all linoleic. In some embodiments, R³And R⁴Includes at least three sites of unsaturation and is independently selected from, for example, dodecatrienyl, tetradecatrienyl, hexadecatrienyl, linolenyl, and eicosatrienyl groups.

The cationic lipid typically comprises from about 2 mol% to about 60 mol%, from about 5 mol% to about 50 mol%, from about 10 mol% to about 50 mol%, from about 20 mol% to about 40 mol%, from about 30 mol% to about 40 mol%, or about 40 mol% of the total lipid present in the particle. Those skilled in the art will readily appreciate that the proportions of the components will vary and the delivery efficiency of a particular formulation can be measured using, for example, an Endosomal Release Parameter (ERP) assay, depending on the intended use of the particle. For example, for systemic delivery, the cationic lipid may comprise from about 5 mol% to about 15 mol% of the total lipid present in the particle, and for local or regional delivery, the cationic lipid may comprise from about 30 mol% to about 50 mol%, or about 40 mol%, of the total lipid present in the particle.

2. Non-cationic lipids

The non-cationic lipid used in the stabilized nucleic acid-lipid particle of the present invention can be any of a variety of neutral uncharged, zwitterionic or anionic lipids that are capable of producing stable complexes. They are preferably neutral, although they may alternatively be positively or negatively charged. Examples of non-cationic lipids include, but are not limited to, phospholipid-related substances such as phosphatidylcholine, phosphatidylethanolamine, lysophosphatidylcholine, lysophosphatidylethanolamine, phosphatidylserine, phosphatidylinosine, sphingomyelin, lecithine (ESM), cephalin, cardiolipin, phosphatidic acid, cerebroside, hexacosanyl phosphate, Distearoylphosphatidylcholine (DSPC), Dioleoylphosphatidylcholine (DOPC), Dipalmitoylphosphatidylcholine (DPPC), Dioleoylphosphatidylglycerol (DOPG), Dipalmitoylphosphatidylglycerol (DPPG), Dioleoylphosphatidylethanolamine (DOPE), palmitoyloleoylphosphatidylcholine (POPE), palmitoyloleoylphosphatidylglycerol (POPG), dioleoylphosphatidylethanolamine 4- (N-maleimidomethyl) -cyclohexane-1-carboxylate (DOPE-mal), dipalmitoyl-phosphatidylethanolamine (DPPE), dimyristoyl-phosphatidylethanolamine (DMPE), distearoyl-phosphatidylethanolamine (DSPE), monomethyl-phosphatidylethanolamine, dimethyl-phosphatidylethanolamine, dioelaidoyl-phosphatidylethanolamine (DEPE), and stearoyl-oleoyl-phosphatidylethanolamine (SOPE). Non-cationic lipids or sterols such as cholesterol may also be present. Additional lipids that do not include phosphorus include, for example, octadecylamine, dodecylamine, hexadecylamine, acetylhexadecylamine, glycerol ricinoleate, hexadecylstearate, isopropyl myristate, amphoteric acrylic polymers, triethanolamine-lauryl sulfate, alkyl-arylsulfate polyethoxylated (polyethyloxylated) fatty acid amides, dioctadecyldimethylammonium bromide, ceramides, diacylphosphatidylcholine, diacylphosphatidylethanolamine, and the like. Other lipids may be present, such as lysophosphatidylcholine and lysophosphatidylethanolamine. Non-cationic lipids also include polyethylene glycol-based polymers such as PEG 2000, PEG 5000, and polyethylene glycol conjugated to phospholipids or to ceramides (referred to as PEG-Cer), as described in U.S. patent application No. 08/316,429.

In preferred embodiments, the non-cationic lipid is a diacylphosphatidylcholine (e.g., distearoylphosphatidylcholine, dioleoylphosphatidylcholine, dipalmitoylphosphatidylcholine, and dilinoleoylphosphatidylcholine), a diacylphosphatidylethanolamine (e.g., dioleoylphosphatidylethanolamine and palmitoyloleoylphosphatidylethanolamine), a ceramide, or a sphingomyelin. The acyl groups in these lipids are preferably derived from a compound having C₁₀-C₂₄Acyl groups of fatty acids of the carbon chain. More preferably, the acyl group is lauroyl, myristoyl, palmitoyl, stearoyl, or oleoyl. In particularly preferred embodiments, the non-cationic lipid comprises one or more of cholesterol, DOPE, or ESM.

The non-cationic lipid typically comprises from about 5 mol% to about 90 mol%, from about 10 mol% to about 85 mol%, from about 20 mol% to about 80 mol%, or about 20 mol% of the total lipid present in the particle. The particles may also include cholesterol. If present, the cholesterol typically comprises from about 0 mol% to about 10 mol%, from about 2 mol% to about 10 mol%, from about 10 mol% to about 60 mol%, from about 12 mol% to about 58 mol%, from about 20 mol% to about 55 mol%, from about 30 mol% to about 50 mol%, or about 48 mol% of the total lipid present in the particle.

3. Double layer stabilizing component

In addition to cationic and non-cationic lipids, the stabilized nucleic acid-lipid particles of the invention can include Bilayer Stabilizing Components (BSCs), such as ATTA-lipids or PEG-lipids, such as PEG coupled to dialkoxypropyl groups (PEG-DAAs), for example, as described in PCT publication No. WO 05/026372, PEG coupled to diacylglycerols (PEG-DAGs), for example, as described in U.S. patent publications nos. 20030077829 and 2005008689, PEG coupled to phospholipids such as phosphatidylethanolamine (PEG-PE), PEG conjugated to ceramides, or mixtures thereof (see, e.g., U.S. patent No. 5,885,613). In a preferred embodiment, the BSC is a conjugated lipid that inhibits aggregation of particles. Suitable conjugated lipids include, but are not limited to, PEG-lipid conjugates, ATTA-lipid conjugates, cation-polymer-lipid Conjugates (CPLs), and mixtures thereof. In another preferred embodiment, the particle comprises a PEG-lipid conjugate or an ATTA-lipid conjugate together with CPL.

PEG is a linear, water-soluble polymer of ethylene PEG repeating units having two terminal hydroxyl groups. PEGs are classified according to their molecular weight: for example, PEG 2000 has an average molecular weight of about 2,000 daltons and PEG 5000 has an average molecular weight of about 5,000 daltons. PEGs are commercially available from Sigma Chemical company (Sigma Chemical Co.) and other companies, and include, for example, the following: monomethoxypolyethylene glycol (MePEG-OH), monomethoxypolyethylene glycol-succinate (MePEG-S), monomethoxypolyethylene glycol-succinimidyl succinate (MePEG-S- -NHS), monomethoxypolyethylene glycol-amine (MePEG-NH) ₂) Monomethoxypolyethylene glycol-tresylate (MePEG-TRES), and monomethoxypolyethylene glycol-imidazolyl-carbonyl (MePEG-IM). In addition, monomethoxypolyethylene glycol-acetic acid (MePEG-CH)₂COOH) are particularly useful for preparing PEG-lipid conjugates, including, for example, PEG-DAA conjugates.

In a preferred embodiment, the PEG has an average molecular weight of about 550 daltons to about 10,000 daltons, more preferably about 750 daltons to about 5,000 daltons, more preferably about 1,000 daltons to about 5,000 daltons, more preferably about 1,500 daltonsAn average molecular weight of from about 3,000 daltons, and even more preferably about 2,000 daltons, or about 750 daltons. PEG may be optionally substituted with alkyl, alkoxy, acyl, or aryl groups. PEG may be conjugated directly to the lipid, or may be attached to the lipid through a linker moiety. Any linker moiety suitable for coupling PEG to a lipid can be used, including, for example, ester-free linker moieties and ester-containing linker moieties. In a preferred embodiment, the linker moiety is an ester-free linker moiety. As used herein, the term "ester-free linker moiety" refers to a linker moiety that does not contain a carboxylic ester linkage (-OC (O) -). Suitable ester-free linker moieties include, but are not limited to, amido (-C (O) NH-), amino (-NR-), carbonyl (-C (O) -), carbamate (-NHC (O) O-), urea (-NHC (O) NH-), disulfide (-S-), ether (-O-), succinyl (- (O) CCH-) ₂CH₂C (O) -), succinamido (-NHC (O) CH)₂CH₂C (O) NH-), ethers, disulfides, and combinations thereof (such as linkers containing a urethane linker moiety and an amino linker moiety). In a preferred embodiment, a carbamate linker is used to couple the PEG to the lipid.

In other embodiments, an ester-containing linker moiety is used to couple the PEG to the lipid. Suitable ester-containing linker moieties include, for example, carbonates (-OC (O) O-), succinyls (-Succinyls (-O-), phosphates (-O- (O) POH-O-), sulfonates, and combinations thereof.

Phosphatidylethanolamines with multiple acyl chain groups of different chain lengths and degrees of saturation can be conjugated to PEG to form a bilayer stabilizing component. Such phosphatidylethanolamines are commercially available or can be isolated or synthesized using conventional techniques known to those skilled in the art. Preferably having a carbon chain length of C₁₀To C₂₀Phosphatidylethanolamines of a range of saturated or unsaturated fatty acids. Phosphatidylethanolamines with mono-or di-unsaturated fatty acids and mixtures of saturated and unsaturated fatty acids can also be used. Suitable phosphatidylethanolamines include, but are not limited to, dimyristoyl-phosphatidylethanolamine (DM) PE), dipalmitoyl-phosphatidylethanolamine (DPPE), dioleoyl phosphatidylethanolamine (DOPE), and distearoyl-phosphatidylethanolamine (DSPE).

The term "ATTA" or "polyamide" refers to, but is not limited to, the compounds disclosed in U.S. Pat. Nos. 6,320,017 and 6,586,559. These compounds include compounds having the formula:

wherein R is a member selected from the group consisting of hydrogen, alkyl, and acyl; r¹Is a member selected from the group consisting of hydrogen and alkyl; or optionally, R and R¹Form an azide moiety with the nitrogen to which they are bound; r²Is a member selected from the group of hydrogen, optionally substituted alkyl, optionally substituted aryl and the side chain of an amino acid; r³Is selected from the group consisting of hydrogen, halo, hydroxy, alkoxy, mercapto, hydrazino, amino and NR⁴R⁵A member of the group consisting of wherein R⁴And R⁵Is independently hydrogen or alkyl; n is 4 to 80; m is 2 to 6; p is 1 to 4; and q is 0 or 1. It will be appreciated by those skilled in the art that other polyamides may be used in the compounds of the present invention.

The term "diacylglycerol" refers to a compound having a 2-fatty acyl chain R¹And R²A compound of (1), R¹And R²Each independently having from 2 to 30 carbon atoms bonded to the 1-and 2-positions of glycerol via ester linkages. The acyl groups may be saturated or have different degrees of unsaturation. Suitable acyl groups include, but are not limited to, lauryl (C12), myristyl (C14), palmityl (C16), stearyl (C18), and eicosyl (C20). In a preferred embodiment, R ¹And R²Are identical, i.e. R¹And R²Are all myristyl (i.e., dimyristyl), R¹And R²Are both stearoyl (i.e., distearoyl), and the like. Diacylglycerols have the following general formula:

the term "dialkoxypropyl" refers to a compound having a 2-alkyl chain R¹And R²The compound of (1), R¹And R²Each independently having from 2 to 30 carbons. The alkyl groups may be saturated or have different degrees of unsaturation. The dialkoxypropyl group has the general formula:

in a preferred embodiment, the PEG-lipid is a PEG-DAA conjugate having the formula:

wherein R is¹And R²Independently selected, and is a long chain alkyl group having from about 10 to about 22 carbon atoms; PEG is polyethylene glycol; and L is an ester-free linker moiety or an ester-containing linker moiety as described above. The long chain alkyl group may be saturated or unsaturated. Suitable alkyl groups include, but are not limited to, lauryl (C12), myristyl (C14), palmityl (C16), stearyl (C18), and eicosyl (C20). In a preferred embodiment, R¹And R²Are identical, i.e. R¹And R²Are all myristyl (i.e., dimyristyl), R¹And R²Are both stearoyl (i.e., distearoyl), and the like.

In formula VI above, the PEG has an average molecular weight of about 550 daltons to about 10,000 daltons, more preferably about 750 daltons to about 5,000 daltons, more preferably about 1,000 daltons to about 5,000 daltons, more preferably about 1,500 daltons to about 3,000 daltons, and even more preferably about 2,000 daltons, or about 750 daltons. PEG may be optionally substituted with alkyl, alkoxy, acyl, or aryl groups. In a preferred embodiment, the terminal hydroxyl group is substituted with a methoxy or methyl group.

In a preferred embodiment, "L" is an ester-free linker moiety. Suitable ester-free linkers include, but are not limited to, amido linker moieties, amino linker moieties, carbonyl linker moieties, carbamate linker moieties, urea linker moieties, ether linker moieties, disulfide linker moieties, succinamido linker moieties, and combinations thereof. In a preferred embodiment, the ester-free linker moiety is a urethane linker moiety (i.e., a PEG-C-DAA conjugate). In another preferred embodiment, the ester-free linker moiety is an amido linker moiety (i.e., a PEG-a-DAA conjugate). In another preferred embodiment, the ester-free linker moiety is a succinamide-based linker moiety (i.e., a PEG-S-DAA conjugate).

The PEG-DAA conjugates are synthesized using standard techniques and reagents known to those skilled in the art. It should be recognized that the PEG-DAA conjugates should contain various amide, amine, ether, thio, carbamate, and urea linkages. Those skilled in the art will recognize that methods and reagents for forming these bonds are well known and readily available. See, for example, March, advanced ORGANIC CHEMISTRY (Wiley 1992), Larock, Complex ORGANIC CHEMISTRY TRANSFORMATIONS (VCH 1989); and Furniss, VOGEL' S TEXTBOOK OF organic CHEMISTRY, 5 th edition (langman, 1989). It will be appreciated that at various points in the synthesis of the PEG-DAA conjugate, any functional groups present may need to be protected and deprotected. Those skilled in the art will appreciate that such techniques are well known. See, for example, Green and Wuts, PROTECTIVE group in ORGANIC Synthesis ORGANIC SYNTHESIS (Wiley 1991).

Preferably, the PEG-DAA conjugate is a dilauryloxypropyl (C12) -PEG conjugate, a dimyristoyloxypropyl (C14) -PEG conjugate, a dipalmitoyloxypropyl (C16) -PEG conjugate, or a distearoyloxypropyl (C18) -PEG conjugate. It will be readily understood by those skilled in the art that other dialkoxypropyl species may be used in the PEG-DAA conjugates of the present invention.

In addition to the foregoing, one skilled in the art will readily appreciate that other hydrophilic polymers may be used in place of PEG. Examples of suitable polymers that may be used in place of PEG include, but are not limited to, polyvinylpyrrolidone, polymethyloxazoline, polyethyloxazoline, polyhydroxypropylmethacrylamide, polymethacrylamide and polydimethylacrylamide, polylactic acid, polyglycolic acid, and derivatized celluloses, such as hydroxymethylcellulose or hydroxyethylcellulose.

In addition to the foregoing ingredients, the particles of the invention (e.g., SNALPs or SPLPs) may further comprise cationic poly (ethylene glycol) (PEG) lipids or CPLs designed to be inserted into the lipid bilayer to impart a positive charge (see, e.g., Chen et al, bioconj. chem. (bioconjugate chemistry), 11: 433-437 (2000)). Suitable SPLPs and SPLP-CPLs for use in the present invention, as well as methods of making and using the same, are disclosed, for example, in U.S. Pat. No. 6,852,334 and PCT publication No. WO 00/62813. The Cationic Polymeric Lipids (CPLs) used in the present invention have the following structural features: (1) a lipid anchor, such as a hydrophobic lipid, for incorporating the CPLs into a lipid bilayer; (2) a hydrophilic spacer, such as polyethylene glycol, for linking the lipid anchor to the cationic head group; and (3) a polycationic moiety, such as a naturally occurring amino acid, to produce a protonated cationic head group.

Suitable CPLs include compounds of formula VII:

A-W-Y(VII)，

wherein A, W, and Y are as described below.

Referring to formula VII, "a" is a lipid moiety, such as an amphiphilic lipid, a neutral lipid, or a hydrophobic lipid that acts as a lipid anchor. Examples of suitable lipids include vesicle-forming lipids, or lipids employing vesicles, and include, but are not limited to, diacylglycerols, dialkylglycerols, N-N-dialkylamino, 1, 2-diacyloxy-3-aminopropanes, and 1, 2-dialkyl-3-aminopropanes.

