Small molecule conjugates for intracellular delivery of nucleic acids

- 1 SMALL MOLECULE CONJUGATES FOR INTRACELLULAR DELIVERY OF NUCLEIC ACIDS The present invention relates to use of novel small molecule conjugates for the delivery of nucleid acids, such as siRNA.

The delivery of nucleic acids into a living cell is highly restricted by the complex membrane system of the cell.

One means that has been used to deliver nucleic acids in vivo has been to attach the nucleic acid to either a small targeting molecule or a hydrophobic molecule such as a lipid or sterol.

While some delivery and activity has been observed with these conjugates when administered to rodents, the dose required has been prohibitively large, resulting often in undesired toxicity effects in vivo and in high costs and impracticable treatment regimen when translated to humans.

Provided herein is the use of small molecule compounds for the delivery of nucleid acids, such as siRNA.

When the small molecule compounds are conjugated to the nucleic acid, they mediate successful delivery of the nucleic acid into a cell.

Surprisingly it has been found that significantly decreased doses of the nucleic acid are now sufficient for successful delivery when using the novel compounds provided herein.

Thus, the use of the compounds provides a powerful tool for the delivery of nucleic acids with considerably limited toxicity in vivo.

In one embodiment, the present invention is directed to the use of compounds of formula 110 eelR1 0 R2 0,µ ).N.....õõNsirNH . n R3 H 0 0 CA 2976966 2017-08-21 -2for the delivery of nucleic acids, wherein Y is a linker group selected from -(CH2)3- or -C(0)-N-(CH2-CH2-0)p-CII2-CH2-;

RI is -(C1-6) alkyl;

-(CH2)-naphthyl; or -(CH2)111-phenyl, which phenyl is unsubstituted or up to four times substituted with a substituent independently selected from -NO2, -CN, Halogen, -0-(CH2)-phenyl, -0-(C1-6) alkyl, or -C(0)-NH2;

R2 is hydrogen;

-(CH2)k-N-C(Ph)3, which phenyl rings are unsubstitutcd or independently substituted with ¨0-(C1-4)alkyl;

-(CH2)1 -C(0)-NH2;

-(CH2)k -phenyl;

-(C1-6) alkyl, which is unsubstituted or once substituted with ¨S-CH3;

R3 is -NH-phenyl, which phenyl group is further substituted with a substituent independently selected from -(CH2)-0H; or -(CH2)-0-C(0)-0-(4-nitro-phenyl);

CA 2976966 2017-08-21 -3k is 1, 2, 3, 4, 5, 6;

m is I, 2, 3 or 4;

n is 0 or 1; and p is an integer from 1 to 20, In another embodiment, use of the compounds of formula (I) with the specific conformation as shown in formula (Ia) H.., R1 0 R2 H 0 hr R3 0 `N 0 0 (Ia), for the delivery of nucleic acids is provided, wherein all substituents RI, R2, R3 and Y as well as the variables k, m, n, and p have the meaning given above.

In yet another embodiment, the present invention is directed to the use of compounds of formula (I) or (Ia) for the delivery of nucleic acids, wherein y is -(cH2)3-; and all remaining substituent groups have the meaning given above.

In yet another embodiment, the present invention is directed to the use of compounds of formula (I) or (Ia) for the delivery of nucleic acids, wherein Y is -C(0)-N-(CH2-CH20)p-CH2-CH2-; and all substituent groups have the meaning given above.

In yet another embodiment, use of the compounds of formulae (I) or (Ia) the delivery of nucleic acids is provided, wherein Y is -(CH2)3-;

R2 is -(CF2)k-N-C(Ph)3, which phenyl rings are unsubstituted or independently substituted with ¨0-(C1-4)alkyl; and CA 2976966 2017-08-21 -4R3 is -NH-phenyl, which phenyl group is further substituted with -(CH2)-0-C(0)-0-(4-nitro-phenyl);

n is 0; and RI and k have the meanings given above.

In yet another embodiment, there is provided the use of compounds of formulae (I) or (Ia) for the delivery of nucleic acids, wherein Y is ¨C(0)-NH-(CH2-012-0)p-CH2-CH2-;

R2 is -(CH2)k-N1-1-C(Ph)3, which phenyl rings are unsubstituted or independently substituted with ¨0-(C1-4)alkyl; and R3 is -NH-phenyl, which phenyl group is further substituted with -(CH2)-0-C(0)-0-(4-nitro-phenyl);

n is 0; and RI, k and p have the meanings given above The term "(C1-6) alkyl" as used herein means a linear or branched, saturated hydrocarbon containing from 1 to 6 carbon atoms.

Preferred C1-6 alkyl groups include methyl, ethyl, propyl, iso-propyl, butyl, 2-butyl and the like.

The term "halogen" as used herein means fluorine, chlorine, bromine, or iodine with fluorine and chlorine being preferred.

The compounds for use in delivery of nucleic acids according to the present invention can be generally obtained using methods known to the person of ordinary skill in the art of organic- or medicinal chemistry.

Likewise, it is understood that the cholesterol moiety can be replaced by other natural or chemically synthesized compounds of the steroid class (e.g.

cholanic acid, lithocholic acid etc) or other small molecules (e.g. vitamins) known to be effective in nucleic acid delivery such as tocopherol (Molecular Therapy, 2008, 16, 734).

CA 2976966 2017-08-21 -5For successful delivery of the nucleic acids, the compounds of formula (I) or (Ia) are covalently attached to the nucleic acids.

Preferably, the covalent bond is created by the reaction of a suitable functional group, such as i.e. a primary amine group, in the nucleic acid with the activated carbonyl group in the ¨0-C(0)-0- moiety of R3 as defined herein before.

Hence provided herein is a conjugate comprising the compounds of formula (I) or (Ia) and a nucleic acid.