"W" is a polymer or oligomer, such as a hydrophilic polymer or oligomer. Preferably, the hydrophilic polymer is a biocompatible polymer that is non-immunogenic or has low inherent immunogenicity. Alternatively, the hydrophilic polymer may be weakly antigenic if used with an appropriate adjuvant. Suitable non-immunogenic polymers include, but are not limited to, PEG, polyamides, polylactic acid, polyglycolic acid, polylactic acid/polyglycolic acid copolymers, and combinations thereof. In a preferred embodiment, the polymer has a molecular weight of about 250 to about 7,000 daltons.

"Y" is a polycationic moiety. The term polycationic moiety refers to a compound, derivative or functional group having a positive charge, preferably at least 2 positive charges, at a selected pH, preferably physiological pH. Suitable polycationic moieties include basic amino acids and their derivatives such as arginine, asparagine, glutamine, lysine and histidine; spermine; spermidine; a cationic dendrimer; a polyamine; a polyamino sugar; and an aminoglycan. The polycation moiety may be linear in structure, such as linear tetra-lysine, branched or dendrimeric. At the selected pH, the polycation moiety has from about 2 to about 15 positive charges, preferably from about 2 to about 12 positive charges, and more preferably from about 2 to about 8 positive charges. The choice of which polycationic moiety to use may be determined by the type of particulate application desired.

The charges on the polycationic moieties may be distributed along the entire particle moiety or alternatively they may be discrete concentrations of charge intensity on a particular region of the particle moiety, for example, a charge spike. If the charge intensity is distributed on the particles, the charge intensity may be uniformly distributed, or may be non-uniformly distributed. All changes in the charge distribution of the polycationic moiety are encompassed by the present invention.

The lipid "a" and the non-immunogenic polymer "W" may be attached by various methods, and preferably by covalent attachment. Methods known to those skilled in the art can be used for covalent attachment of "a" and "W". Suitable linkages include, but are not limited to, amide, amine, carboxyl, carbonate, carbamate, ester, and hydrazone linkages. It should be clear to one skilled in the art that "A" and "W" must have complementary functional groups to complete the linkage. The reaction of such two groups, one on the lipid and the other on the polymer, will provide the desired bonding. For example, when the lipid is a diacylglycerol and the terminal hydroxyl group is activated with, for example, NHS and DCC to form an active ester, followed by reaction with a polymer containing an amino group, such as with a polyamide (see, e.g., U.S. Pat. nos. 6,320,017 and 6,586,559), an amide bond will be formed between the two groups.

In some cases, the polycationic moiety may have a ligand attached to it, such as a targeting ligand or a chelating moiety that complexes calcium. Preferably, the cationic moiety maintains a positive charge after the ligand is attached. In some cases, the attached ligand has a positive charge. Suitable ligands include, but are not limited to, compounds or devices having reactive functional groups and include lipids, amphipathic lipids, carrier compounds, bioaffinity compounds, biomaterials, biopolymers, biomedical devices, analytically detectable compounds, therapeutically active compounds, enzymes, peptides, proteins, antibodies, immunostimulants, radiolabels, fluorophores, biotin, drugs, haptens, DNA, RNA, polysaccharides, liposomes, viral particles, micelles, immunoglobulins, functional groups, other targeting moieties or toxins.

The bilayer stabilizing component (e.g., PEG-lipid) typically comprises from about 0 mol% to about 20 mol%, from about 0.5 mol% to about 20 mol%, from about 1.5 mol% to about 18 mol%, from about 4 mol% to about 15 mol%, from about 5 mol% to about 12 mol%, or about 2 mol% of the total lipid present in the particle. It will be appreciated by those of ordinary skill in the art that the concentration of the bilayer stabilizing component may vary depending on the bilayer stabilizing component used and the rate at which the nucleic acid-lipid particle becomes fusogenic.

By controlling the composition and concentration of the bilayer stabilizing component, one can control the rate at which the bilayer stabilizing component is exchanged out of the nucleic acid-lipid particle, and in turn, the rate at which the nucleic acid-lipid particle becomes fusogenic. For example, when a polyethylene glycol-phosphatidylethanolamine conjugate or a polyethylene glycol-ceramide conjugate is used as the bilayer stabilizing component, the rate at which the nucleic acid-lipid particle becomes fusogenic may be varied, for example, by varying the concentration of the bilayer stabilizing component, by varying the molecular weight of the polyethylene glycol, or by varying the chain length and degree of saturation of the acyl chain group on the phosphatidylethanolamine or ceramide. In addition, other variables including, for example, pH, temperature, ionic strength, etc., can be used to alter and/or control the rate at which the nucleic acid-lipid particle becomes fusogenic. Other methods that can be used to control the nucleic acid-lipid particle becoming fusogenic will be apparent to those of skill in the art upon reading this disclosure.

B. Other carrier systems

Non-limiting examples of other lipid-based carrier systems suitable for use in the present invention include lipid-nucleic acid complexes (see, e.g., U.S. patent publication No. 20030203865; and Zhang et al, j.control Release (journal of controlled Release), 100: 165-180(2004)), pH-sensitive lipid-nucleic acid complexes (see, e.g., U.S. patent publication No. 20020192275), reversibly masked lipid-nucleic acid complexes (see, e.g., U.S. patent publication No. 20030180950), cationic lipid-based compositions (see, e.g., U.S. patent No. 6,756,054; and U.S. patent publication No. 20050234232), cationic liposomes (see, e.g., U.S. patent publication No. 20030229040, 20020160038, and 20020012998; U.S. patent publication No. 5,908,635; and PCT publication No. WO01/72283), anionic liposomes (see, e.g., U.S. patent publication No. 20030026831), pH-sensitive liposomes (see, for example, U.S. patent publication nos. 20020192274; and AU2003210303), antibody-coated liposomes (see, e.g., U.S. patent publication nos. 20030108597; and PCT publication nos. WO 00/50008), cell-type specific liposomes (see, e.g., U.S. patent publication No. 20030198664), liposomes containing nucleic acids and peptides (see, e.g., U.S. patent publication No. 6,207,456), liposomes containing lipids derivatized with a releasable hydrophilic polymer (see, e.g., U.S. patent publication No. 20030031704), lipid-embedded nucleic acids (see, e.g., PCT publication nos. WO 03/057190 and WO 03/059322), lipid-encapsulated nucleic acids (see, e.g., U.S. patent publication No. 20030129221; and U.S. patent No. 5,756,122), other liposome compositions (see, e.g., U.S. patent publication nos. 20030035829 and 20030072794; and U.S. patent No. 6,200,599), mixtures of stable liposomes and emulsions (see, e.g., EP1304160), emulsion compositions (see, e.g., U.S. patent No. 6,747,014), and nucleic acid microemulsions (see, e.g., U.S. patent publication No. 20050037086).

Examples of polymer-based carrier systems suitable for use in the present invention include, but are not limited to, cationic polymer-nucleic acid complexes (i.e., polymer-nucleic acid complexes). To form polymer-nucleic acid complexes, nucleic acids (e.g., siRNA) are typically complexed with cationic polymers having a linear, branched, star-shaped, or dendrimer structure that concentrates the nucleic acids into positively charged particles that are capable of interacting with anionic proteoglycans on the cell surface and entering the cell by endocytosis. In some embodiments, the polymer-nucleic acid complex comprises a nucleic acid complexed with a cationic polymer (e.g., a cationic polymerE.g., siRNA), such as Polyethylenimine (PEI) (see, e.g., U.S. patent nos. 6,013,240; in vivo jetPEI commercially available from Qbiogene, Inc. (Carlsbad, Calif.)^TMLinear forms of PEI), polypropyleneimine (PPI), polyvinylpyrrolidone (PVP), poly-L-lysine (PLL), Diethylaminoethyl (DEAE) -dextran, poly (β -amino ester) (PAE) polymers (see, e.g., Lynn et al, j.am.chem.soc. (journal of the american chemical association), 123: 8155-8156(2001)), chitosan, Polyamidoamine (PAMAM) dendrimers (see, e.g., Kukowska-Latallo et al, proc.natl.acad.sci.usa (proceedings of the national academy of sciences usa), 93: 4897-4902(1996)), porphyrins (see, e.g., U.S. patent No. 6,620,805), polyvinyl ethers (see, e.g., U.S. patent publication No. 20040156909), polycyclic amidinium (see, e.g., U.S. patent publication No. 20030220289), other polymers that include primary amine, imine, guanidine, and/or imidazole groups (see, e.g., U.S. patent No. 6,013,240; PCT publication Nos. WO/9602655; PCT publication No. W095/21931; zhang et al, j.control Release (journal of controlled Release), 100: 165-180 (2004); and tira et al, curr. gene Ther (modern gene therapy), 6: 59-71(2006)), and mixtures thereof. In other embodiments, the polymer-nucleic acid complexes include cationic polymer-nucleic acid complexes as described in U.S. patent publication nos. 20060211643, 20050222064, 20030125281, and 20030185890, and PCT publication No. WO 03/066069; biodegradable poly (β -amino ester) polymer-nucleic acid complexes as described in U.S. patent publication No. 20040071654; microparticles comprising a polymer matrix as described in U.S. patent publication No. 20040142475; other particulate compositions as described in U.S. patent publication No. 20030157030; concentrated nucleic acid complexes as described in U.S. patent publication No. 20050123600; and nanocapsule and microcapsule compositions as described in AU2002358514 and PCT publication No. WO 02/096551.

In some cases, the modified siRNA molecule may be complexed with a cyclodextrin or a polymer thereof. Non-limiting examples of cyclodextrin-based carrier systems include the cyclodextrin-modified polymer-nucleic acid complexes described in U.S. patent publication No. 20040087024; linear cyclodextrin copolymer-nucleic acid complexes described in U.S. Pat. nos. 6,509,323, 6,884,789, and 7,091,192; and cyclodextrin polymer-complexing agent-nucleic acid complexes described in U.S. patent No. 7,018,609. In certain other cases, the modified siRNA molecule may be complexed with a peptide or polypeptide. Examples of protein-based carrier systems include, but are not limited to, cationic oligopeptide-nucleic acid complexes described in PCT publication No. WO 95/21931.

V. preparation of nucleic acid-lipid particles

The serum-stable nucleic acid-lipid particles of the invention, wherein the modified siRNA described herein is encapsulated in a lipid bilayer and protected from degradation, can be prepared by any method known in the art including, but not limited to, continuous mixing methods, direct dilution treatments, detergent dialysis methods, or modified reverse phase methods that use organic solvents to provide a single phase during component mixing.

In a preferred embodiment, the cationic lipid is a lipid of formula I and II or a combination thereof. In other preferred embodiments, the non-cationic lipid is Egg Sphingomyelin (ESM), distearoyl phosphatidylcholine (DSPC), dioleoyl phosphatidylcholine (DOPC), 1-palmitoyl-2-oleoyl-phosphatidylcholine (POPC), dipalmitoyl-phosphatidylcholine (DPPC), monomethyl-phosphatidylethanolamine, dimethyl-phosphatidylethanolamine, 14:0PE (1, 2-dimyristoyl-phosphatidylethanolamine (DMPE)), 16:0PE (1, 2-dipalmitoyl-phosphatidylethanolamine (DPPE)), 18:0PE (1, 2-distearoyl-phosphatidylethanolamine (DSPE)), 18:1PE (1, 2-dioleoyl-phosphatidylethanolamine (DOPE)), 18:1 trans PE (1, 2-elaidoleoyl-phosphatidylethanolamine (DEPE)), 18:0-18:1PE (1-stearoyl-2-oleoyl-phosphatidylethanolamine (SOPE)), 16:0-18:1PE (1-palmitoyl-2-oleoyl-phosphatidylethanolamine (POPE)), a polyethylene glycol-based polymer (see, e.g., PEG 2000, PEG 5000, PEG-modified diacylglycerol, or PEG-modified dialkyloxypropyl), cholesterol, or a combination thereof. In other preferred embodiments, the organic solvent is methanol, chloroform, dichloromethane, ethanol, diethyl ether, or a combination thereof.

In a preferred embodiment, the present invention provides a method for producing nucleic acid-lipid particles by a continuous mixing process, for example, the method comprising providing an aqueous solution comprising a nucleic acid, such as siRNA, in a first reservoir, providing an organic lipid solution in a second reservoir, and mixing the aqueous solution and the organic lipid solution such that the organic lipid solution is mixed with the aqueous solution to substantially instantaneously produce liposomes encapsulating the nucleic acid (e.g., siRNA). Such a method and apparatus for carrying out the method are described in detail in U.S. patent publication No. 20040142025.

The continuous introduction of lipid and buffer solution into a mixing environment, such as into a mixing chamber, acts to continuously dilute the lipid solution with the buffer solution, thereby producing liposomes substantially at the instant of mixing. As used herein, the phrase "serially diluting a lipid solution with a buffer solution" (and variations) generally refers to diluting the lipid solution sufficiently rapidly during hydration with sufficient force to accomplish vesicle formation. By mixing an aqueous solution including nucleic acid and an organic lipid solution, the organic lipid solution undergoes continuous stepwise dilution in the presence of a buffer solution (i.e., an aqueous solution), thereby producing nucleic acid-lipid particles.

The serum-stable nucleic acid-lipid particles formed using a continuous mixing process typically have a size of about 50nm to about 150nm, about 60nm to about 130nm, about 70nm to about 110nm, or about 70nm to about 90 nm. The particles so formed are not agglomerated and are optionally sized to obtain a uniform particle size.

In another embodiment, the present invention provides a method for producing nucleic acid-lipid particles by a direct dilution method comprising forming a liposome solution and immediately and directly introducing the liposome solution into a collection vessel containing a controlled amount of dilution buffer. In a preferred aspect, the collection vessel comprises one or more elements arranged to agitate the contents of the collection vessel to facilitate dilution. In one aspect, the amount of dilution buffer present in the collection vessel is substantially equal to the volume of liposome solution introduced therein. As a non-limiting example, a solution of liposomes in 45% ethanol will advantageously produce smaller particles in about 22.5%, about 20%, or about 15% ethanol when introduced into a collection vessel containing an equal volume of ethanol.

In another embodiment, the present invention provides for the production of nucleic acid-lipid particles by a direct dilution method, wherein a third reservoir containing a dilution buffer is fluidly coupled to the second mixing zone. In this embodiment, the liposome solution formed in the first mixing zone is mixed immediately and directly with the dilution buffer in the second mixing zone. In a preferred aspect, the second mixing zone comprises a T-connector arranged such that the liposome solution and dilution buffer streams merge as opposing 180 ° streams; however, connectors providing smaller angles may be used, for example, from about 27 ° to about 180 °. The pump mechanism delivers a controllable flow of buffer to the second mixing zone. In one aspect, the flow rate of the dilution buffer provided to the second mixing zone is controlled to be substantially equal to the flow rate of the liposome solution introduced therein from the first mixing zone. This embodiment advantageously allows for more control of the dilution buffer flow mixed with the liposome solution in the second mixing zone, and thus also more control of the concentration of the liposome solution in the buffer throughout the second mixing process. Such dilution buffer flow rate control advantageously allows small particle sizes to be formed at reduced concentrations.

These methods and apparatus for carrying out these direct dilution methods are described in detail in U.S. patent application No. 11/495,150.

The serum-stable nucleic acid-lipid particles formed using the direct dilution method typically have a size of about 50nm to about 150nm, about 60nm to about 130nm, about 70nm to about 110nm, or about 70nm to about 90 nm. The particles so formed are not aggregated and are optionally sized to achieve a uniform particle size.

In some embodiments, the particles are formed using detergent dialysis. Without wishing to be bound by any particular mechanism of formation, a nucleic acid, such as an siRNA, is contacted with a detergent solution of a cationic lipid to form a coated nucleic acid complex. These coated nucleic acids can aggregate and precipitate. However, the presence of detergent reduces this aggregation and allows the coated nucleic acids to react with excess lipid (typically, non-cationic lipid) to form particles in which the nucleic acids are encapsulated in a lipid bilayer. Thus, the serum-stable nucleic acid-lipid particles can be prepared as follows:

(a) combining a nucleic acid with a cationic lipid in a detergent solution to form a coated nucleic acid-lipid complex;

(b) Contacting a non-cationic lipid with the coated nucleic acid-lipid complex to form a detergent solution comprising the nucleic acid-lipid complex and the non-cationic lipid; and

(c) dialyzing the detergent solution in step (b) to provide a solution of serum-stable nucleic acid-lipid particles, wherein the nucleic acid is encapsulated in a lipid bilayer and the particles are serum-stable and have a size between about 50 to about 150 nm.

An initial solution of coated nucleic acid-lipid complexes is formed by combining the nucleic acid with a cationic lipid in a detergent solution. In these embodiments, the detergent solution is preferably an aqueous solution of a neutral detergent having a critical micelle concentration of 15-300mM, more preferably 20-50 mM. Examples of suitable detergents include, for example, N' - ((octanoylimino) -di- (1, 3-propylene)) -di- (D-glucamide) (BIGCHAP); BRIJ 35; deoxy-BIGCHAP; dodecyl poly (ethylene glycol) ether; tween 20; tween 40; tween 60; tween 80; tween 85; mega 8; mega 9; zwittergent3-08；Zwittergent3-10; triton X-405; hexyl-, heptyl-, octyl-and nonyl-beta-D-glucopyranosides; and heptyl thioglucopyranoside; among them, octyl β -D-glucopyranoside and Tween 20 are most preferable. The concentration of the detergent in the detergent solution is typically about 100mM to about 2M, preferably about 200mM to about 1.5M.

The cationic lipid and nucleic acid will typically combine to produce a charge ratio (+/-) of about 1: 1 to about 20: 1, a ratio of about 1: 1 to about 12: 1, a ratio of about 2: 1 to about 6: 1. In addition, the total concentration of nucleic acid in the solution will typically be from about 25. mu.g/ml to about 1mg/ml, from about 25. mu.g/ml to about 200. mu.g/ml, or from about 50. mu.g/ml to about 100. mu.g/ml. The binding of nucleic acids to cationic lipids in detergent solutions is typically maintained at room temperature for a period of time sufficient to form a coated complex. Alternatively, the nucleic acid and cationic lipid may be combined in a detergent solution and heated to a temperature of up to about 37 ℃, about 50 ℃, about 60 ℃, or about 70 ℃. For nucleic acids that are particularly sensitive to temperature, the coated complexes can be formed at lower temperatures, typically as low as about 4 ℃.

In some embodiments, the ratio of nucleic acid to lipid (mass/mass ratio) in the formed nucleic acid-lipid particle ranges between about 0.01 to about 0.2, about 0.02 to about 0.1, about 0.03 to about 0.1, or about 0.01 to about 0.08. The ratio of the starting materials is also within this range. In other embodiments, the nucleic acid-lipid particle formulation uses about 400 μ g nucleic acid/10 mg total lipid, or a nucleic acid to lipid ratio of about 0.01 to about 0.08, more preferably about 0.04, the ratio of 0.04 corresponding to 1.25mg total lipid/50 μ g nucleic acid. In other preferred embodiments, the particle has a nucleic acid of about 0.08: lipid mass ratio.

Next, a detergent solution of the coated nucleic acid-lipid complex is allowed to stand with a non-cationic lipidContacting the lipid to provide a detergent solution of the nucleic acid-lipid complex and the non-cationic lipid. Non-cationic lipids useful in this step include diacylphosphatidylcholine, diacylphosphatidylethanolamine, ceramide, sphingomyelin, cephalin, cardiolipin and cerebroside. In preferred embodiments, the non-cationic lipid is a diacylphosphatidylcholine, a diacylphosphatidylethanolamine, a ceramide, or a sphingomyelin. The acyl groups in these lipids are preferably derived from having C₁₀-C₂₄Acyl groups of fatty acids of the carbon chain. More preferably, the acyl group is lauroyl, myristoyl, palmitoyl, stearoyl, or oleoyl. In particularly preferred embodiments, the non-cationic lipid is DSPC, DOPE, POPC, Egg Phosphatidylcholine (EPC), cholesterol, or mixtures thereof. In a most preferred embodiment, the nucleic acid-lipid particle is a fused particle having enhanced properties in vivo, and the non-cationic lipid is DSPC or DOPE. In addition, the nucleic acid-lipid particle of the present invention may further comprise cholesterol. In other preferred embodiments, the non-cationic lipid may also comprise polyethylene glycol-based polymers, such as PEG 2,000, PEG 5,000, and PEG conjugated to diacylglycerol, ceramide, or phospholipids, as described, for example, in U.S. patent No. 5,820,873 and in U.S. patent publication No. 2003/0077829. In other preferred embodiments, the non-cationic lipid may also comprise polyethylene glycol-based polymers such as PEG 2,000, PEG 5,000 and PEG conjugated to a dialkoxypropyl group.

The amount of non-cationic lipid used in the present methods is typically about 2 to about 20mg of total lipid to 50 μ g of nucleic acid. Preferably, the amount of total lipid is from about 5 to about 10mg/50 μ g of nucleic acid.

After formation of the detergent solution of the nucleic acid-lipid complex and the non-cationic lipid, the detergent is removed, preferably by dialysis. Removal of the detergent results in the formation of a lipid bilayer surrounding the nucleic acid, thereby providing serum-stable nucleic acid-lipid particles having a size of about 50nm to about 150nm, about 60nm to about 130nm, about 70nm to about 110nm, or about 70nm to about 90 nm. The particles so formed are not agglomerated and are optionally sized to achieve a uniform particle size.

The serum-stable nucleic acid-lipid particles can be sized by any useful method to size the liposomes. Sizing may be performed to achieve a desired size range and relatively narrow particle size distribution.

Several techniques may be used to size the particles to a desired size. U.S. patent No. 4,737,323 describes a method for liposomes and equally applicable to one size arrangement of the present particles. Sonication of the particle suspension by a water bath or probe sonication produces progressive size reduction to particles of a size less than about 50 nm. Homogenization is another method that relies on shear forces to break larger pieces of particles into smaller particles. In a typical homogenization process, the particles are recirculated through a standard emulsion homogenizer until a selected particle size is observed, typically between about 60 and about 80 nm. In both methods, the particle size distribution can be monitored by conventional laser beam particle size discrimination, or QELS.

Extrusion of particles through small pore polycarbonate membranes or asymmetric ceramic membranes is also an effective method of reducing particle size to a relatively well defined size distribution. Typically, the suspension is circulated through the membrane once or several times until the desired particle size distribution is obtained. The particles may be pressed through a continuous smaller pore membrane, thereby obtaining a gradual reduction in size.

In another set of embodiments, the serum-stable nucleic acid-lipid particle can be prepared as follows:

(a) preparing a mixture comprising a cationic lipid and a non-cationic lipid in an organic solvent;

(b) contacting an aqueous solution of nucleic acid with the mixture in step (a) thereby providing a clear single phase; and

(c) removing the organic solvent to provide a suspension of nucleic acid-lipid particles, wherein the nucleic acid is encapsulated in a lipid bilayer, the particles being stable in serum and having a size of about 50 to about 150 nm.

The nucleic acids (e.g., siRNA), cationic lipids, and non-cationic lipids used in this set of embodiments are as described for the detergent dialysis methods described above.

The choice of organic solvent will typically include considerations of solvent polarity and the ease with which the solvent can be removed late in particle formation. The organic solvent, which may also be used as a solubilizing agent, is present in an amount sufficient to provide a clear single phase mixture of nucleic acids and lipids. Suitable solvents include, but are not limited to, chloroform, dichloromethane, diethyl ether, cyclohexane, cyclopentane, benzene, toluene, methanol, or other aliphatic alcohols such as propanol, isopropanol, butanol, tert-butanol, isobutanol, pentanol, and hexanol. Combinations of two or more solvents may also be used in the present invention.

Contacting the nucleic acids with organic solutions of cationic and non-cationic lipids is accomplished by mixing together a first solution of the nucleic acids, which is typically an aqueous solution, and a second organic solution of the lipids. It will be appreciated by those skilled in the art that this mixing can occur by a variety of methods, for example, by mechanical means such as by using a vortex mixer.

After the nucleic acid has been contacted with the organic solution of lipids, the organic solvent is removed, thus forming an aqueous suspension of serum-stable nucleic acid-lipid particles. The method for removing the organic solvent will typically comprise evaporation under reduced pressure or blowing a stream of inert gas (e.g. nitrogen or argon) through the mixture.

The serum-stable nucleic acid lipid particles so formed will typically range in size from about 50nm to about 150nm, from about 60nm to about 130nm, from about 70nm to about 110nm, or from about 70nm to about 90 nm. To obtain further particle size reduction or uniformity of size, sieving may be performed as described above.

In other embodiments, the method will further comprise the addition of a non-lipid polycation that can be used to effect delivery to cells using the present compositions. Examples of suitable non-lipid polycations include, but are not limited to, hexadimethrine bromide (POLYBRENE bromide), under the trade name POLYBRENE Sold, commercially, from Aldrich Chemical Co., Milwaukee, Wisconsin, USA) or other hexadimorphine salts. Other suitable polycations include, for example, salts of poly-L-ornithine, poly-L-arginine, poly-L-lysine, poly-D-lysine, polyallylamine, and polyethyleneimine.

In certain embodiments, the formation of the nucleic acid-lipid particles can be performed in a single phase system (e.g., Bligh and Dyer single phase or similar mixtures of aqueous and organic solvents) or a two phase system, with appropriate mixing.

When complex formation is performed in a single-phase system, the cationic lipid and nucleic acid are each dissolved in a volume of single-phase mixture. The combination of the two solutions provides a single mixture in which the complex is formed. Alternatively, the complex may be formed in a biphasic mixture in which a cationic lipid binds to the nucleic acid (which is present in the aqueous phase) and "pushes" it into the organic phase.

In another embodiment, the serum-stable nucleic acid-lipid particle may be prepared as follows:

(a) contacting a nucleic acid with a solution comprising a non-cationic lipid and a detergent to form a nucleic acid-lipid mixture;

(b) Contacting a cationic lipid with the nucleic acid-lipid mixture to neutralize negatively charged portions of the nucleic acids and form a neutralized charge mixture of the nucleic acids and the lipid; and

(c) removing the detergent from the charge-neutralized mixture to provide nucleic acid-lipid particles, wherein the nucleic acid is protected from degradation.

In one set of embodiments, the solution of non-cationic lipid and detergent is an aqueous solution. Contacting the nucleic acid with a solution of a non-cationic lipid and a detergent is typically accomplished by mixing a first solution of the nucleic acid with a second solution of the lipid and the detergent. It will be appreciated by those skilled in the art that the mixing may occur by any number of methods, for example, by mechanical means such as by using a vortex mixer. Preferably, the nucleic acid solution is also a detergent solution. The amount of non-cationic lipid used in the present method is typically determined based on the amount of cationic lipid used and is typically from about 0.2 to about 5 times the amount of cationic lipid, preferably from about 0.5 to about 2 times the amount of cationic lipid used.

In some embodiments, the nucleic acid is pre-concentrated as described, for example, in U.S. patent application No. 09/744,103.

The nucleic acid-lipid mixture thus formed is contacted with a cationic lipid, thereby neutralizing the negatively charged moieties that are associated with the nucleic acids (or other polyanionic material) present. The amount of cationic lipid used will typically be sufficient to neutralize at least 50% of the negative charge of the nucleic acid. Preferably, the negative charge will be at least 70% neutralized, more preferably at least 90% neutralized. Cationic lipids useful in the present invention include, for example, DLinDMA and DLenDMA. These lipids and related analogs have been described in U.S. patent publication No. 20060083780.

Contacting the cationic lipid with the nucleic acid-lipid mixture can be accomplished by any of a number of techniques, preferably by mixing together a solution of the cationic lipid and a solution comprising the nucleic acid-lipid mixture. After the two solutions are mixed (or contacted in any other way), the negative charge associated with the nucleic acid is partially neutralized. However, the nucleic acid remains in an uncompressed state and acquires hydrophilic properties.

After the cationic lipid has been contacted with the nucleic acid-lipid mixture, the detergent (or combination of detergent and organic solvent) is removed, thereby forming the nucleic acid-lipid particles. The method for removing the detergent will typically comprise dialysis. When an organic solvent is present, it is typically successfully removed by evaporation under reduced pressure or by blowing a stream of an inert gas (e.g. nitrogen or argon) through the mixture.

The size of the particles so formed will typically range from about 50nm to several microns, about 50nm to about 150nm, about 60nm to about 130nm, about 70nm to about 110nm, or about 70nm to about 90 nm. To further achieve size reduction or uniformity in size of the particles, the nucleic acid-lipid particles can be sonicated, filtered, or otherwise subjected to size-aligning techniques used in liposomal formulations and well known to those skilled in the art.

In other embodiments, the method will further comprise the addition of a non-lipid polycation that can be used to perform lipofection using the compositions of the invention. Examples of suitable non-lipid polycations include, hexadimethrine bromide (under the trade name POLYBRENE)Sold, commercially, from Aldrich chemical Co., Milwaukee, Wisconsin, USA) or other hexadimethrine salts. Other suitable polycations include, for example, salts of poly-L-ornithine, poly-L-arginine, poly-L-lysine, poly-D-lysine, polyallylamine and polyethyleneimine. The addition of these salts is preferably carried out after the particles have been formed.

In another aspect, the serum-stable nucleic acid-lipid particle is prepared by:

(a) Contacting an amount of a cationic lipid with a nucleic acid in a solution; the solution comprises about 15-35% water and about 65-85% organic solvent, and the amount of cationic lipid is sufficient to produce a +/-charge ratio from about 0.85 to about 2.0, thereby providing a hydrophobic nucleic acid-lipid complex;

(b) contacting in solution a hydrophobic, nucleic acid-lipid complex with a non-cationic lipid, thereby providing a nucleic acid-lipid mixture; and

(c) removing the organic solvent from the nucleic acid-lipid mixture to provide a nucleic acid-lipid particle in which the nucleic acid is protected from degradation.

The nucleic acids (e.g., siRNA), non-cationic lipids, cationic lipids and organic solvents effective for use in this aspect of the invention are the same as those described above for the methods using detergents. In one set of embodiments, the solution of step (a) is monophasic. In another set of embodiments, the solution of step (a) is biphasic.

In preferred embodiments, the non-cationic lipid is ESM, DSPC, DOPC, POPC, DPPC, monomethyl-phosphatidylethanolamine, dimethyl-phosphatidylethanolamine, DMPE, DPPE, DSPE, DOPE, DEPE, SOPE, POPE, a PEG-based polymer (e.g., PEG 2000, PEG5000, a PEG-modified diacylglycerol, or a PEG-modified dialkoxypropyl group), cholesterol, or a combination thereof. In other preferred embodiments, the organic solvent is methanol, chloroform, dichloromethane, ethanol, diethyl ether or a combination thereof.

In one embodiment, the nucleic acid is an siRNA as described herein; the cationic lipid is DLindMA, DLenDMA, DODAC, DDAB, DOTMA, DOSPA, DMRIE, DOGS, or a combination thereof; the non-cationic lipid is ESM, DOPE, PEG-DAG, DSPC, DPPC, DPPE, DMPE, monomethyl-phosphatidylethanolamine, dimethyl-phosphatidylethanolamine, DSPE, DEPE, SOPE, POPE, cholesterol, or combinations thereof (e.g., DSPC and PEG-DAA); and the organic solvent is methanol, chloroform, dichloromethane, ethanol, diethyl ether or a combination thereof.

As above, contacting the nucleic acid with the cationic lipid is typically accomplished, preferably mechanically, such as by mixing together a first solution of the nucleic acid, and a second solution of the lipid using a vortex mixer. The resulting mixture comprises the complex as described above. These complexes are then converted into particles by adding non-cationic lipids and removing the organic solvent. The addition of the non-cationic lipid is typically accomplished by simply adding a solution of the non-cationic lipid to the mixture containing the complex. Reverse addition may also be used. Subsequent removal of the organic solvent can be accomplished by methods known to those skilled in the art and also as described above.

The amount of non-cationic lipid used in this aspect of the invention is typically an amount from about 0.2 to about 15 times the amount of cationic lipid used to provide the charge-neutralized nucleic acid-lipid complex (on a molar basis). Preferably, the amount is about 0.5 to about 9 times the amount of cationic lipid used.

In one embodiment, the nucleic acid-lipid particles prepared according to the above methods are net charge neutral or carry an overall charge that provides the particle with greater lipofection activity of the gene. Preferably, the nucleic acid component of the particle is a nucleic acid that interferes with the production of an unwanted protein. In other preferred embodiments, the non-cationic lipid may further comprise cholesterol.