The term "nucleic acid" as used herein refers to any form of DNA, including cDNA, or RNA, or a fragment thereof, nucleotide, nucleoside, oligonucleotides (including antisense oligonucleotides, LNA and siRNA), that causes a biological effect when administered in vivo to an animal, including but not limited to birds and mammals, including humans.

Preferred nucleic acids used herein are siRNAs.

The conjugate comprising the compounds covalently attached to a nucleic acid shows an improved ability to be taken up by cells compared to said nucleic acid alone.

Once the conjugate is delivered into the cell and trafficking to the lysosome, the corresponding nucleic acid is released by enzymatic cleavage.

This cleavage preferably takes place when a dipeptide motif, preferably consisting of the sequence a- or 13-(phenyl)alanine and lysine as present in the compounds of formula (I) or (Ia) is incorporated in the conjugate (see scheme 1).

Most preferably the conjugate contains the di-peptide motif and a spacer such as the paminobenzylcarbamatc spacer (Bioconjugate Chem. 2002,13,855) that spontaneously fragments once the amide bond C-terminal of the di-peptide motif is cleaved as exemplified for siRNAs in scheme 2.

Hence the conjugates comprising compounds of formula (I) or (Ia) are also referred to as dipeptide containing cholesterol conjugates.

Enzymatic cleavage of the nucleic acid from the dipeptide containing cholesterol conjugates of this invention is catalyzed by innate proteases of the cell.

One example of an innate protease capable of cleaving the di-peptide motif present in the compounds of formula (I) or (Ia) is Cathepsin B.

Cathepsin B is a known ubiquitous cysteine protease located in the lysosomes of mammalian cells (Bioconjugate Chem.

2002,13,855;

J.Med.Chem. 2005,48,1344; Nat.

Biotechnology 2003.21,778).

Thus, the dipeptide motif described above is also referred to as Cathepsin-cleavable dipeptide-motif.

The present invention therefore also provides a method for delivery of a nucleic acid, into cells wherein said nucleic acid may subsequently be cleaved off the conjugate to unfold a therapeutic activity.

CA 2976966 2017-08-21 -6In a further embodiment of the present invention, there is provided the use of a conjugate of the compounds of formula (I) or (Ia) covalently attached to a siRNA for intracellular delivery.

Conjugates of formula (I) or (Ia) covalently attached to a nucleic acid are designated herein as formula (II) or (Ha), respectively.

Di-peptide motif with R1-(phenyl)alanine moiety R2-lysine moiety Sit H 0 R2 H H 0-0 R1 N_R3 0 n N ' '`N = 0 enzymatic cleavage with biologically active substance attached in R3 Scheme 1 Therefore, in a further embodiment, the present invention provides a compound of formula *0 0411 R1 0 Fr 0 sY nucleic acid n H 0 0 (11) wherein CA 2976966 2017-08-21 -7Ra is ¨(CH2)k-NH2;

R1 and k have the meanings given for formula (I) above.

In a more specific embodiment, the present invention provides compounds of formula H R1 0 Ra H 0 N _______________ N 0 nucleic acid 0 (Ha) wherein Ra is ¨(CH2)k-NH2;

RI and k have the meanings given for formula (I) above.

In a preferred embodiment, the nucleic acid in formula (11) or (IIa) is a siRNA.

The compounds of formula (II) or (IIa) may have valuable properties in therapy.

Therefore, in a further embodiment, there are provided the compounds of formula (II) or (IIa) for use as medicaments.

Another embodiment of the invention is a pharmaceutical composition comprising the conjugates of the compounds of formula (I) or (Ia) covalently attached to a nucleic acid.

In still another embodiment of the invention there is provided a pharmaceutical composition comprising the compounds of formula (IIa) together with pharmaceutically acceptable excipients.

Below embodiments arc exemplified for conjugates of the compounds of formula (I) or (Ia) covalently attached to siRNA.

It is understood that these embodiments are also applicable for other types of nucleic acids as defined above.

The covalent attachment of the siRNA to the compounds of formula (I) or (Ia) is achieved via reaction of a suitable nucleophilic group, i.e. a primary amine group, in the siRNA CA 2976966 2017-08-21 -8with the activated ¨C(0)- group in R3 of said compounds of formula (I) or (Ia).

The activation of that ¨C(0)- group is obtained by a p-nitrophenoxy carbonate as shown in scheme 2 below.

Cholesterol moiety 1. siRNA binding site;

H4-0 ;e activated -C(0)- group Cw0 R1 0R2 NNh H Ol y 0 o \SI 0 \ [I 0 2. site for enzymatic cleavage in lysosome 15 (scheme 2) The p-nitrophenyl activated carbonate may for example be reacted with the siRNA equipped with a suitable nucleophile such as the primary amine of a hexylaminolinker to generate a carbamate linkage to yield the covalent siRNA conjugate.

Once the siRNA is taken up intracellularly and transfered to the lysosome the compounds of formula (II) or (IIa) wherein the 25 biological active substance is a siRNA are cleaved by the protease activity releasing the siRNA via a 1,6-elimination reaction as also shown in scheme 3.

The cholesterol moiety of the conjugate of the compounds of formula (II) or (Ha) modifies the PK properties of siRNA in such a way that systemic administration enables gene silencing in vivo.

In one embodiment the compounds of formula (II) or (IIa) wherein the nucleic acid is 30 siRNA is co-administered with a delivery polymer.

Delivery polymers provide a means of disrupting cell membranes and mediate endosomal release.

In another embodiment, said delivery polymer and thc siRNA conjugate of the invention are not covalently attached and synthesized separately and may be supplied in separate containers or a single container.

Delivery polymers for oligonucleotides such as siRNA are well known in the art.

35 For example, Rozema et al., in U.S.

Patent Publication 20040162260 demonstrated a means to reversibly regulate membrane disruptive activity of a membrane active polyamine.

Reversible regulation provided a means to limit activity to the endosomes of target cells, thus CA 2976966 2017-08-21 -9limiting toxicity.

Their method relied on reaction of amines on the polyamine with 2-propionic3-methylmaleic anhydride.

This modification converted the polycation to a polyanion via conversion of primary amines to carboxyl-containing groups and reversibly inhibited membrane activity of the polyamine.

1 ,SiFINA N IL' 0 I. 0 N NN/NN moietr 0 R" 0 proteolysis 1,6-elirrination 140 + C 2 + H2N,siRNA Eliolesterdmoie)y VN7N R HN R" CH 0 Scheme 3 To enable co-delivery of the nucleic acid with the delivery vehicle, the nucleic acid was covalently linked to the delivery polymer.

In US provisional patent application 61/307490 a new generation of delivery polymers is described.

Therein, membrane active polyamine comprising an amphipathic terpolymer formed by random polymerization of amine-containing monomers, lower hydrophobic monomers, and higher hydrophobic monomers are provided.

This new generation of delivery polymers removed the requirement that polynucleotide and polymer are associated either by covalent linkage or by charge-charge interaction.

Non-limiting examples of delivery polymers used for co-administration with the siRNA conjugates of the present invention are membrane active polyamines and poly(vinyl ether) (PBAVE), Dynamic PolyConjugates (DPC; Rozema et al. 2007) and improved DPCs as disclosed in US provisional patent application 61/307490.

CA 2976966 2017-08-21 -10In a further embodiment, a new chemical siRNA modification pattern for functional in vivo delivery is provided.

This new chemical siRNA modification pattern is especially useful with delivery vehicles which display a relatively strong endosomal / lysosomal retention.

It was found that siRNA stabilization against degradation by endosomal / lysosomal-localized nucleases such as DNAse II strongly improves target knock down.

Such stabilization may directly effect the amount of siRNA released into the cytoplasm where the cellular RNAi machinery is located.

Only the siRNA portion available in the cytoplasm is able to trigger the RNAi effect.

In addition to poor pharmacokinetic characteristics, siRNAs are susceptible to nucleases in the biological environment when administered as such into the circulation without a protecting delivery vehicle.

Accordingly, many siRNAs are rapidly degraded either extracellularly in the tissue and blood stream or after intracellular uptake (e.g. in the endosome).

One well known nuclease localized in the endosomal / lysosomal compartment is DNase II.

This enzyme is active at a pH below 6-6.5 with maximum activity in the pH-range of 4.5-5, reflecting conditions present in the acidified environment of the endosomal/ lysosomal compartment.

The following RNA degradation pathways induced by DNase II were identified in vitro and are disclosed in this invention:

A. RNA strands containing at least one 2'-OH nucleotide are rapidly degraded via a cyclic pentavalent phosphorus intermediate, leading to 2'-3' cyclic phosphates at the 5'-cleavage product.

The formation of the pentavalent intermediate can be inhibited by nucleotides lacking a 2'-OH group such as 2'-deoxy, 2'-0-methyl (2'-0Me)or 2'-deoxy-2'-fluoro (2'-F) nucleotides.

B.

Additionally, RNA is degraded in a 5'-exonucleolytic pathway independent of the 2% modification on the 5'-terminal nucleotides.

This degradation pathway can be inhibited by 5'terminal non-nucleotide moieties, like e.g. cholesterol, aminoalkyl-linker or a phosphorothioate at the first internucleotide linkage.

C. A 5'-phosphate also protects and slows down the exonucleolytic cleavage kinetics, but can not fully block this pathway.

This is most probably due to the cleavage of the 5'-phosphate by phosphatases or an inherent phosphatase activity of the DNase II enzyme preparation used in the in vitro stability assay.

CA 2976966 2017-08-21 -11D.

The best protection was achieved with oligonucleotides lacking any 2'-OH nucleotide within the strand, starting with a 2'-0Me nucleotide at the 5'-end connected by a phosphorothioate (PTO) linkage to the second nucleotide.

Other terminal nucleotides lacking a 2'-OH group also protect against the 5'-exo degradation, but to a lower extent compared to the 2'-0Me tnodification.

Hence the inventors of the present invention found that siRNAs can be significantly stabilized when using the following design, wherein an oligonucleotide is provided with an antisense strand with the modification pattern: 5'-(w)- (Z1)- (Z2)- (Z3)na-3' and a sense strand with the modification pattern 5'- (Z3)n , wherein w is independently a 5'-phosphate or 5'-phosphothioate or H, Z1 is independently a 2'-modified nuleoside.

Z2 is independently a 2'-deoxy nucleoside or 2'-Fluoro-modified nucleoside, Z3 is independently a 2'-modified nucleoside, na is 8-23 and ns is 8-25.

In one preferred embodiment an oligonucleotide is provided with an antisense strand with the modification pattern: 5'-(w)- (Z1)- (Z2)- (Z3) na -3' and a sense strand with the modification pattern 5'- (Z3) ns -3' , wherein Z1 is a 2'-Fluoro-modified nucleoside or a 2deoxy-nucleoside and all remaining substituents as well as the variables n, and Its have the meaning given above.

In one preferred embodiment an oligonucleotide is provided with an antisense strand with the modification pattern: 5'-(w)- (Z1)- (Z2)- (Z3) na -3' and a sense strand with the modification pattern 5'- (Z3) ns -3' , wherein Z3 is a 2'-0-Methyl modified nucleoside, a 2'-Fluoro-modified nucleoside or a 2deoxy-nucleoside and all remaining substituents as well as the variables na and ns have the meaning given above.