Various general methods of making SNALP-CPLs (SNALPs including CPL) as discussed herein. Two common techniques include "post-insertion" techniques, i.e., the insertion of CPL into, for example, preformed SNALP, and "standard" techniques, in which CPL is included in a lipid mixture during, for example, a SNALP formation step. The post-insertion technique results in SNALPs with CPLs primarily in the outer face of the SNALP bilayer membrane, whereas standard techniques provide SNALPs with CPLs on both the inner and outer faces. The method is particularly applicable to vesicles prepared from phospholipids (which may contain cholesterol), and also to vesicles containing PEG-lipids (such as PEG-DAAs and PEG-DAGs). In, for example, U.S. patent nos. 5,705,385; 6,586,410, respectively; 5,981,501, respectively; 6,534,484, respectively; and 6,852,334; U.S. patent publication numbers 20020072121; and PCT publication No. WO 00/62813 teaches a method of preparing SNALP-CPL.

VI. kit

The invention also provides nucleic acid-lipid particles in the form of a kit. The kit may include a container partitioned to accommodate various components of the nucleic acid-lipid particle (e.g., the nucleic acid and individual lipid components of the particle). In some embodiments, the kit further comprises a destabilizing agent (e.g., calcium ions) for the endosomal membrane. The kit typically comprises the nucleic acid-lipid particle compositions of the invention, preferably in dehydrated form, together with instructions for their rehydration and administration. In certain instances, the particles and/or compositions comprising the particles can have a targeting moiety attached to the surface of the particle. Methods of attaching targeting moieties (e.g., antibodies, proteins) to lipids, such as those used in the present particles, are known to those of skill in the art.

Administration of nucleic acid-lipid particles

Once formed, the serum-stable nucleic acid-lipid particles of the invention are effectively used to introduce nucleic acids (e.g., siRNA) into cells. Accordingly, the invention also provides methods of introducing nucleic acids (e.g., siRNA) into cells. The method is performed in vitro or in vivo by first forming a particle as described above, followed by contacting the particle with the cell for a time sufficient for delivery of the nucleic acid to the cell to occur.

The nucleic acid-lipid particles of the invention can be adsorbed by virtually any cell type with which they are mixed or contacted. Once adsorbed, the particles can be endocytosed by parts of the cell, exchange lipids with the cell membrane, or fuse with the cell. Transfer or binding of the nucleic acid portion of the particle may occur by any of these pathways. Specifically, when fusion occurs, the membrane of the particle is integrated into the cell membrane, and the contents of the particle are in contact with the fluid within the cell.

The nucleic acid-lipid particles of the invention can be administered alone or in admixture with a pharmaceutically acceptable carrier (e.g., saline or phosphate buffer) selected with regard to the route of administration and standard pharmaceutical practice. Typically, normal buffered saline (e.g., 135-. Other suitable carriers include, for example, water, buffered water, 0.4% saline, 0.3% glycerol, and the like, including glycoproteins for improved stability, such as albumin, lipoprotein, globulin, and the like. Other suitable carriers are described, for example, in REMINGTON' S SPHARMACEMENT SCIENCES (Ramington pharmacopoeia science), Mack publishing company, Philadelphia, PA, 17 th edition (1985). As used herein, the term "carrier" includes any and all solvents, dispersion media, excipients, coatings, diluents, antibacterial and antifungal agents, isotonic and absorption delaying agents, buffers, carrier solutions, suspensions, colloids, and the like. The phrase "pharmaceutically acceptable" refers to molecular entities and compositions that do not produce an allergic or similar untoward reaction when administered to a human.

The pharmaceutically acceptable carrier is typically added after the particles are formed. Thus, after the particles are formed, the particles may be diluted into a pharmaceutically acceptable carrier, such as normal buffered saline.

The concentration of the particles in the pharmaceutical formulation can vary widely, i.e., from less than about 0.05% by weight, typically at or at least about 2-5% by weight, to as much as about 10-90% by weight, and will be selected primarily by fluid volume, viscosity, etc., in accordance with the particular mode of administration selected. For example, the concentration may be increased to reduce the fluid load associated with the treatment. This may be particularly desirable in patients with atherosclerosis-related congestive heart failure or severe hypertension. Alternatively, particles composed of stimulating lipids may be diluted to low concentrations to reduce inflammation at the site of administration.

The pharmaceutical compositions of the present invention may be sterilized by conventional, well-known sterilization techniques. The aqueous solution may be packaged for use, or filtered under sterile conditions and lyophilized, which is combined with a sterile aqueous solution prior to administration. The compositions may contain pharmaceutically acceptable auxiliary substances as required to adapt to physiological conditions, such as pH adjusting and buffering agents, tonicity adjusting agents and the like, for example, sodium acetate, sodium lactate, sodium chloride, potassium chloride, and calcium chloride. Additionally, the particle suspension may include a lipid-protecting agent that protects lipids from free radical and lipid-peroxide damage during storage. Lipophilic radical quenchers, such as alpha tocopherol, and water soluble iron ion specific chelators, such as ferrioxamine, are suitable.

A. In vivo administration

The use of nucleic acid-lipid particles, such as described in PCT publication No. WO 96/40964, and U.S. patent No. 5,705,385; 5,976,567, respectively; 5,981,501, respectively; and 6,410,328, has achieved systemic delivery of therapeutic in vivo through body systems such as the circulation, i.e., delivery of therapeutic nucleic acids to remote target cells. This latter approach provides well-encapsulated nucleic acid-lipid particles that protect nucleic acids from nuclease degradation in serum, are non-immunogenic, are small in size, and are suitable for repeated administration.

For in vivo administration, administration can be by any route known in the art, e.g., by injection, oral administration, inhalation (e.g., intranasally or intratracheally), transdermal administration, or rectal administration. Administration may be effected by a single or separate administration. The pharmaceutical composition may be administered parenterally, i.e., intra-articularly, intravenously, intraperitoneally, subcutaneously, or intramuscularly. In some embodiments, the pharmaceutical composition is administered intravenously or intraperitoneally by bolus (bolus) injection (see, e.g., U.S. patent No. 5,286,634). Intracellular nucleic acid delivery has also been described in Straubringer et al, Methods Enzymol (Methods in enzymology), 101: 512 (1983); mannino et al, Biotechniques (Biotechniques), 6: 682 (1988); nicolau et al, crit.rev.ther.drug carrier system (important review of therapeutic drug carrier systems), 6: 239 (1989); and Behr, acc. chem, Res. (description of chemical research), 26: 274 (1993). Other methods of administering lipid-based therapeutics are described, for example, in U.S. patent nos. 3,993,754; 4,145,410, respectively; 4,235,871; 4,224,179, respectively; 4,522,803; and 4,588,578. The lipid-nucleic acid particles can be administered by direct injection at the site of disease or by injection at a site remote from the site of disease (see, e.g., Culver, HUMAN GENE THERAPY (HUMAN gene therapy), maryan Liebert, inc., press, new york, pages 70-71 (1994)).

The compositions of the invention, alone or in combination with other suitable compositions, may be formulated into aerosol formulations (i.e., they may be "nebulized") for administration by inhalation (e.g., intranasally or intratracheally) (see, e.g., Brigham et al, am.j.sci. (journal of U.S. science), 298: 278 (1989)). The aerosol formulation may be placed in a pressurized, available propellant such as dichlorodifluoromethane, propane, nitrogen, and the like.

In certain embodiments, the pharmaceutical composition may be delivered by intranasal spray, inhalation, and/or other aerosol delivery vehicles. Methods for delivering nucleic acid compositions directly to the lungs by nasal aerosol spray have been described, for example, in U.S. Pat. nos. 5,756,353 and 5,804,212. Likewise, the use of intranasal microparticle resins and lysophosphatidyl-glycerol compounds (U.S. patent 5,725,871) for drug delivery is also well known in the pharmaceutical art. Similarly, transmucosal drug delivery in the form of polytetrafluoroethylene (polytetrafluoroethylene) support matrices is described in U.S. Pat. No. 5,780,045.

Formulations suitable for parenteral administration, such as, for example, by the intra-articular (at the joint), intravenous, intramuscular, intradermal, intraperitoneal, and subcutaneous routes, include aqueous and non-aqueous, isotonic sterile injection solutions, which may contain antioxidants, buffers, bacteriostats, and solutes that render the formulation isotonic with the blood of the intended recipient, and aqueous and non-aqueous sterile suspensions, which may include suspending agents, solubilizers, thickening agents, stabilizers, and preservatives. In the practice of the present invention, the composition is preferably administered, for example, by intravenous infusion, orally, topically, intraperitoneally, intravesically, or intrathecally.

Typically, when administered intravenously, the nucleic acid-lipid formulation is formulated with a suitable pharmaceutical carrier. A number of pharmaceutically acceptable carriers can be used in the compositions and methods of the invention. Suitable formulations for use in the present invention are found, for example, in REMINGTON' S PHARMACEUTICAL SCIENCES (Remington pharmacopoeia science), Mack publishing company, Philadelphia, Pa., 17 th edition (1985). Many aqueous carriers can be used, for example, water, buffered water, 0.4% saline, 0.3% glycerol, and the like, and can include glycoproteins that improve stability, such as albumin, lipoproteins, globulins, and the like. Normally buffered saline (135-. These compositions may be sterilized by conventional liposome sterilization techniques, such as filtration. The compositions may contain pharmaceutically acceptable auxiliary substances as required to adapt to the physiological conditions, such as pH adjusting and buffering agents, tonicity adjusting agents and the like, wetting agents and the like, for example, sodium acetate, sodium lactate, sodium chloride, potassium chloride, calcium chloride, sorbitan monolaurate, triethanolamine oleate, and the like. These compositions may be sterilized using the techniques mentioned above, or alternatively, they may be produced under sterile conditions. The resulting aqueous solution may be packaged for use, or filtered under sterile conditions and lyophilized, which is combined with a sterile aqueous solution prior to administration.

In certain applications, the nucleic acid-lipid particles disclosed herein can be delivered by oral administration to an individual. The granules can be combined with excipients and used in the form of ingestible tablets, buccal tablets, troches, capsules, pills, lozenges, elixirs, mouthwash, suspensions, oral sprays, syrups, wafers, and the like (see, e.g., U.S. Pat. nos. 5,641,515, 5,580,579, and 5,792,451). These oral dosage forms may also comprise the following: a binder, gelatin; excipients, lubricants, and/or flavoring agents. When the unit dosage form is a capsule, it may contain, in addition to the materials described above, a liquid carrier. Various other materials may be present, as coatings, or to otherwise modify the physical form of the dosage unit. Of course, any material used in preparing any unit dosage form should be pharmaceutically pure and substantially non-toxic in the amounts used.

Typically, these oral formulations may comprise at least about 0.1% of the nucleic acid-lipid particle, or more, although the percentage of the particle may of course vary, and may readily be between about 1% or 2% and about 60% or 70% or more by weight or volume of the total formulation. Naturally, the amount of particles that can be prepared in each therapeutic composition is such that, in any given unit dose of the compound, an appropriate dosage should be obtained. One skilled in the art of preparing such pharmaceutical formulations should consider such factors as solubility, bioavailability, biological half-life, route of administration, shelf-life of the product, and other pharmaceutical considerations, and thus, many dosages and treatment regimens may be desirable.

Formulations suitable for oral administration may consist of: (a) a lipid solution, such as an effective amount of packaged nucleic acids (e.g., siRNA), suspended in a diluent such as water, saline, or PEG 400; (b) capsules, sachets, or tablets, each containing a predetermined amount of nucleic acid (e.g., siRNA), as a liquid, solid, granules, or gelatin; (c) suspensions in suitable liquids; and (d) a suitable emulsion. The tablet form may include one or more of the following: lactose, sucrose, mannitol, sorbitol, calcium phosphate, corn starch, potato starch, microcrystalline cellulose, gelatin, colloidal silicon dioxide, talc, magnesium stearate, stearic acid, and other excipients, colorants, fillers, binders, diluents, buffers, moisturizers, preservatives, flavors, dyes, disintegrants, and pharmaceutically compatible carriers. Lozenge forms can include nucleic acids (e.g., siRNA) in a flavoring agent such as sucrose, as well as pastilles comprising nucleic acids (e.g., siRNA) in an inert matrix such as gelatin and glycerin, or emulsions, gels, and the like of sucrose and gum arabic containing carriers known in the art in addition to the nucleic acids (e.g., siRNA).

In another example of their use, nucleic acid-lipid particles can be incorporated into a wide range of topical dosage forms. For example, suspensions containing the nucleic acid-lipid particles can be formulated and administered as gels, oils, emulsions, topical creams, pastes, ointments, lotions, foams, foaming agents (mousses), and the like.

When preparing a pharmaceutical formulation of the nucleic acid-lipid particles of the invention, it is preferred to use a plurality of already purified particles to reduce or eliminate empty particles or particles with nucleic acid associated with an outer surface.

The methods of the invention may be practiced in a variety of hosts. Preferred hosts include mammalian species such as birds (e.g., ducks), primates (e.g., humans and chimpanzees, and other non-human primates), canines, felines, equines, bovines, ovines, caprines, rodents (e.g., rats and mice), lagomorphs, and swine.

The amount of particles administered will depend on the ratio of nucleic acid to lipid, the particular nucleic acid used, the disease state diagnosed, the age, weight, and condition of the patient, and the judgment of the clinician, but generally should be in the range of about 0.01 to about 50mg/kg body weight, preferably about 0.1 to about 5mg/kg body weight, or about 10 per administration (e.g., injection) ⁸-10¹⁰And (4) granules.

B. In vitro administration

For in vitro applications, nucleic acids (e.g., siRNA) can be delivered to any cell cultured in culture, whether of plant or animal origin, vertebrate or invertebrate, and of any tissue or type. In a preferred embodiment, the cell will be an animal cell, more preferably a mammalian cell, and most preferably a human cell.

When performed in vitro, the contact between the cells and the nucleic acid-lipid particles occurs in a biocompatible culture medium. The concentration of particles varies over a wide range depending on the particular application, but is typically between about 1. mu. mol and about 10 mmol. The treatment of the cells with the nucleic acid-lipid particles is generally carried out at physiological temperature (about 37 ℃) for a time period of about 1 to about 48 hours, preferably from about 2 to 4 hours.

In a preferred group of embodiments, the nucleic acid-lipid particle suspension is added to seeded cells at 60-80% confluence, the cells having about 10³To about 10⁵Individual cells/ml, more preferably about 2X 10⁴Density of individual cells/ml. The concentration of the suspension added to the cells is preferably about 0.01-0.2. mu.g/ml, more preferably about 0.1. mu.g/ml.

The delivery efficiency of SNALP or other lipid-based carrier systems can be optimized using an Endosomal Release Parameter (ERP) assay. The ERP assay is described in detail in U.S. patent publication No. 20030077829. More specifically, the objective of ERP assays is to distinguish the role of various cationic lipids and helper lipid components of SNALPs based on their relative effects on binding/uptake or fusion/destabilization to the endosomal membrane. This assay allows one to quantitatively determine how each component of SNALP or other lipid-based carrier system affects delivery efficiency, thereby optimizing SNALPs or other lipid-based carrier systems. Typically, ERP assays measure the expression of reporter proteins (e.g., luciferase, β -galactosidase, Green Fluorescent Protein (GFP), etc.), and in some cases, SNALP formulations optimized for the expression plasmid will also be suitable for encapsulating interfering RNAs. In other cases, ERP assays may also be useful for measuring down-regulation of transcription or translation of a target sequence in the presence or absence of interfering RNA (e.g., siRNA). By comparing the ERPs of each of the various SNALPs or other lipid-based formulations, a system that is optimized, such as SNALP or other lipid-based formulation that has the greatest uptake in the cell, can be readily determined.

C. Cells for delivery of interfering RNA

The compositions and methods of the invention are useful for treating a wide variety of cell types in vivo and in vitro. Suitable cells include, for example, hematopoietic precursor (stem) cells, fibroblasts, keratinocytes, hepatocytes, endothelial cells, skeletal and smooth muscle cells, osteoblasts, neurons, quiescent lymphocytes, terminally differentiated cells, slow or non-circulating primary cells, parenchymal cells, lymphoid cells, epithelial cells, bone cells, and the like.