In one preferred embodiment an oligonucleotide is provided with an antisense strand with the modification pattern: 5'-(w)- (Z1)- (Z2)- (Z3) na -3' and a sense strand with the modification pattern 5'- (Z3) rts -3' , wherein Z1 is a 2'-Fluoro-modified nucleoside or a 2'deoxy-nucleoside and Z3 is a 2'-0-Methyl modified nucleoside, a 2'-Fluoro-modified nucleoside or a 2deoxyCA 2976966 2017-08-21 -12nucleoside and all remaining substituents as well as the variables na and ns have the meaning given above.

The nucleosides in the nucleic acid sequence of the oligonucleotide with the novel modification pattern can either be linked by 5'-3' phosphodiesters or 5'-3' phosphorothioates.

As used herein, the "anti-sense" strand is the siRNA strand that is complementary to the target mRNA and that will be binding to the mRNA once the siRNA is unwound.

The sense strand of said siRNA comprising the novel modification pattern is complimentary to the antisense strand.

Said siRNA comprising the novel modification pattern proofed to be particularly advantageous when covalently attached to a delivery polymer as exemplified by Rozema et al.

(Dynamic PolyConjugates (DPC; Rozema et al. 2007).

Potency and duration of effect can be significantly enhanced employing the siRNA modification strategy outlined in this invention.

In another embodiment, said siRNA comprising the novel modification pattern are especially useful when conjugated to small molecules that alter the pharmacokinetic properties of siRNA such as cholesterol or the compounds of formula (I) and (Ia) provided herein.

In one embodiment a conjugate of a small molecule and an oligonucleotide is provided wherein the oligonucleotide has the following modification pattern: the antisense strand with the modification pattern: : 5'-(w)- (Z1)- (Z2)- (Z3) na -3' and a sense strand with the modification pattern 5'- (Z3) ns -, wherein the substituents as well as the variables na and ns have the meaning given above.

In one embodiment said small molecule is cholesterol.

In another embodiment said small molecule is a compound of formula (I) or (I a), resulting in compounds of formula (11) or (IIa).

Preferably, said siRNAs conjugates are co-administered with a delivery polymer.

Suitable delivery polymers are described above.

In one embodhnent, said siRNA comprising the novel modification pattern are especially useful when conjugated to a ligand that is known to bind to a specific receptor which internalizes the conjugate into a cell.

Particularly, the asialoglycoprotein receptor (ASGPR) expressed on hcpatocytes is a well-known receptor enabling the clearance (endocytosis and lysosomal degradation) of desialylated proteins from circulation.

It has been shown that the N-Acetyl-DCA 2976966 2017-08-21 = galactosamine has a high binding affinity for the receptor, especially when presented multivalent and when the galactose residues are properly spaced (J Biol Bhem, 2001, 276, 37577).

In order to utilize this high capacity receptor for receptor mediated endocytosis of the nucleic acid, the synthetic ligand shown below was prepared to be covalently attached to the siRNAs comprising the novel modification pattern.

Since this type of endocytosis leads to lysosomal degradation of the internalized material the siRNA must be prepared in such a way that it is stable in the lysosome, which is now solved by the novel modification pattern outlined above.

Likewise, it is understood that the targeting ligand shown in formula III conjugated to a nucleic acid such as siRNA as shown in formula IV can be replaced by other natural or chemically synthesized compounds (antagonists or agonists) displaying a high binding affinity to cell surface expressed receptors.

Example include folate as ligand for the folate receptor expressed on a variety of cancer cells (Ann. N.Y.

Acad.

Sci., 2009, 1175, 32) or PSMA binding molecules (Nature Biotech, 2006, 24, 1005; Mol Pharm, 2009, 6, 780).

The ligand for the ASGPR is attached via an amide bond to the nucleic acid.

The amide bond formation can be established with the aid of N-Hydroxy-succinimide (NHS) chemistry.

The ligand employed in the conjugation reaction is shown below (formula III).

For interaction with the ASGPR the 0-acetate groups on the sugars residues need to be removed as shown in (formula IV) for siRNA.

n 0 0 0.---Nr 0 0 0 0 0.0( 0 0 N __ N N ==-c-jo 0 0 Oµ 0 0 0 N o 0 0 0 CA 2976966 2017-08-21 -14o 0 z 0 0 0 N, 0 0 0 0 0 0 0 0.4N 0 si RNA (IV) In one embodiment of the invention, a conjugate of a compound of formula IV and an oligonucleotide is provided, wherein the oligonucleotide has the following modification pattern:

the antisense strand with the modification pattern 5'-(w)- (Z1)- (Z2)- (Z3) na -3' and a sense strand with the modification pattern 5'- (Z3) ns -, wherein the substituents as well as the variables n, and n, have the meaning given above.

Said conjugate is also referred to as GalNAc palmitoyl conjugate.

Preferably, said GalNAc palmitoyl conjugate are co-administered with a delivery polymer.

Suitable delivery polymers are described above.

It was found that for these modification patterns cleavable linkers proofed to be advantageous compared to stably linked small molecule ligands.

Possible cleavable linkers are a di-peptide motif as exemplified in scheme 1 or a cleavable RNA-linker comprising 2'-OH containing nucleotides.

The cleavable RNA-linker is especially useful in connection with the siRNAs having the novel modification pattern (fully 2'-modified siRNA) described above.

In principle a nuclease cleavage site can be introduced by 3'- or 5'-overhangs containing at least one 2'-OH nucleotide at either the sense or the antisense strand.

The final active siRNA species is generated by intracellular nuclease processing.

Also, the use of defined cleavage sites implemented by 2'-OH nucleotides within the base paired region is possible.

This can be done using at least one 2'-OH nucleotide complementary to the opposite strand or by introduction of CA 2976966 2017-08-21 -15either at least one mismatched 2'-OH nucleotide or a hairpin/bulge containing at least one 2'-OH nucleotide.