In vivo delivery of nucleic acid-lipid particles encapsulating interfering RNAs (e.g., sirnas) is suitable for targeting cells of any cell type. The methods and compositions can be used with cells of a wide variety of vertebrates, including mammals, such as, for example, canines, felines, equines, bovines, ovines, caprines, rodents (e.g., mice, rats, and guinea pigs), lagomorphs, porcines, and primates (e.g., monkeys, chimpanzees, and humans).

The extent of tissue culture of cells that may be desired is well known in the art. For example, Freekney, Culture of Animal Cells, a Manual of Basic technology (Animal Cell Culture, Basic technical Manual), 3 rd edition, Wiley-Liss, New York (1994), Kuchler et al, Biochemical Methods in Cell Culture and Virology (Biochemical Methods of Cell Culture and Virology), Dowden, Hutchinson and Ross, Inc. (1977), and the references cited therein, provide general guidance for Cell Culture. The cultured cell system will typically be in the form of a monolayer of cells, although suspensions of cells are also used.

Detection of snalps

In some embodiments, the nucleic acid-lipid particle is detectable in the subject after about 8, 12, 24, 48, 60, 72, or 96 hours, or after 6, 8, 10, 12, 14, 16, 18, 19, 22, 24, 25, or 28 days after administration of the particle. The presence of the particles can be detected in a cell, tissue, or other biological sample from the subject. The particles can be detected, for example, by directly detecting the particles, by detecting interfering RNA (e.g., siRNA) sequences, by detecting a target sequence of interest (i.e., by detecting expression or reduced expression of the target sequence), or a combination thereof.

1. Detection of particles

The nucleic acid-lipid particles can be detected using any method known in the art. For example, the label is coupled directly or indirectly to a component of the SNALP or other carrier system using methods well known in the art. A wide variety of labels can be used, with the label being selected depending on the sensitivity required, ease of conjugation to the SNALP component, stability requirements, and preparation of available tools and processes. Suitable labels include, but are not limited to, spectroscopic labels, such as fluorescent dyes (e.g., fluorescein and derivatives, such as Fluorescein Isothiocyanate (FITC) and Oregon Green) ^TM(ii) a Rhodamine and derivatives, such as Texas Red, Tetrarhodamine isothiocyanate (TRITC), and the like, digoxigenin, Biotin, phycoerythrin, AMCA, CyDyes^TMEtc.); radiolabels, e.g.³H，¹²⁵I，³⁵S，¹⁴C，³²P，³³P, etc.; enzymes such as horseradish peroxidase, alkaline phosphatase, etc.; spectrally colorimetric labels such as colloidal gold or coloured glass or plastic beads such as polystyrene, polypropylene, latex etc. The label may be detected using any means known in the art.

2. Detection of nucleic acids

Nucleic acids (e.g., sirnas) are detected and quantified by any of a number of means well known to those skilled in the art. Detection of nucleic acids is performed by methods well known in the art, such as southern blot analysis, northern blot analysis, gel electrophoresis, PCR, radiolabeling, scintillation counting, and affinity chromatography. Other analytical biochemical methods such as spectrophotometry, radiography, electrophoresis, capillary electrophoresis, High Performance Liquid Chromatography (HPLC), Thin Layer Chromatography (TLC), ultra diffusion (Hyperdiffusion) chromatography may also be employed.

The choice of the nucleic acid hybridization format is not critical. Various forms of nucleic acid hybridization are known to those skilled in the art. For example, common formats include sandwich (sandwich) assays and competitive or displacement assays. Hybridization techniques are generally described, for example, in "Nucleic Acid Hybridization, A practical approach," Hames, and Higgins, eds., IRL Press (1985).

The sensitivity of hybridization assays can be increased by using nucleic acid amplification systems that multiply the target nucleic acid being detected. In vitro amplification techniques suitable for amplifying sequences for use as molecular probes or for generating nucleic acid fragments for subsequent subcloning are known. By these in vitro amplification methods, including Polymerase Chain Reaction (PCR), Ligase Chain Reaction (LCR), Q β -replicase amplification and other RNA polymerase mediated techniques (e.g., NASBA)^TM) Examples of techniques sufficient to instruct the skilled person are found in Sambrook et al, Molecular Cloning: a Laboratory Manual, Cold Spring Harbor Laboratory Press, 2000, and Ausubel et al, SHORT PROTOCOLS IN MOLECULAR BIOLOGY (Rapid methods IN MOLECULAR BIOLOGY), eds., Current PROTOCOLS, Greene publishing Associates, Inc. (Greenwich publishing Co., Ltd.) with John Wiley and Sons Inc. (2002); and U.S. Pat. nos. 4,683,202; PCR Protocols, A Guide to Methods and applications (Guide for PCR Methods, Methods and applications) (Innis et al, ed.), Academic Press Inc. (Academic Press Inc.. San Diego, CA (1990); arnheim and Levinson (1990, 10/1/C), C &EN 36; the Journal Of NIH Research (Journal Of NIH Research), 3: 81 (1991); kwoh et al, proc.natl.acad.sci.usa (proceedings of the american academy of sciences), 86: 1173 (1989); guatelli et al, proc.natl.acad.sci.usa (proceedings of the american academy of sciences), 87: 1874 (1990); lomell et al, j.clin.chem. (journal of clinical chemistry), 35: 1826 (1989); landegren et al, Science, 241: 1077 (1988); van Brunt, Biotechnology (Biotechnology), 8: 291 (1990); wu and Wallace, Gene, 4: 560 (1989); barringer et al, Gene, 89: 117(1990), and soknanan and Malek, Biotechnology (Biotechnology), 13: 563(1995). Cloning of in vitro amplified nucleic acidsImproved methods are described in U.S. Pat. No. 5,426,039. Other methods described in the art are Nucleic Acid Sequence Based Amplification (NASBA)^TMCangene, Mississauga, Ontario) and Q β -replicase systems. These systems can be used to directly identify mutants in which the PCR or LCR primers are designed to be extended or ligated only in the presence of the selection sequence. Alternatively, the selected sequence can be amplified, typically using, for example, non-specific PCR primers, and the amplified target region is later probed for specific sequences indicative of mutations.

According to Beaucage et al, Tetrahedron letters, 22: 18591862 (1981) using an automated synthesizer, e.g., as described in needleham Van Devanter et al, Nucleic Acids Res. (Nucleic Acids research), 12: 6159(1984), typically chemically synthesized, for use as a probe, e.g., in an in vitro amplification method, as a gene probe, or as an inhibitor component. Purification of polynucleotides is typically carried out by native acrylamide gel electrophoresis or by anion exchange HPLC, when necessary, as described in Pearson et al, j.chrom. (journal of chromatography), 255: 137149 (1983). The method was performed using Maxam and Gilbert (1980) in Grossman and Moldave (eds.), academic press, new york, Methods in Enzymology, 65: 499 the sequence of the synthesized polynucleotide can be confirmed by the chemical degradation method.

An alternative way to determine the level of transcription is in situ hybridization. In situ hybridization assays are well known and are generally described in anger et al, Methods Enzymol, 152: 649 (1987). In situ hybridization assays, cells are immobilized on a solid support, typically a slide. If the DNA is to be probed, the cells are denatured with heat or alkali. Next, the cells are contacted with a hybridization solution at a suitable temperature to allow annealing of the labeled specific probes. The probe is preferably labeled with a radioisotope or fluorescent reporter.

VIII example

The present invention will be described in more detail by way of the following examples. The following examples are provided for illustrative purposes and are not intended to limit the invention in any way. Those skilled in the art should readily recognize a variety of non-critical parameters that may be varied or modified to produce substantially the same result.

Example 1 design of non-inflammatory synthetic siRNA that regulates potent gene silencing in vivo

This example illustrates that a minimal 2 '-O-methyl (2' OMe) modification at a selected position in one strand of an siRNA duplex is sufficient to reduce or completely eliminate the immunostimulatory activity of the siRNA, regardless of its sequence. In fact, by limiting the 2' OMe modification to the non-target sense strand of the siRNA duplex, the immunostimulatory activity of the siRNA can be abolished, while retaining sufficient RNAi activity.

Results

2' 0Me modifications within the ssRNA abolished the immunostimulatory properties. To examine the extent and type of chemical modification required to inhibit immune cell activation by RNA, 2' OMe nucleotides were selectively introduced into the GU-rich immunostimulatory motif of single-stranded RNA polynucleotides (ssRNA) derived from β -galactosidase (. beta. -gal) siRNA (Judge et al, Nat. Biotechnol. (Nature Biotechnology), 23: 457-. The polynucleotide sequences used in this study are provided in table 1. When human Peripheral Blood Mononuclear Cell (PBMC) cultures were treated with ssRNA-encapsulating lipids, 2 'OMe modification of 5 nucleotides including an immunostimulatory 5' -ugugugu-3 'motif (2' OMe GU) in β -gal sense ssRNA completely abolished induction of interferon- α (IFN- α) (fig. 1A). Inhibition of the interferon response was also obtained by selectively modifying 2 guanosine (2 'OMe 2xG) or 3 uridine (2' OMe 3xU) nucleotides within the motif. Since the 2 guanosine residues at the 3 'end of the ugugugu motif towards the end of β -galssRNA (2' OMe 2xG 3 ') also resulted in a complete abolition of the interferon response in PBMC cultures (fig. 1A), the inhibition of 2' -O-methyl action did not appear to require direct modification of nucleotides within the immunostimulatory GU-rich motif. As previously described, in these assays, the unmodified complementary Antisense (AS) ssRNA sequence is inherently non-immunostimulatory (Judge et al, supra). Similar results were obtained when β -gal ssRNA was delivered to PBMCs using the cationic polymer Polyethylenimine (PEI) (fig. 2A).

Similar methods were used to modify the strands of the 21-and 23-base components of siRNA duplexes targeting human and mouse ApoB (Soutschek et al, Nature, 432: 173-178 (2004)). Unmodified ApoB (AS) ssRNA stimulates a strong IFN-. alpha.response in PBMC cultures even at low concentrations, as predicted by their GU-rich nucleotide sequences (Heil et al, Science 303: 1526-. This reaction was completely inhibited by 2 'OMe modifications to 5 nucleotides (2' OMe) or 6 guanosine (2 'OMe G) or 7 uridine (2' OMe U) residues comprising the 5 '-gugg-3' motif in apob (as) ssRNA (fig. 1B). Unmodified, complementary ApoB sense polynucleotide (ApoB (s)) encapsulated in lipid particles did not induce IFN- α in PBMC (fig. 1B), despite the high dose of this polynucleotide delivered as a result of the discovery that PEI polymer-nucleic acid complexes activate cytokine responses. This weak response to PEI-complexed apob(s) ssRNA was also inhibited by the 2' OMe-uridine modification. These findings indicate that selective binding of 2' OMe-modified nucleotides within ssRNA is sufficient to prevent stimulation of the interferon response from innate immune cells.

TABLE 1 RNA polynucleotides used in this study.

Unmodified (native) and 2' OMe-modified RNA polynucleotides correspond to the sense (S) and Antisense (AS) strands of β -gal, ApoB mismatch and vflp siRNA. The 2' OMe-modified nucleotides are shown in bold and underlined. Asterisks indicate 5' phosphate. "dT" is deoxythymidine.

Selective nucleotide modifications within siRNA eliminate immune stimulation. To test whether selective 2 'OMe modification within siRNA duplexes also inhibited immunostimulatory activity, β -gal and ApoB siRNA series were generated that included 2' OMe-modified sense or AS strands annealed to their complementary unmodified polynucleotides (see, table 1). The lipid encapsulating the double stranded β -gal siRNA comprising the 2' OMe-modified ugugugu, 2xG, or 3xU sense strand annealed to the unmodified (non-immunostimulatory) AS strand did not induce detectable interferon responses from human PBMCs (fig. 3A). Interestingly, selective 2 ' OMe modification of the complementary 5 ' -ACACA-3 ' motif in the AS strand, juxtaposed to the unmodified 5 ' -UGUGU-3 ' motif in the sense strand, also attenuated the level of IFN- α induction, despite the annealed duplex comprising the unmodified (immunostimulatory) sense strand (fig. 3A). Similar results were obtained when β -gal siRNA was delivered to PBMCs using PEI (FIG. 2B). Likewise, unmodified ApoB siRNA induced a strong IFN- α response in PBMC, and this response was completely abolished when 2' OMe GU, U, or G modified AS strands were incorporated into the ApoB duplex (fig. 3B). Notably, the modified ApoBsiRNA comprising the 2' OMe G or U modified sense strand annealed to the unmodified, immunostimulatory AS strand also did not confer immunostimulatory properties (fig. 3B). Since even high concentrations (675nM,. about.9. mu.g/ml) of modified siRNAs failed to induce IFN-. alpha.or inflammatory cytokines such as TNF in PBMC cultures (FIGS. 3B and 3C), it appeared to be absolute to abrogate cytokine induction by 2' OMe G or U modification of the sense strand of the modified ApoB siRNA.

However, since ApoB sirnas containing 2 'OMe-modified cytidine residues induced cytokine levels similar to those induced by natural duplexes (fig. 3B), no inhibition of immune stimulation of the siRNA by 2' -O-methylation was observed using all modification patterns. Binding of 2' OMe adenosine results in significant, rather than absolute, inhibition of cytokine response. Since the 2' OMe G, U, C, and a modified ApoB contain 2, 5, 6, and 8 modified nucleotides in the sense strand, respectively, these differences do not simply reflect the degree of chemical modification. This suggests that unmodified U and/or G residues may play an important role in immune recognition of duplex sirnas.

To confirm that this siRNA design approach will successfully inhibit the inflammatory response to siRNA in vivo, the immunostimulatory activity of 2' OMe-modified β -gal and ApoB sirnas was evaluated in mice. Intravenous administration of lipids encapsulating beta-gal (fig. 4A and 4B) or ApoB (fig. 4C and 4D) sirnas containing a 2' OMe-modified guanosine or uridine residue in the sense or antisense strand did not cause detectable increases in serum IFN- α or inflammatory cytokines such as TNF. This is significantly different from unmodified or cytosine-modified siRNAs, which induce substantial elevation in the levels of these cytokines. These significant effects of selective 2' OMe modification were confirmed by similar methods using modified ApoB mismatches (Sotschek et al, Nature (Nature), 432: 173-178(2004)) and vFLIP (Guaspari et al, J.Exp.Med. (J.Experimental medicine), 199: 993-1003(2004)) siRNA sequences (see, Table 1). For ApoB mismatched (fig. 4E) and vflp (fig. 4F) siRNA duplexes, modifying the GU-rich region or just the uridine residues in either RNA strand completely abolished cytokine induction of the siRNA duplexes. Inhibition of cytokine response against modified ApoB mismatched sirnas was also demonstrated in human PBMC cultures (fig. 3B and 3C). Incorporation of 2' OMe cytosine residues into the vflp siRNA did not substantially reduce IFN- α response as with ApoB (fig. 4F). Similar results have been consistently obtained for each siRNA sequence tested, where binding of 2' OMe-uridine or guanosine residues produced a non-inflammatory siRNA duplex. For example, figures 5-7 show that for the 5 additional siRNA sequences provided in table 2, the introduction of 2' OMe-uridine or guanosine residues results in a non-inflammatory siRNA duplex. Taken together, these findings support the inference that the underlying mechanism for immunologically recognizing short RNA duplexes is conserved between mouse and human (Judge et al, supra; Hornung et al, nat. Med. (Nature medicine), 11: 263-plus 270 (2005)). These results indicate that in either species, this mechanism can be severely interrupted by binding as few as 2' OMe-modified nucleotides in either strand of the siRNA duplex.

TABLE 2 other RNA polynucleotides used in this study.

Unmodified (native) and 2' OMe-modified RNA polynucleotides correspond to the sense (S) and Antisense (AS) strands of β -gal, luciferase (Luc), cyclophilin b (cyp b), influenza Nucleocapsid Protein (NP) and influenza Polymerase (PA) sirnas. The 2' OMe-modified nucleotides are shown in bold and underlined. Asterisks indicate 5' phosphate. "dT" is deoxythymidine.