In contrast to other cleavable linker chemistries the use of defined cleavage sites by introduction of 2'-OH nucleotides lead to a more versatile conjugation approach.

By introducing selective cleavage sites on one or on both strands of the siRNA either at the 3' and / or the 5'-end or within the duplex structure, multiple conjugation is possible.

Accordingly, in one embodiment, a conjugate of a small molecule and an oligonucleotide is provided wherein a) the small molecule comprises a nucleotide linker comprising 1-10 preferably 1-5, most preferably 1-3 2' OH-nucleotides;

b) the oligonucleotide has the following modification pattern: the antisense strand with the modification pattern 5'-(w)- (Z1)- (Z2)- (Z3) na -3' and a sense strand with the modification pattern 5' - (Z3) ns -, wherein the substituents as well as the variables na and ns have the meaning given above; and c) the oligonucleotide is covalently attached via the nucleotide linker to the small molecule.

The nucleotide linker is cleaved e.g. in the endosome by intracellular nucleases such as DNAse II after internalization of the conjugate, thus releasing the siRNA.

Preferably, said conjugate is co-administered with a delivery polymer.

Suitable delivery polymers are described above.

In another embodiment of the invention a compound of formula (V) is provided.

This compound comprises a cholesterol moiety, and a nucleotide linker comprising 110 preferably 1-5, most preferably 1-3 2'0H-nucleotides.This nucleotide linker is useful for covalently attaching an oligonucleotide such as a siRNA to the compound of formula (V).

Preferably, said oligonucleotide has the novel modification pattern outlined above.

Hence in another embodiment a conjugate of a compound of formula (V) and an oligonucleotide is provided, wherein the oligonucleotide is covalently attached to the nucleotide linker of the compound of formula (V).

The nucleotide linker is cleaved by intracellular nucleases such as DNAse II after internalization of the conjugate of a compound of formula (V) and an oligonucleotide into the endosome, thus releasing the siRNA.

CA 2976966 2017-08-21 -16,õµ,, 0 OH N H 0 0 I OH 0, R = nucleotide linker (V) Preferably, said conjugate of a compound of formula (V) and an oligonucleotide is coadministered with a delivery polymer.

Suitable delivery polymers are described above.

In another embodiment, said delivery polymer and the conjugate of a compound of formula (V) and an oligonucleotide of the invention are not covatently attached and synthesized separately and may be supplied in separate containers or a single container.

Definitions The term "small molecule" as used herein, refers to organic or inorganic molecules either synthesized or found in nature, generally having a molecular weight less than 10,000 grams per mole, optionally less than 5,000 grams per mole, and optionally less than 2.000 grams per mole.

The term "peptide" as used herein refers to any polymer compound produced by amide bond formation between an .alpha.-carboxyl group of one D- or L-amino acid and an .alpha.amino group of another D- or L-amino acid.

The term "protein" as used herein refers to polypeptides of specific sequence of more than about 50 residues.

The term "di-peptide motif' as used herein refers to any motif comprising an amide bond formed by either the D- or L- alpha or beta amino group of a first amino acid with the alpha carboxyl group of a second D- or L-amino acid.

CA 2976966 2017-08-21 -17As used herein, the term "amino acid" refers to any molecule that contains both amine and carboxyl functional groups.

Thus the term "amino acid" refers to both natural, non-natural and synthetic amino acids.

Any natural amino acids used in the present invention are referred to herein by their common abbreviations.

The term "ligand" as used herein refers to a moiety that is capable of covalently or otherwise chemically binding a nucleic acid.

The term "ligand" in the context of the invention is preferably a compound of formula (I) or (la) covalently attached to a nucleic acid.

The term "nucleic acid" as used herein means an oligomer or polymer composed of nucleotides, e.g., deoxyribonucleotides or ribonucleotides, or compounds produced synthetically (e.g., PNA as described in U.S.

Pat.

No. 5,948,902 and the references cited therein) which can hybridize with naturally occurring nucleic acids in a sequence specific manner analogous to that of two naturally occurring nucleic acids, e.g., can participate in WatsonCrick base pairing interactions.

Non-naturally occurring nucleic acids are oligomers or polymers which contain nucleobase sequences which do not occur in nature, or species which contain functional equivalents of naturally occurring nucleobases, sugars, or inter-sugar linkages, like peptide nucleic acids (PNA), threose nucleic acids (TNA), locked nucleic acids (LNA), or glycerol nucleic acids (GNA).

This term includes oligomers that contain the naturally occurring nucleic acid nucleobases adenine (A), guanine (G), thymine (T), cytosine (C) and uracil (U), as well as oligomers that contain base analogs or modified nucleobases.

Nucleic acids can derive from a variety of natural sources such as viral, bacterial and eukaryotic DNAs and RNAs.

Other nucleic acids can be derived from synthetic sources, and include any of the multiple oligonucleotides that are being manufactured for use as research reagents, diagnostic agents or potential and definite therapeutic agents.

The term includes oligomers comprising of a single strand nucleic acid or a double strand nucleic acid.

The term "2'-modified" as used herein refers to a 13-D-ribonucleoside or fi-Dribonucleotide comprising of naturally occurring nucleobases having the 2'-OH group replaced by H, F, O-CH3 or other substituents known in the art.

The term "2'-OH ¨nucleotide" as used herein refers to 13-D-ribonucleotide comprising of naturally occurring nucleobases having a 2'-OH group.

CA 2976966 2017-08-21 -18The term "5'-phosphate" as used herein refers to the formula -0-P(--.0)(OH)OH.

In another aspect the phosphate is modified such that one of the 0 or OH groups is replaced by S and termed herein as "5 -phosphothioate" The term "phosphorothioate" as used herein refers to an internucleotide linkage in which one of the non-bridging oxygens is replaced by sulfur.

The term "delivery polyrner" as used herein refers to polymers suitable for functional delivery of a nucleic acid.

In the context of the present invention the delivery polymer is either covalently attached to or coadministered with the biologically substance conjugated to the compounds described herein and mediates endosomal escape after internalization into the cell and uptake into the endosome.

The term "polymer" in this context means any compound that is made up of two or more monomeric units covalently bonded to each other, where the monomeric units may be the same or different, such that the polymer may be a homopolymer or a heteropolyrner.

Representative polymers include peptides, polysaccharides, nucleic acids and the like, where the polymers may be naturally occurring or synthetic.

Non-limiting examples of delivery polymers are for example reviewed in INTERNATIONAL JOURNAL OF PHARMACEUTICAL RESEARCH AND DEVELOPMENT, October - 2010 / Volume - 2 / Issue - 8 / Article No -2.

Non-limiting examples of delivery polymers useful for delivery of nucleic acids are disclosed in EP applications 10165502.5 and 10191030.5, PCT publication WO 2008/0022309, and references cited herein, As used herein, "pharmaceutical composition" includes the conjugates of the invention, a pharmaceutical carrier or diluent and any other media or agent necessary for formulation.

As used herein, "pharmaceutical carrier" includes any and all solvents, dispersion media, coatings, antibacterial and antifungal agents, isotonic and absorption delaying agents, and the like that are physiologically compatible.

Preferably, the carrier is suitable for intravenous, intramuscular, subcutaneous, parenteral, spinal or epidermal administration (e.g. by injection or infusion).

A conjugate of the present invention can be administered by a variety of methods known in the art.

As will be appreciated by the skilled artisan, the route and/or mode of administration will vary depending upon the desired results.

To administer a conjugate of the invention by CA 2976966 2017-08-21 -19certain routes of administration, it may be necessary to coat the conjugate with, or co-administer the conjugate with, a material to prevent its inactivation.

For example, the conjugate may be administered to a subject in an appropriate carrier or a diluent.

Pharmaceutically acceptable diluents include saline and aqueous buffer solutions.

Pharmaceutical carriers include sterile aqueous solutions or dispersions and sterile powders for the extemporaneous preparation of sterile injectable solutions or dispersion.

The use of such media and agents for pharmaceutically active substances is known in the art.

The phrases "parenteral administration" and "administered parenterally" as used herein means modes of administration other than enteral and topical administration, usually by injection, and includes, without limitation, intravenous, intramuscular, intraarterial, intrathecal, intracapsular, intraorbital, intracardiac, intradermal, intraperitoneal, transtracheal, subcutaneous, subcuticular, intraarticular, subcapsular, subarachnoid, intraspinal, epidural and intrasternal injection and infusion.

These carriers may also contain adjuvants such as preservatives, wetting agents, emulsifying agents and dispersing agents.

Prevention of presence of microorganisms may be ensured both by sterilization procedures, supra, and by the inclusion of various antibacterial and antifungal agents, for example, paraben, chlorobutanol, phenol, sorbic acid, and the like.

It may also be desirable to include isotonic agents, such as sugars, sodium chloride, and the like into the compositions.

In addition, prolonged absorption of thc injectable pharmaceutical form may be brought about by the inclusion of agents which delay absorption such as aluminum monostearate and gelatin.

Regardless of the route of administration selected, the conjugates of the present invention, which may be used in a suitable hydrated form, and/or the pharmaceutical compositions of the present invention, are formulated into pharmaceutically acceptable dosage forms by conventional methods known to those of skill in the art.

Actual dosage levels of the active ingredients in the pharmaceutical compositions of the present invention may be varied so as to obtain an amount of the active ingredient which is effective to achieve the desired therapeutic response for a particular patient, composition, and mode of administration, without being toxic to the patient.

The selected dosage level will depend upon a variety of pharmacokinetic factors including the activity of the particular compositions of the present invention employed, the route of administration, the time of administration, the rate CA 2976966 2017-08-21 -20of excretion of the particular compound being employed, the duration of the treatment, other drugs, compounds and/or materials used in combination with the particular compositions employed, the age, sex, weight, condition, general health and prior medical history of the patient being treated, and like factors well known in the medical arts.

The pharmaceutical composition must be sterile and fluid to the extent that the composition is deliverable by syringe.

In addition to water, the carrier preferably is an isotonic buffered saline solution.

Proper fluidity can be maintained, for example, by use of coating such as lecithin, by maintenance of required particle size in the case of dispersion and by use of surfactants.

In many cases, it is preferable to include isotonic agents, for example, sugars, polyalcohols such as mannitol or sorbitol, and sodium chloride in the composition.

Short description of the figures Figure 1 shows Co-Administration of siRNA-conjugates comprising the compounds of formula (I) or (Ia) and a delivery polymer in vivo.

Figure 2 shows Co-Administration of siRNA-conjugates comprising the compounds of formula (I) or (Ia) and a delivery polymer in vivo.

Figure 3 shows Co-Administration of siRNA-conjugates comprising the compounds of formula (I) or (Ia) and a delivery polymer in vivo.

Figure 4 shows Co-Administration of siRNA-conjugates comprising the compounds of formula (I) or (Ia) and a delivery polymer in vivo.

Figure 5a shows antisense strand mediated gene silencing with fully 2'modified siRNAs. COS7 cells were cotransfected with EGFP-directed siRNAs at 3 nM and psiCHECK2-AT.

The knockdown activity of the siRNAs was assessed by measuring renilla versus firefly luciferase CA 2976966 2017-08-21 -21activity from the reporter construct. siRNAs were sorted by knockdown activity of unmodified (2-19-2) reference siRNAs.