Restriction modification of the sense strand of the siRNA retained RNAi activity. Both native and 2' OMe-modified ApoB siRNAs were evaluated for gene silencing activity in vitro. Unmodified ApoB encapsulated within liposomes caused potent, dose-dependent inhibition of ApoB protein in HepG2 cell culture supernatant (figure 8). Estimated IC₅₀The values (. about.1.5 nM) are consistent with those determined for this siRNA sequence using Oligofectamine transfection in a similar in vitro model (Soutschek et al, supra). The modified ApoB duplexes in which the 2' OMe modification was restricted in the non-target sense or passenger strand (passenger strand) exhibited ApoB silencing activity similar to that of the natural siRNA (figure 8). In contrast, modifications to the target Antisense (AS) or guide strand severely affect the RNAi activity of the duplex. Incorporation of 2 ' OMe uridine or guanosine residues in the AS strand abolished ApoB gene silencing, whereas duplexes containing 5 ' -gugg-3 ' modified AS strands exhibited substantially reduced activity (estimated IC) compared to natural or sense modified duplexes ₅₀15 nM). Unmodified or modified ApoB mismatch control siRNAs did not produce significant inhibition of ApoB protein expression (fig. 8). A similar strategy that restricted 2' OMe modification to the sense strand of β -gal 728 and luciferase siRNA also demonstrated successful generationNon-inflammatory sirnas that retained full RNAi activity were generated (fig. 6-7). Although the adverse effect of AS strand modification on gene silencing activity is consistent with previous work that demonstrated that 2 ' OMe modification of the AS strand of an siRNA duplex, particularly at the 5 ' end, reduced RNAi activity (Prakash et al, J.Med.Chem. (J.Med.Chem., 48: 4247-4 (2005)), siRNA sequences that could tolerate extended 2 ' OMe modification of the AS strand have been identified (Morrissey et al, Hepatology, 41: 3149-1356 (2005); Czauderna et al, Nucl.acids Res. (nucleic acid research), 31: 2705-2716 (2003)). These data indicate that selective 2' OMe modification limited to the sense strand of siRNA provides a powerful approach to overcome the problem of siRNA immune activation while reducing the chance of adversely affecting RNAi activity. These results indicate that this approach can be applied to many, if not all, siRNA sequences that have the inherent ability to stimulate the innate immune response, including a large number of synthetic sirnas of conventional design.

Potent RNAi activity in vivo without immunostimulation. 2' OMe-modified ApoBsiRNAs were evaluated for their ability to silence gene expression and immune stimulation in vivo. 2' OMe U (S) and GU (AS) modified ApoB were selected as non-inflammatory duplexes (see, FIGS. 3 and 4). This also provides the opportunity to evaluate the effect of chemical modifications that reduce RNAi activity of AS modified sirnas in vitro (see, figure 8). Natural or 2' OMe-modified ApoB and mismatch siRNAs were formulated in stable nucleic acid-lipid particles (SNALPs) that previously demonstrated delivery of siRNAs to the liver (Morrissey et al, nat. Biotechnol (Nature Biotechnology), 23: 1002-1007 (2005)). For use in systemic applications, nucleic acid-based drugs require stabilization or protection from nuclease degradation. Encapsulation inside the lipid bilayer protected unmodified and otherwise labile sirnas from serum nuclease degradation for more than 24 hours at 37 ℃ in vitro, indicating that encapsulation provided sufficient nuclease protection without chemical modification to the siRNA for extension. By comparison, naked siRNA was completely degraded within 4 hours under similar conditions (fig. 9).

The encapsulated ApoB siRNA was administered intravenously to BALB/c mice at 5 mg/kg/day for 3 days. This protocol represents a 10-fold reduction in ApoB siRNA dose (Soutschek et al, supra) originally reported to be effective in experiments using cholesterol-conjugated, chemically modified ApoB sirnas. Animals receiving native, immunostimulatory ApoB or mismatched siRNA showed significant toxicity symptoms as indicated by a 10.5% and 9% reduction in initial body weight by day 3 (fig. 10A) and a slight worsening of the overall physical condition throughout the treatment period, respectively. In contrast, treatment with 2' OMe-modified siRNA tolerated well with minimal (less than 1%) or no weight loss (fig. 10A). The abrogation of the innate cytokine response in these efficacy studies was confirmed by serum IFN- α analysis throughout life (fig. 10B), and thus the toxicity associated with administration of unmodified siRNA was attributed to systemic cytokine responses. It should be noted that cytokine levels and body weight loss induced by unmodified mismatched sirnas were lower than the corresponding active ApoB duplexes. In this case, the mismatch control was generated by 4G/C substitutions in the ApoB sequence (Sotschek et al, supra), which provides further evidence for the sequence-dependent effect of the RNA duplex on immune stimulation.

As a direct measure of RNAi-mediated knockdown, ApoB mRNA in liver was determined two days after the last siRNA treatment (fig. 10C). In the native and 2' OMe U (S) modified ApoB-treated groups, ApoB mRNA levels were significantly reduced compared to PBS-treated animals (18 ± 2% and 18 ± 5% PBS control, respectively). By comparison, mice treated with 2' OMe GU (AS) -modified ApoB siRNA exhibited less pronounced ApoB mRNA silencing (44 ± 4% control), which correlates with reduced in vitro RNAi activity of such modified siRNA (see, fig. 8). ApoB mRNA levels in the modified mismatch group were equivalent to those in the PBS control (fig. 10C), while the native mismatch siRNA caused a moderate reduction in ApoB mRNA levels (79 ± 12% PBS control). In these three independent experiments, the moderate reduction in hepatic ApoB mRNA observed with the natural mismatched siRNA was evident and correlated with interferon release and toxicity symptoms associated with systemic administration and delivery of the unmodified siRNA.

Silencing of ApoB mRNA in the liver results in a proportional, sequence-specific reduction in serum ApoB protein. Mice treated with native, 2' OMe U (S), or gu (as) modified ApoB siRNA had serum ApoB protein levels of 26%, 28%, and 47% of those of PBS-treated animals, respectively (fig. 10D). Functional silencing of ApoB expression is also reflected in a significant reduction in serum cholesterol, which correlates with the relative potency of the ApoB duplex to mRNA and protein knockdown. Mice treated with native, 2' OMe U (S), or gu (as) modified ApoB siRNA exhibited serum cholesterol levels that were 48%, 51%, and 69% of those of the PBS control group (fig. 10E). The mismatched sirnas did not have any effect on serum cholesterol (fig. 10E). In a separate experiment, non-inflammatory 2' OMe G (S) -modified ApoB sirnas regulated similar reductions in ApoB mRNA, protein, and serum cholesterol in the absence of IFN induction.

The results from these studies indicate that lipid encapsulation of siRNA provides sufficient serum stability for systemic application and negates the need for extended chemical modification of RNA. In the case of the liver, coupled with efficient delivery of siRNA payload to the target organ, this promoted silencing of endogenous genes, exemplified in these studies by ApoB, a protein that represents a potential therapeutic target for hypercholesterolemia. Importantly, the 2' OMe-modified siRNA, designed to be non-inflammatory, exhibited efficacy in vivo equivalent to unmodified siRNA but without the immunotoxicity and other off-target effects associated with systemic administration of unmodified siRNA. The methods described herein can be generally applicable to a wide range of gene targets, and are suitable for use in a variety of therapeutic approaches.

Discussion of the related Art

Based on the finding that immune activation of siRNA is sequence-dependent, it has previously been shown that it is possible to design active siRNAs with negligible immunostimulatory activity by selecting sequences lacking a GU-rich motif (Judge et al, nat. Biotechnol (Nature Biotechnology), 23: 457-. However, this measure significantly limits the number of novel siRNA sequences that can be designed for a given target. Furthermore, due to the relatively unclear nature of putative RNA immunostimulatory motifs, some degree of screening is currently required. This study highlights a novel and powerful approach to abrogate synthetic siRNA-mediated immune stimulation by selective incorporation of 2' OMe-modified nucleotides into siRNA duplexes. Significantly, the incorporation of as few as 2' OMe guanosine or uridine residues in the highly immunostimulatory siRNA molecules completely abolished siRNA-mediated induction of interferon and inflammatory cytokines in human PBMCs and in mice. This degree of chemical modification represents-5% of the natural 2' -OH positions in the siRNA duplex. Since complete elimination of the immune response requires selective modification of only one of the RNA strands, the 2' OMe modification can be restricted to the sense strand of the duplex, thus minimizing the potential for reducing siRNA efficacy. These findings have provided a simple rationale for the synthesis of non-immunostimulatory sirnas based on natural sequences with demonstrated RNAi activity. By combining selectively modified sirnas with effective systemic delivery vehicles, such as nucleic acid-lipid particles, robust silencing of endogenous gene targets can be achieved in vivo at therapeutically feasible doses without the adverse side effects associated with systemic activation of the innate immune response.

Since the 2' -OH in the ribose backbone is a distinguishing feature of RNA, extended chemical substitutions at this position are expected to disrupt recognition of the modified nucleic acid by the RNA binding receptor pathway. However, the present study unexpectedly shows that 2' OMe-modified sirnas exhibit no immunostimulatory activity, despite retaining up to 95% of their native ribonucleotides, including those comprising defined immunostimulatory regions of the RNA. The 2 'OMe is believed to be a relatively large chemical group at the 2' position, which is located in the minor groove of the RNA duplex without significantly distorting its A-type helical structure (Chiu et al, RNA, 9: 1034. Res. (2003); Cummins et al, Nucl. acids Res. (nucleic acid research), 23: 2019. sup. 2024 (1995)). This may be sufficient to disrupt the interaction between the double stranded RNA duplex and its putative immune receptor or helper molecule. Trans-inhibition of 2 '-O-methylation, whereby 2' OMe-modified ssRNA anneals to unmodified immunostimulatory ssRNA to produce an immunostimulatory duplex, is consistent with such a hypothesis including recognition of siRNA as a double-stranded molecule.

Many other stabilizing chemistries are routinely used in synthetic siRNA design in an effort to confer nuclease resistance that may also affect immune recognition and RNAi. Locked Nucleic Acids (LNA) containing a 2 '-O, 4' -C methylene bridge in the sugar ring have been shown to partially reduce the immunostimulatory activity of siRNA (Hornung et al, nat. Med. (Nature medicine), 11: 263-270 (2005)). It has been found that siRNA containing a reverse deoxy non-basic end-cap retains immunostimulatory activity (Morrissey et al, nat. Biotechnol (Nature Biotechnology), 23: 1002-1007 (2005)). No evidence of trans-inhibition was observed with LNA-modified duplexes. These observations indicate that the immunostimulatory properties of siRNA are particularly sensitive to inhibition of 2' OMe modification relative to other well-described stabilization chemistries.

This study shows that unmodified and 2' OMe-modified synthetic sirnas can regulate robust silencing of endogenous gene target ApoB when encapsulated in lipid particles and administered systemically. Intravenous administration of encapsulated unmodified or modified ApoB siRNA results in a significant reduction of ApoB mRNA levels in the liver, with a concomitant reduction of ApoB protein in the blood. Importantly, assuming interest in ApoB as a therapeutic target for hypercholesterolemia, ApoB silencing results in a significant reduction in serum cholesterol. Lipid encapsulation confers excellent resistance to serum nuclease degradation, allowing for the use of minimally modified siRNA duplexes in vivo. By preventing the induction of interferons and inflammatory cytokines, the potential for non-specific effects on gene expression is limited, while the tolerance of siRNA formulations is increased. In particular, intravenous administration of encapsulated 2' OMe-modified siRNA is effective and well tolerated. These findings facilitate the use of synthetic sirnas in a wide range of in vivo and therapeutic applications.

Method of producing a composite material

siRNA: all sirnas used in these studies were chemically synthesized by Dharmacon (Lafayette, CO) and received as desalted, deprotected polynucleotides. The duplexes are annealed by standard methods. Equimolar concentrations of the complementary strand were heated to 90 ℃ for 2 minutes and then slowly cooled at 37 ℃ for 60 minutes. The formation of annealed duplexes was confirmed by non-denaturing PAGE analysis. All natural and 2' OMe-modified sequences used in this study are listed in tables 1 and 2.

Lipid encapsulation of RNA: by spontaneous vesicle formation, followed by stepwise ethanol dilutions, as in Jeffs et al, pharm. res. (pharmaceutical research), 22: 362-372(2005) description of pDNA, encapsulation of siRNA or ssRNA in liposomes. Liposomes consist of the following lipids: synthetic cholesterol (Sigma; Louis, MO), phospholipid DSPC (1, 2-distearoyl-sn-glycero-3-phosphocholine; Avanti Polar Lipids (Avanti Polar Lipids); Alabaster, AL), PEG-lipid PEG-cDMA (3-N- [ (methoxypoly (ethylene glycol) 2000) carbamoyl)]1, 2-dimyryltrypropylamine) and the cationic lipid DLinDMA (1, 2-dioleyloxy-3- (N, N-dimethyl) aminopropane) in a molar ratio of 48: 20: 2: 30. The lipids PEG-cDMA and DLinDMA (Heyes et al, J. control Release, 107: 276-287(2005)) were synthesized in Proloving vitamin therapy (Protiva Biotherapeutics). The obtained stable lipid particles were dialyzed in PBS and filter sterilized through a 0.2 μm filter before use. The particle size of each liposome formulation is in the range of 100-130nm and typically comprises 90-95% siRNA encapsulated within the liposome. Membrane-impermeable fluorescent dye, RiboGreen, was used before and after the detergent addition to disrupt the lipid bilayer (molecular probes; Eugene, OR) determine the concentration and percent encapsulation of formulated siRNA (Jeffs et al, supra).

Serum nuclease protection assay: unmodified naked or lipid-encapsulated siRNA (0.25mg/ml) were incubated in 50% mouse serum at 37 ℃. At the indicated times, the aspirated aliquots were added directly to the gel loading buffer containing 0.1% SDS and frozen in liquid nitrogen. After the final time point, siRNA samples were run on a non-denaturing 20% polyacrylamide TBE gel and visualized by ethidium bromide staining. To confirm the nuclease protection conferred by lipid encapsulation of siRNA, 0.1% triton-X100 was added immediately prior to incubation with serum to disrupt the integrity of the lipid bilayer.

Cell isolation and culture: human PBMC were isolated from whole blood of healthy donors by standard Ficoll-Hypaque density centrifugation techniques. For the immunostimulatory assay, 3 × 10⁵Freshly isolated PBMCs were plated in triplicate in 96-well plates and cultured in RPMI 1640 medium containing 10% FBS, 2mM glutamine, 100U/ml penicillin, and 100. mu.g/ml streptomycin. Liposomal encapsulated siRNA was added to cells at the final nucleic acid concentrations indicated, and culture supernatants were collected after 16-20 hours and assayed for IFN- α, IL-6, and TNF- α by sandwich ELISA.

In vitro RNA interference assay: HepG2 cells were seeded into 24-well plates at 20,000 cells/well. To determine the in vitro RNAi activity of 2' OMe-modified ApoB sirnas, HepG2 cultures were treated with encapsulated sirnas in triplicate at nucleic acid concentrations between 0.6nM and 45 nM. The medium was changed 24 hours after the addition of siRNA and then incubated for another 48 hours. Human ApoB protein levels in culture supernatants were determined by sandwich ELISA as described in Soutschek et al, Nature, 432: 173(2004), polyclonal goat anti-human ApoB capture antibody (Chemicon International) and horseradish peroxidase-conjugated goat anti-human ApoB-100 antibody (Academy Bio-medical) were used to detect bound ApoB, as described in detail. The ELISA plates were developed using TMB substrate, stopped with 2N sulfuric acid, and the absorbance read at 450nm-570 nm. A. the₄₅₀Values were normalized to a standard curve generated from untreated HepG2 conditioned media to determine the linear range of the ELISA. In siRNA-treated culture supernatantWas calculated as a percentage of PBS-treated control.

In vivo cytokine induction: animal studies were completed following approval by the local Prolo vitamin treatment company (Protiva Biotherapeutics) Animal Care and use committee, according to the Canadian Council on Animal Care Guideline. Prior to use, 6-8 week old CD1 ICR mice (Harlan; Indianapolis, IN) were subjected to 3 weeks of quarantine and acclimation. The encapsulated siRNA formulation in 0.2ml PBS was administered into the lateral tail vein by standard intravenous injection. Blood was collected by cardiac puncture 6 hours after administration and processed as plasma for cytokine analysis. In the RNAi efficacy experiment, plasma was collected from 50 μ l of test blood samples 6 hours after initial siRNA administration.

Cytokine ELISA: all cytokines were quantified using a sandwich ELISA kit according to the instructions of the supplier. These include mouse and human IFN-alpha (PBL biomedical; Piscataway, NJ), human IL-6 and TNF-alpha (eBioscience; San Diego, Calif.), and mouse IL-6, TNF-alpha, and IFN-gamma (BD bioscience; San Diego, Calif.).

In vivo RNA interference: groups of 5 Balb/C mice were treated once daily for 3 consecutive days with lipid-encapsulated siRNA (natural, 2 ' OMe U (S), 2 ' OMe, or GU (AS) ApoB and natural or 2 ' OMe U (S) mismatches) by standard intravenous injection in the lateral tail vein at 5 mg/kg. Body weights and general observations were recorded throughout the duration of the study. Mice were sacrificed 48 hours after the last siRNA treatment. Blood was collected by cardiac puncture for serum analysis of ApoB protein and cholesterol. The livers were weighed and collected into 6ml RNAlater (Sigma) for ApoB mRNA analysis by QuantiGene assay (Genospectra; Fremont, CA).

Serum cholesterol was measured using a commercially available cholesterol measurement kit according to the supplier's instructions (Thermo Electron Corp; Melbourne, Australia) Amount of the compound (A). Serum from individual animals was tested for ApoB-100 by sandwich ELISA (Zlot et al, J.Lipid Res. (J.Lipid. Res., 40: 76-84(1999)) using the monoclonal mouse ApoB-100 capture antibody LF 3. Bound ApoB-100 was detected using polyclonal rabbit anti-mouse ApoB (International Biodesign; Saco, Maine) and horseradish peroxidase-conjugated goat anti-rabbit Ig's (Jackson Immunoresearch; West Grove, Pa.). Standard curves generated using normal mouse serum, from A₄₅₀Values serum ApoB levels were determined to define the linear range of the ELISA and are expressed as a percentage of the PBS-treated control group.

The QuantiGene assay (genoscope) was used to quantify the reduction of mouse ApoB mRNA in liver tissue after siRNA treatment. Small homogeneous tissue samples were taken from livers collected 48 hours after the last injection and stored in RNAlater (sigma). Lysates were used directly for ApoB and GAPDH mRNA quantification and the ratio of ApoB and GAPDH mRNA was calculated and expressed as group mean relative to PBS control group. Specific probe sets for mRNA detection were designed by Genospectra to target the following regions: for ApoB mRNA, positions 5183-5811 of accession number XM _ 137955; for GAPDH mRNA, positions 9-319 of accession no NM — 008084.

Example 2 design of ApoB siRNA with Selective chemical modification

This example illustrates that minimal 2' OMe modification at selected positions in the sense and antisense strands of an ApoB siRNA duplex is sufficient to reduce the immunostimulatory properties of an ApoB siRNA while retaining RNAi activity. In particular, selective 2 'OMe-uridine and/or 2' OMe-guanosine modifications at less than about 30% of the nucleotide positions in both strands provide conjugated ApoB sirnas with desirable silencing and non-immunostimulatory properties.

Results

A female BALB/c mouse model was used to determine the efficacy and toxicity profile of SNALP formulations containing ApoB siRNA with selective chemical modification in the sense and antisense strands. The ApoB siRNA duplexes used in this study are provided in table 3.

TABLE 3 siRNA duplexes comprising sense and antisense ApoB RNA polynucleotides.

Column 1: numbers refer to the nucleotide position at the 5' base of the sense strand relative to the mouse ApoB mRNA sequence XM _ 137955. Column 2: the numbers refer to the distribution of 2' OMe chemical modifications in each chain. For example, "U2/4" refers to 2 uridine 2 'OMe modifications in the sense strand and 4 uridine 2' OMe modifications in the antisense strand. Column 3: the 2' OMe-modified nucleotides are shown in bold and underlined.

Column 4: the number and percentage of 2' OMe-modified nucleotides in the siRNA duplex are provided. Column 5: the number and percentage of modified nucleotides in the double-stranded (DS) region of the siRNA duplex is provided.

For the ApoB 10048siRNA family sequences, 2' OMe modifications at 16.7% or 23.8% of the nucleotide positions in the siRNA duplex resulted in similar silencing activity as the unmodified siRNA (fig. 11, rows 2-6). Similar results were obtained for ApoB 10048siRNA sequences with 2' OMe modifications at 19% of the nucleotide positions. However, 2' OMe modifications at 35.7% of the nucleotide positions in the siRNA duplex resulted in reduced activity (fig. 11, lines 7-8). For the ApoB 10886 siRNA sequence, 2' OMe modifications at 16.7% of the nucleotide positions in the siRNA duplex resulted in increased activity compared to the unmodified siRNA (fig. 11, lines 9-10). For the ApoB 10346 siRNA sequence, 2' OMe modifications at 16.7% of the nucleotide positions in the siRNA duplex resulted in reduced activity compared to unmodified siRNA (fig. 11, lines 11-12).

The use of 2' OMe instead of chemically modified siRNA improves the toxicity profile of siRNA treatment in vivo by eliminating cytokine response. As shown in figure 12, none of the modified apobsirnas tested in this group stimulated interferon- α release, whereas treatment with any of the three unmodified siRNA negatives resulted in a substantial interferon- α concentration in plasma at hour 6.

Discussion of the related Art

This example demonstrates that SNALP-formulated ApoB-targeted sirnas comprising minimal 2' OMe modifications at selected positions within the sense and antisense strands are capable of silencing up to 77% of plasma ApoB protein levels at extended time points 7 days after IV treatment relative to PBS control. Indeed, 2 'OMe-uridine and/or 2' OMe-guanosine selective modifications at less than about 30% (e.g., 16.7%, 19%, or 23.8%) of the nucleotide positions of both strands of the siRNA duplex typically produce similar silencing activity as the unmodified siRNA sequence. In addition, such 2' OMe modifications improve the toxicity profile of in vivo therapy by reducing the immunostimulatory properties of ApoB siRNA.

Method of producing a composite material

siRNA: siRNA duplexes are prepared by annealing two deprotected and desalted RNA polynucleotides. Each polynucleotide is designed to be 21 bases in length and each duplex is designed to contain a 19 base duplex region with 2 3' overhangs at each end of the duplex region. All duplexes were designed to be cross-reactive against mouse and human ApoB. The sense strand of ApoB 10048 siRNA corresponds to nucleotides 10164-10184 of human ApoB mRNA sequence NM-000384. The sense strand of ApoB10886 siRNA corresponds to nucleotides 11002-11022 of the human ApoB mRNA sequence NM-000384. The sense strand of the ApoB 10346 siRNA corresponds to nucleotides 10462 to 10482 of the human ApoB mRNA sequence NM-000384.

Lipid encapsulation of RNA: a "2: 40: 10" formulation of DSPC: cholesterol: PEG-C-DMA: DLinDMA (10: 48: 2: 40 molar ratio) SNALP was prepared using a direct dilution method at a target nucleic acid to lipid ratio of 0.04.

In vivo treatment protocol: BALB/c mice (female, 4 weeks old) were obtained from Harlan laboratories. Following an adaptation period of at least 7 days, animals were administered the SNALP formulations shown in table 4 by Intravenous (IV) injection in the lateral tail vein once daily (3 total doses per animal) on study days 0, 1 and 2. The dose was 2mg siRNA/kg body weight, corresponding to 10ml/kg (rounded to the nearest 10. mu.l). As a negative control, one group of animals was given IV injections of PBS vehicle. In-cage observations of body weight and animal behavior and/or appearance were recorded on study days 0-3, 9 and 16. Tail-biting was performed to collect a small amount (50 μ Ι) of whole blood, which was processed as plasma. On study day 16, animals were anesthetized with a lethal dose of ketamine/xylazine and blood was collected by cardiac puncture prior to cervical dislocation. Blood was collected in lavendar EDTA microcuvettes (microtainers) and plasma was processed.

Table 4. in vivo treatment regimens and SNALP formulations used in this study.

The analysis method comprises the following steps: ELISA methods are used, as described in Zlot et al, j. lipid Res, (journal of lipid research), 40: 76-84(1999), ApoB protein levels in plasma were measured. Interferon-alpha levels in plasma were measured using a sandwich ELISA method, following the supplier's instructions (PBL biomedicine; Piscataway, NJ).

Example 3 design of Eg5 siRNA with Selective chemical modification

This example illustrates that minimal 2' OMe modification at selected positions in the sense and/or antisense strand of an Eg5 siRNA duplex is sufficient to reduce the immunostimulatory properties of Eg5 siRNA while retaining RNAi activity. In particular, selective 2 'OMe-uridine and/or 2' OMe-guanosine modifications at less than about 20% of the nucleotide positions in one or both strands provide a bound Eg5 siRNA with desirable silencing and non-immunostimulatory properties.

Results

Selective modification of Eg5 siRNA retained antiproliferative activity. A panel of 2' OMe-modified Eg5 siRNA molecules was prepared and their RNAi activity was evaluated in human HeLa cells and mouse Neuro2A cells. The Eg5 siRNA duplexes used in this study are provided in table 5. The modifications comprise introducing 2 ' OMe-uridine and/or 2 ' OMe-guanosine at selected positions on the sense and/or antisense strand of the Eg 52263 siRNA sequence, wherein the siRNA duplex comprises less than about 20% 2 ' OMe-modified nucleotides. Antiproliferative activity was evaluated in a cell viability bioassay. In particular, cell viability of cell cultures was measured 48 hours after treatment with SNALP formulations containing Eg 52263 siRNA and expressed as mean fluorescence units. Figure 13 shows that selective chemical modification of Eg 52263 siRNA duplexes retained RNAi activity in human HeLa cells. Similarly, figure 14 shows that selective chemical modification of Eg 52263 siRNA duplexes retained RNAi activity in mouse Neuro2A cells.

TABLE 5 siRNA duplexes comprising sense and antisense Eg5 RNA polynucleotides.

Column 1: "0/0" ═ unmodified siRNA duplexes; "U/0" ═ 2' OMe-uridine modified Sense Strand (SS); "G/0" ═ 2' OMe-guanosine modified SS; "0/U" ═ 2' OMe-uridine modified Antisense Strand (AS); "0/G" ═ 2' OMe-guanosine modified AS; "U/U" ═ 2' OMe-uridine modified siRNA duplexes; "U/G" ═ 2 'OMe-uridine-modified SS and 2' OMe guanosine-modified AS; "G/G" ═ 2' OMe-guanosine modified siRNA duplexes; and "G/U" ═ 2 'OMe-guanosine modified SS and 2' OMe-uridine modified AS. Column 2: 2' OMe-modified nucleotides are in bold and underlined; "dT" is deoxythymidine.

Column 3: the number and percentage of 2' OMe-modified nucleotides in the siRNA duplex are provided. Column 4: the number and percentage of modified nucleotides in the double-stranded (DS) region of the siRNA duplex is provided.

Selective modification of Eg5 siRNA abrogated cytokine induction in vivo. Unmodified Eg 52263 siRNA (i.e., 0/0) and certain 2' OMe-modified variants thereof (i.e., U/0, G/0, U/U, and G/G) were encapsulated in SNALPs with 2 mol% PEG-cDMA, 40 mol% DLinDMA, 10 mol% DSPC, and 48 mol% cholesterol. These SNALP-formulated Eg 5-targeted sirnas were tested in vivo to look for induction of immune responses, e.g., induction of cytokines. BALB/c mice (n-3/treatment group) were injected with 40 μ g SNALP formulation containing Eg 52263 siRNA. At 6 hours post-treatment, samples were collected and levels of interferon- α were detected by ELISA assay. Figure 15 shows that selective 2' OMe modification of Eg 52263 siRNA abolished interferon induction associated with systemic administration of natural (i.e., unmodified) duplexes.

Selective modification of Eg5 siRNA abolished antibody response to the delivery vehicle. Unmodified Eg52263 siRNA (i.e., 0/0) and certain 2' OMe-modified variants thereof (i.e., U/0 and U/U) were encapsulated in SNALPs with 2 mol% PEG-cDMA, 40 mol% DLinDMA, 10 mol% DSPC, and 48 mol% cholesterol. These SNALP-formulated Eg 5-targeted sirnas were tested in mice in search of induction of an immune response against a delivery vehicle component such as PEG. Specifically, mice (n ═ 4) were treated with 3 × 2mg/kg daily with SNALP formulations containing Eg52263 siRNA, and serum levels of anti-PEG IgM and IgG antibodies were detected on day 10. Fig. 16 shows that the selective 2' OMe modification (i.e., U/U) of both strands of an Eg52263 siRNA duplex is a complete abolishment of the antibody response against the PEG component of the SNALP delivery vehicle associated with systemic administration of the native (i.e., unmodified) duplex.

Method of producing a composite material

siRNA: all siRNAs used in these studies were chemically synthesized by Protozoan vitamin therapy (Protiva Biotherapeutics) (Burnaby, BC), University of Caragary (University of Calgary) (Calgary, AB), or Dharmacon. siRNA was desalted and annealed using standard methods. The Eg52263 siRNA sense strand corresponds to nucleotides 2263-2281 of the human Eg5 mRNA sequence NM-004523.

Lipid encapsulation of siRNA: unless otherwise indicated, siRNAs are encapsulated in liposomes composed of the following lipids: synthetic cholesterol (sigma; St. Louis, Mo.), the phospholipid DSPC (1, 2-distearoyl-sn-glycerol-3-phosphocholine; Avanti Polar lipids; Alabaster, AL), the PEG-lipid PEG-cDMA (3-N- [ (-methoxypoly (ethylene glycol) 2000) carbamoyl ] -1, 2-dimyristyloxy-propylamine), and the cationic lipid DLinDMA (1, 2-dioleyloxy-3- (N, N-dimethyl) aminopropane), in a molar ratio of 48: 10: 2: 40. In other words, unless otherwise indicated, siRNAs were encapsulated in liposomes of the SNALP formulation described below: 2 mol% PEG-cDMA, 40 mol% DLinDMA, 10 mol% DSPC, and 48 mol% cholesterol. For the vector control, in the absence of siRNA, empty liposomes with the same lipid composition were formed.

In vivo cytokine induction: animal studies were completed following approval by the local Prolo vitamin treatment company (Protiva Biotherapeutics) animal care and use Committee, according to the Canadian animal Care guide Committee. Prior to use, 6-8 week old CD1 ICR mice (Harlan; Indianapolis, IN) were subjected to 3 weeks of quarantine and acclimation. The encapsulated siRNA formulation in 0.2ml PBS was administered into the lateral tail vein by standard intravenous injection. Blood was collected by cardiac puncture 6 hours after administration and processed as plasma for cytokine analysis. In the RNAi efficacy experiment, plasma was collected from 50 μ l of test blood samples 6 hours after initial siRNA administration. Interferon-alpha levels in plasma were measured using a sandwich ELISA method, following the supplier's instructions (PBL biomedicine; Piscataway, NJ).

Cell viability assay: using the commercial reagent CellTiter _ Blue^TM(Promega Corp.; Madison)WI), a resazurin dye, which is reduced by metabolically active cells to the fluorescent product resorufin (resorvufin), to assess cell viability of in vitro cell cultures. Various cancer cell lines were cultured in vitro using standard tissue culture techniques, and CellTiter-Blue was administered 48-72 hours after treatment with siRNA formulations or small molecule drugs^TMReagents were added to the cultures to quantify the metabolic activity of the cells, a measure of cell viability.

Antibody detection: an ELISA was developed to detect IgM and IgG antibodies to PEG-lipids and other lipid components of SNALP. Briefly, 10. mu.g of PEG-cDSA was added to 20. mu.l of 100% ethanol in 96-well plates containing PVDF membranes (Millipore Corp.); Bedford, Mass.). PEG-cDSA-coated membranes were allowed to air dry completely for 2 hours before blocking with 10% FBS in PBS for 1 hour. Then 100 μ l of serum samples serially diluted in blocking buffer were added to two wells in a duplicate for 1 hour and washed 4 times with 1% FBS in PBS. Plate bound antibodies were detected with HRP-conjugated goat anti-IgM Fc μ or IgG Fc γ. The bound enzyme was developed with TMB substrate, stopped with 2N sulfuric acid and then read on a spectrophotometer at 450nm-570 nm.

Example 4 design of anti-influenza siRNA with selective chemical modification.

This example illustrates that minimal 2' OMe modifications at selected positions in the sense strand of an influenza Nucleocapsid Protein (NP) siRNA duplex are sufficient to reduce the immunostimulatory properties of NP siRNA while retaining RNAi activity. In particular, selective 2' OMe-uridine modifications at less than about 20% of the nucleotide positions in the sense strand provide bound NPSiRNA with desirable silencing and non-immunostimulatory properties.