Figure 5b shows sense strand mediated gene silencing with fully 2'-modified siRNAs. COS7 cells were cotransfected with EGFP-directed siRNAs at 3 nM and psiCHECK2-ST.

The knockdown activity of the siRNAs was assessed by measuring luciferase expression from the reporter construct. siRNAs were sorted by knockdown activity of unmodified (219-2) reference siRNAs.

Figure 6a shows reduction of serum FVII activity in non-human primates upon intravenous injection of various 2'-modified siRNAs covalently attached to a delivery polymer.

Figure 6b shows the development of the prothrombin time in non-human primates upon treatment with 2'-modified siRNAs covalently conjugated to a delivery polymer.

Examples The following examples are meant as references example solely, in order to illustrate the synthesis of the compounds for use in delivery of nucleic acids.

They are not intended to form part of the invention.

Example 1 Step 1: 3-(3S,8S,9S,10R,13R,14S,17R)-174(R)-1,5-Dimethyl-hexyl)-10,13dimethyl2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahydro-1H-cyc1opentakiphenanthren3-y1oxy]propylamine Chiral 001H N CA 2976966 2017-08-21 -22The title amine was prepared from its nitrile precursor according to a literature protocol [ Lollo et al , W02001/070415].

Step 2: N-{ 3-[(3S ,8S ,9S ,10R,13R,14S ,17R)-174(R)-1,5-Dimethylhexyl)-10,13-dimethyl2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahydro-1H-cyclopenta[a]phenanthren3-yloxy] propyl I -succinamic acid Chiral H ìO H100 0 0 In a 2 L round-bottomed flask, 3-((3S,8S,9S,10R,13R,14S ,17R)-10,13-dimethy117-((R)-6methylheptan-2-y1)-2,3 ,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahydro-1Hcyclopenta[a]phenanthren-3-yloxy)propan-1-amine (21.15 g, 47.7 mmol, Eq: 1.00) and Huenig's base (12.3 g, 16.6 ml, 95.3 mmol, Eq: 2.00) were combined with AcOEt (845 ml) to give a colorless solution.

Dihydrofuran-2,5-dione (4.77 g, 47.7 mmol, Eq: 1.00) in THF (42 ml) was added and the reaction mixture was stirred at ambient temperature over night => white suspension.

All volatiles were removed i. v., the residue dissolved in CH2C12, the organic layer washed with NH4C1 and brine, dried over Na2SO4, and evaporated to dryness.

The crude product was dissolved in CH3CN / H20 and lyophilized to yield 29.8 g of the title compound as fluffy powder.

MS (ISP): (M-H) 542.5.

Step 3: N1-(34(3S ,8S ,9S ,10R,13R,14S ,17R)- 10,13-Dimethyl- 17-((R)-6methylheptan-2-y1)2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahydro-1H-cyclopenta[a]phenanthren3yloxy)propy1)-N4-((S)-1-((S)-1-(4-(hydroxymethyl)phenylamino)-6-((4methoxyphenyl)diphenylmethylamino)-1-oxohexan-2-ylamino)-3-(4-nitropheny1)-1oxopropan2-y1)succinamide CA 2976966 2017-08-21 -23Chiral 40 i= 0 = H 0 0 14111 0 0 00 In a 10 mL round-bottomed flask, the above preparcd 4-(343S,8S,9SJOR,13R,14S,17R)-10,13dimethy1-174(R)-6-methylhept an-2-yI)-2,3,4,7,8,9,10,11,12,13,14,15,16,17tetradec ahydro- 1Hcyc1opentala]phenanthren-3-y1oxy)propy1amino)-4-oxobutanoic acid (106 mg, 184 iamol, Eq:

1.00), (S)-2-4S )-2-amino-3-(4-nitrophenyl)propanamido)-N-(4(hydroxymethyl)pheny1)-6-((4methoxyphenyl)diphenylmethylamino)hexanamide (132 mg, 184 mot, Eq: 1.00), HOAt (25.0 mg, 184 mot, Eq: 1.00) and EDC hydrochloride (35.3 mg, 184 mot, Eq: 1.00) were mixed together in CH2C12 (1.8 ml) to give a yellow solution.

Huenig'sBase (47.5 mg, 64.2 [11, 368 mot, Eq: 2.00) was added and the reaction stirred at ambient temperature over night. TLC indicated the consumption of starting material.

All volatiles were removed i.

V. and the crude product purified by flash chromatography SiO2 / 7% Me0H / 0.1% NEt3 in CH2C12 to produce 128 mg of the title compound as light yellow solid.

MS: expected mass: 1240.7552, found mass: 1240. 7518.

Step 4:

CA 2976966 2017-08-21 -24Chiral H 0 it 0 H N o ¨ 0 0 0 0 In a 10 mL round-bottomed flask, the above prepared N1(34(3S,85,9S,10R,13R,14S,17R)10,13-dimethy1-174(R)-6-methylheptan-2-y1)-2,3,4,7,8,9,10,11,12,13,14,15,16,17 tetradecah ydro-1H-c yc lopenta[a]phenanthren-3-ylox y)prop y1)-N4-((S)-1 ((S)-1-(4(hydroxymethyl)phenylamino)-6-((4-methoxyphenyl)diphenylmethylamino)-1oxohexan-2ylamino)-3-(4-nitropheny1)-1-oxopropan-2-ypsuccinamide (126 mg, 101 pmol, Eq:

1.00) and Huenig's base (39.3 mg, 53.2 pl, 304 pmol, Eq: 3.00) were combined with CH2C12 (1.4 ml) and DMF (1.0 ml) to give a yellow suspension; bis(4-nitrophenyl) carbonate (46.3 mg, 152 mol, Eq:

1.50) was added and the reaction allowed to proceed over night.

The mixture was poured onto crashed ice, extracted 2 x with AcOEt, washed with H20, dried over Na2SO4, and evaporated to dryness.