Results

Selective modification of NP siRNA retained viral knockdown activity. A panel of 2' OMe-modified NP siRNAs was prepared and their RNAi activity was evaluated in Madin-Darby canine kidney (MDCK) cells. NP siRNA duplexes used in this study are provided in table 6. The modification comprises introducing 2 'OMe-uridine at selected positions on the sense strand of the NP siRNA sequence, wherein the siRNA duplex comprises less than about 20% 2' OMe-modified nucleotides. NP siRNA molecules were formulated into lipid nucleic acid complexes and tested for their ability to significantly reduce cytopathic effect (CPE) produced by influenza virus approximately 48 hours after infection. The NP siRNA molecules were also tested for the amount of HA produced (i.e., HA units/well) and the percentage of HA produced (i.e., percentage knockdown) relative to the virus alone control.

TABLE 6siRNA duplexes comprising sense and antisense NP RNA polynucleotides.

Column 1: the numbers refer to the nucleotide position of the 5' base of the sense strand relative to the influenza a virus NP ssRNA sequence NC _ 004522. Column 2: the numbers refer to the distribution of 2' OMe chemical modifications in each chain. For example, "U5/0" indicates 5 uridine 2 'OMe modifications in the sense strand, and uridine-free 2' OMe modifications in the antisense strand. Column 3: 2' OMe-modified nucleotides are in bold and underlined; "dT" is deoxythymidine.

Column 4: the number and percentage of 2' OMe-modified nucleotides in the siRNA duplex are provided. Column 5: the number and percentage of modified nucleotides in the double-stranded (DS) region of the siRNA duplex is provided.

Figures 17 and 18 show that selective 2' OMe modification of the sense strand of the NPsiRNA duplex did not adversely affect influenza knockdown activity when compared to the unmodified negative sequence or control sequence. Figure 19 shows that various combinations of these 2' OMe-modified NP siRNA molecules provide increased knockdown of influenza virus in MDCK cells relative to controls.

Selective modification of NP siRNA abolished cytokine induction in vitro and in vivo. Unmodified NP 1496siRNA (i.e., 0/0) and its 2' OMe-modified variant (i.e., U8/0) were encapsulated in SNALPs with 2 mol% PEG-cDMA, 40 mol% DLinDMA, 10 mol% DSPC, and 48 mol% cholesterol, or complexed with Polyethylenimine (PEI) to form polymer-nucleic acid complexes. SNALP-formulated NP-targeted sirnas were tested in vitro for induction of immune responses, e.g., cytokine induction. Human Peripheral Blood Mononuclear Cells (PBMCs) were transfected with 40 μ g of SNALP preparation containing NP 1496siRNA and at 16 hours, supernatants were collected for cytokine analysis. The polymer-nucleic acid complex preparations are tested in vivo for induction of an immune response, e.g., cytokine induction. Mice were injected intravenously with 120 μ g siRNA/mouse and plasma samples were collected 6 hours after treatment and assayed for interferon- α levels by ELISA. Figure 20 shows that selective 2' OMe modification of NP 1496siRNA abrogated interferon induction in vitro cell culture systems. Figure 21 shows that selective 2' OMe modification of NP 1496siRNA abolished induction of interferon associated with systemic administration of natural (i.e., unmodified) duplexes.

Method of producing a composite material

siRNA: all siRNAs used in these studies were chemically synthesized by Protozoan vitamin therapy (Protiva Biotherapeutics) (Burnaby, BC), University of Caragary (University of Calgary) (Calgary, AB), or Dharmacon. siRNA was desalted and annealed using standard methods.

Lipid encapsulation of siRNA; unless otherwise indicated, siRNAs are encapsulated in liposomes composed of the following lipids: synthetic cholesterol (sigma; St. Louis, Mo.), the phospholipid DSPC (1, 2-distearoyl-sn-glycerol-3-phosphocholine; Avanti Polar lipids; Alabaster, AL), the PEG-lipid PEG-cDMA (3-N- [ (-methoxypoly (ethylene glycol) 2000) carbamoyl ] -1, 2-dimyristyloxy-propylamine), and the cationic lipid DLinDMA (1, 2-dioleyloxy-3- (N, N-dimethyl) aminopropane), in a molar ratio of 48: 10: 2: 40. In other words, unless otherwise indicated, siRNAs were encapsulated in liposomes of the SNALP formulation described below: 2 mol% PEG-cDMA, 40 mol% DLinDMA, 10 mol% DSPC, and 48 mol% cholesterol.

Lipid nucleic acid complex treatment and in vitro influenza infection: when trypsin is present, influenza virus (e.g., influenza A/PR/8/34H1N1) produces a cytopathic effect in MDCK cells upon infection. Lipid nucleic acid complex treatment of MDCK cells and in vitro influenza infection were performed as follows:

1. MDCK cells were plated at about 8000 cells/well (about 4X 10)⁴Individual cells/ml) were seeded in 96-well plates such that 24 hours after seeding, the cells were at a density of about 50%.

2. After about 24 hours, the medium was changed to fresh complete medium (no antibiotics) and the cells were used in Lipofectamine^TM2000(LF2000) (Invitrogen; Camarillo, CA) comprising nucleic acid liponucleic acid complexes transfection, nucleic acids: LF2000 ratio was 1: 4.

3. After about 4 hours, the complexes were removed, the cells were washed with PBS, and infected with various influenza virus dilutions in virus infection medium (DMEM, 0.3% BSA, 10mM HEPES), added to about 50 μ l of diluted virus per well.

4. The virus was incubated on the cells at 37 ℃ for about 1-2 hours, then the virus was removed and about 200. mu.l of virus growth medium (DMEM, 0.3% BSA, 10mM HEPES, 0.25. mu.g/ml trypsin) was added.

5. At about 48 hours, the cells were monitored for cytopathic effects.

6. The supernatants were subjected to influenza HA Enzyme Immunoassay (EIA).

Polymer-nucleic acid complex therapy and in vivo cytokine induction: animal studies were completed following approval by the local Prolo vitamin treatment company (Protiva Biotherapeutics) animal care and use Committee, according to the Canadian animal Care guide Committee. Prior to use, 6-8 week old CD1 ICR mice (Harlan; Indianap) olis, IN) were performed for 3 weeks quarantine and acclimation time. In vivo jetPEI for siRNAs^TM(Qbiogene, Inc.; Carlsbad, Calif.) was mixed at room temperature for 20 minutes at a N/P ratio of 5 according to the supplier's instructions. Mice were administered with jetPEI in vivo in 0.2ml PBS^TMPolymer-nucleic acid complexes, corresponding to 120. mu.g siRNA/mouse, were administered into the lateral tail vein by standard intravenous injection. Blood was collected by cardiac puncture 6 hours after administration and processed as plasma for cytokine analysis. Interferon-alpha levels in plasma were measured using a sandwich ELISA method, following the supplier's instructions (PBL biomedicine; Piscataway, NJ). Other methods for PEI polymer-nucleic acid complex formation are described in Judge et al, nat. biotechnol, (natural biotechnology), 23: 457-462 (2005).

In vitro cytokine induction: PBMCs were transfected with 40 μ g of SNALP formulated siRNA and the interferon- α levels in the cell culture supernatant were measured after 16 hours using a sandwich ELISA method according to the supplier's instructions (PBL biomedicine; Piscataway, N.J.).

It is to be understood that the above description is intended to be illustrative, and not restrictive. Many embodiments will be apparent to those of skill in the art upon reading the above description. The scope of the invention should, therefore, be determined not with reference to the above description, but should instead be determined with reference to the appended claims, along with the full scope of equivalents to which such claims are entitled. The disclosures of all papers and references, including patent applications, patents, PCT publications, and Genbank accession numbers, are incorporated herein by reference for all purposes.

Claims (29)

1. A modified siRNA comprising a double-stranded region of 19 to 25 nucleotides in length, wherein 15% -30% of the nucleotides in said double-stranded region comprise 2' -O-methyl nucleotides,

wherein the modified siRNA comprises 2' -O-methyl nucleotides in both strands of the modified siRNA,

wherein the modified siRNA comprises at least one 2 '-O-methylguanosine nucleotide and at least one 2' -O-methyluridine nucleotide in the double-stranded region,

wherein 2 '-O-methylguanosine nucleotides and 2' -O-methyluridine nucleotides are the only modified nucleotides present in the double-stranded region,

wherein the corresponding unmodified siRNA sequence of said modified siRNA is capable of silencing expression of a target sequence and said modified siRNA is less immunostimulatory than the corresponding unmodified siRNA sequence, and

wherein the modified siRNA is capable of silencing expression of the target sequence.

2. A modified siRNA comprising a double-stranded region of 19 to 25 nucleotides in length, wherein 15% -30% of the nucleotides in said double-stranded region comprise 2' -O-methyl nucleotides,

wherein the modified siRNA comprises 2' -O-methyl nucleotides in both strands of the modified siRNA,

Wherein the modified siRNA comprises at least one 2 '-O-methylguanosine nucleotide and at least one 2' -O-methyluridine nucleotide in the double-stranded region,

wherein the modified siRNA does not comprise a 2' -O-methylcytosine nucleotide in the double-stranded region,

wherein the 2' -O-methyl nucleotide is the only modified nucleotide present in the double-stranded region,

wherein the corresponding unmodified siRNA sequence of said modified siRNA is capable of silencing expression of a target sequence and said modified siRNA is less immunostimulatory than the corresponding unmodified siRNA sequence, and

wherein the modified siRNA is capable of silencing expression of the target sequence.

3. The modified siRNA of claim 1 or 2, wherein the modified siRNA is chemically synthesized.

4. The modified siRNA of claim 1 or 2, wherein the modified siRNA comprises a 3' overhang in one or both strands of the modified siRNA.

5. The modified siRNA of claim 4, wherein the 3' -end overhang of one or both strands of said modified siRNA comprises unmodified nucleotides, modified nucleotides, or a combination thereof.

6. The modified siRNA of claim 1 or 2, wherein the modified siRNA does not comprise a 3' overhang.

7. The modified siRNA of claim 1 or 2, wherein 20% to 30% of the nucleotides in the double-stranded region comprise 2' -O-methyl nucleotides.

8. The modified siRNA of claim 1 or 2, wherein 25% to 30% of the nucleotides in the double-stranded region comprise 2' -O-methyl nucleotides.

9. The modified siRNA of claim 1 or 2, wherein the modified siRNA has an interferon- α response that is lower than the interferon- α response detected for the corresponding unmodified siRNA sequence.

10. The modified siRNA of claim 1 or 2, wherein the modified siRNA has an IC less than or equal to the corresponding unmodified siRNA sequence₅₀2 times of IC₅₀。

11. The modified siRNA of claim 1 or 2, further comprising a carrier system.

12. The modified siRNA of claim 11, wherein the carrier system is selected from the group consisting of nucleic acid-lipid particles, liposomes, micelles, viral particles, nucleic acid complexes, and mixtures thereof.

13. A nucleic acid-lipid particle, comprising:

the modified siRNA of claim 1 or 2;

a cationic lipid; and

a non-cationic lipid.

14. The nucleic acid-lipid particle of claim 13, wherein the cationic lipid is a member selected from the group consisting of: n, N-dioleyl-N, N-dimethylammonium chloride, N, N-distearyl-N, N-dimethylammonium bromide, N- (1- (2, 3-dioleyloxy) propyl) -N, N, N-trimethylammonium chloride, N, N-dimethyl-2, 3-dioleyloxypropylamine, 1, 2-dioleyloxy-N, N-dimethylaminopropane, 1, 2-dillenyloxy-N, N-dimethylaminopropane, and mixtures thereof.

15. The nucleic acid-lipid particle of claim 13, wherein the non-cationic lipid is a member selected from the group consisting of: distearoylphosphatidylcholine, dioleoylphosphatidylethanolamine, palmitoyloleoylphosphatidylcholine, palmitoyloleoylphosphatidylethanolamine, palmitoyloleoylphosphatidylglycerol, dipalmitoylphosphatidylcholine, dipalmitoylphosphatidylethanolamine, dimyristoylphosphatidylethanolamine, distearoylphosphatidylethanolamine, monomethyl-phosphatidylethanolamine, dimethyl-phosphatidylethanolamine, dioleoylphosphatidylethanolamine, stearoyloleoylphosphatidylethanolamine, egg phosphatidylcholine, cholesterol, and mixtures thereof.

16. The nucleic acid-lipid particle of claim 13, further comprising a conjugated lipid that inhibits aggregation of the particle.

17. The nucleic acid-lipid particle of claim 16, wherein the conjugated lipid that inhibits aggregation of particles is a member selected from the group consisting of polyethylene glycol-lipid conjugates, polyamide-lipid conjugates, and mixtures thereof.

18. The nucleic acid-lipid particle of claim 16, wherein the conjugated lipid that inhibits aggregation of particles comprises a polyethylene glycol-dialkoxypropyl conjugate.

19. The nucleic acid-lipid particle of claim 13, wherein the cationic lipid comprises from 2 mol% to 60 mol% of the total lipid present in the particle.

20. The nucleic acid-lipid particle of claim 13, wherein the non-cationic lipids comprise from 5 mol% to 90 mol% of the total lipid present in the particle.

21. The nucleic acid-lipid particle of claim 16, wherein the conjugated lipid that inhibits aggregation of particles comprises 0.5 mol% to 20 mol% of the total lipid present in the particle.

22. The nucleic acid-lipid particle of claim 13, wherein the modified siRNA is completely encapsulated within the nucleic acid-lipid particle.

23. A pharmaceutical composition comprising the modified siRNA of claim 1 or 2 or the nucleic acid-lipid particle of claim 13 and a pharmaceutically acceptable carrier.

24. An in vitro method of introducing into a cell an siRNA that silences expression of a target sequence, the method comprising:

contacting the cell with the modified siRNA of claim 1 or 2 or the nucleic acid-lipid particle of claim 13.

25. The modified siRNA of claim 1 or 2 or the nucleic acid-lipid particle of claim 13 for use in a method of delivering an siRNA that silences expression of a target sequence in vivo,

Wherein said siRNA or said particle is administered to a mammalian subject.

26. An in vitro method for preparing the modified siRNA of claim 1, the method comprising:

(I) providing an unmodified siRNA sequence capable of silencing expression of a target sequence and comprising a double-stranded region 19 to 25 nucleotides in length; and

(II) modifying the siRNA by replacing 15% to 30% of the nucleotides in the double-stranded region with 2' -O-methyl nucleotides, thereby producing a modified siRNA that is less immunostimulatory than an unmodified siRNA sequence and is capable of silencing expression of the target sequence,

wherein the modified siRNA comprises 2' -O-methyl nucleotides in both strands of the modified siRNA,

wherein the modified siRNA comprises at least one 2 '-O-methylguanosine nucleotide and at least one 2' -O-methyluridine nucleotide in the double-stranded region,

wherein 2 '-O-methylguanosine nucleotides and 2' -O-methyluridine nucleotides are the only modified nucleotides present in the double-stranded region.

27. An in vitro method for preparing the modified siRNA of claim 2, the method comprising:

(I) providing an unmodified siRNA sequence capable of silencing expression of a target sequence and comprising a double-stranded region 19 to 25 nucleotides in length; and

(II) modifying the siRNA by replacing 15% to 30% of the nucleotides in the double-stranded region with 2' -O-methyl nucleotides, thereby producing a modified siRNA that is less immunostimulatory than an unmodified siRNA sequence and is capable of silencing expression of the target sequence,

wherein the modified siRNA comprises 2' -O-methyl nucleotides in both strands of the modified siRNA,

wherein the modified siRNA comprises at least one 2 '-O-methylguanosine nucleotide and at least one 2' -O-methyluridine nucleotide in the double-stranded region,

wherein the modified siRNA does not comprise a 2' -O-methylcytosine nucleotide in the double-stranded region,

wherein the 2' -O-methyl nucleotide is the only modified nucleotide present in the double-stranded region.

28. The method of claim 26 or 27, wherein step (I) comprises:

(a) contacting the unmodified siRNA sequence with a mammalian effector cell under conditions suitable for the effector cell to produce a detectable immune response; and

(b) identifying the unmodified siRNA sequence as an immunostimulatory siRNA by the presence of a detectable immune response in the effector cell.

29. Use of the modified siRNA of claim 1 or 2 or the nucleic acid-lipid particle of claim 13 to formulate a pharmaceutical composition for silencing expression of a target sequence in a cell.