After trituration with ¨10 ml of diethyl ether, 99 mg of the title product was obtained as an off-white solid.

MS: expected mass: 1405.7614, found mass: 1405.7518.

The necessary dipeptide building block for step 3 was prepared as follows:

Step a: (S)-2{(S)-2-(9H-Fluoren-9-ylmethoxycarbonylamino)-3-(4-nitro-pheny1)propionylamino] -6- { [(4-methoxy-phen y1)-diphenyl-methyl] -amino I -hexanoic acid CA 2976966 2017-08-21 -25NO, Chiral 0 0 0 .411 0 0 0¨ ln a 25 mL round-bottomed flask, (S)-2-amino-6-((4methoxyphenyl)diphenylmethylamino)hexanoic acid (Bioconjugate Chem. 2002, 13, 855-869, 968 mg, 2.31 mmol, Eq: 1.00) was dissolved in CH2C12 (20 ml) to give a light yellow solution.

Huenigts base (897 mg, 1.21 ml, 6.94 mmol, Eq: 3.00) and trimethylchlorosilane (528 mg, 621 I, 4.86 mmol, Eq:

2.10) were added and the reaction mixture was stirred for 15 min.

In a second 50 mL round-bottomed flask,(S)-2-(((9H-fluoren-9yOmethoxy)carbonylamino)-3(4-nitrophenyl)propanoic acid (1 g, 2.31 mmol, Eq: 1.00) was dissolved in DMF (20 ml) to give a colorless solution.Huenig's base (359 mg, 485 R1, 2.78 mmol, Eq: 1.20) and TPTU [125700-712] (687 mg, 2.31 mmol, Eq: 1.00) were added and the reaction mixture was stirred for 20.

The solution from the first flask containing the corresponding silyl ester monosilylamine was added and the reaction was stirred for another 3 hours.

The mixture was poured onto crashed ice / NH4C1, extracted 2 x with AcOEt, washed with 1120 and brine, dried over Na2SO4, and evaporated to dryness.

Flash chromatography Si02 / 10% Me0H / 0.1% NEt3 in CH2C12 afforded 1.38 g of the title compound as brownish foam.

MS (ISP): (M+H) 833.5, (M+Na) 855.4.

Step b: [(S)-14(S)-1-(4-Hydroxymethyl-phenylcarbamoy1)-5-{ [(4-methoxy-phenyl)diphenylmethyThaminol-penty1carbamoy1)-2-(4-nitro-pheny1 )-ethyl[ -carbamic acid 9H-fluoren-9ylmethyl ester CA 2976966 2017-08-21 -26NO, Chiral O el ))¨N 0 0 Nr,7\7 114M MIT 0 0¨ O In a 250 mL pear-shaped flask, the above synthesized (S)-2-((S)-2-(((9Hfluoren-9yl)methoxy)carbonylamino)-3-(4-nitrophenyl)propanamido)-6-((4methoxyphenyl)diphenylmethylamino)hexanoic acid (1.38 g, 1.66 mmol, Eq: 1.00), (4aminophenyl)methanol (204 mg, 1.66 mmol, Eq: 1.00), HOAt (226 mg, 1.66 mmol, Eq: 1.00) and EDC hydrochloride (318 mg, 1.66 mmol, Eq: 1.00) were dissolved in CH2C12 (16.6 ml) to give a yellow solution.

Huenig's base (428 mg, 579 1, 3.31 mmol, Eq: 2.00) was added and the reaction allowed to proceed over night.

The mixture was poured onto crashed ice / NH4C1 (pH ¨7), extracted 2 x with AcOEt, washed with H20, dried over Na2SO4, and evaporated to dryness.

The crude product was triturated with diethyl ether (1 x 50 mL); the resultant solid was filtered off and dryed to yield 1.214 g of the title compound as light-brown solid.

MS (ISP): (M+H) 938.7.

Step c: (S )-2- )-2-Amino-3-(4-nitro-pheny1)-propionylamino]-6-1[(4methoxy-pheny1)diphcnyl-mcthyl]-aminol-hexanoic acid (4-hydroxymethyl-pheny1)-amide NO, Chiral 0 NI N 0 O CA 2976966 2017-08-21 -27In a 50 mL round-bottomed flask, the above prepared [(S)-1-((S)-1-(4hydroxymethylphenylcarbamoy1)-5- [(4-methoxy-phenyl)-diphenyl-methyl] -amino } pentylcarbamoy1)-2-(4nitro-pheny1)-ethyl]-carbamic acid 9H-fluoren-9-ylmethyl ester (1.214 g, 1.29 mmol, Eq: 1.001) was combined with THF (19 ml) to give a brown solution.

At 0 , diethylamine (1.77 g, 2.49 ml, 24.2 mmol, Eq: 18.70) was added.

The reaction was stirred at ambient temperature for 3h when MS indicated the disappearance of the starting material.

All volatiles were evaporated i. V.;

ensuing flash chromatography Si02 / 0.1% NEt3 in CH2C12 => 10% Me0H / 0.1% NEt3 in CH2C12, followed by a second flash chromatography Si02 / 5% Me0H / 0.1% NEt3 in CH2C12 afforded 502 mg of the title compound as light brown foam.

MS: expected mass: 715.337, found mass: 715.3362.

Example 2 0-Benzyl-N[4-({34(3beta)-cholest-5-en-3-yloxy]propyl lamino)-4-oxobutanoyll-LtyrosylN-6-4(4-methoxyphenyl)(diphenyl)methyl]-N44-({ [(4nitrophenoxy)carbonyl]oxy}methyl)pheny11-L-1ysinamide Chiral 0 40 *, 0 = H 0 N 0 == 0 ¨ 0 0 1411 0 '0 Was prepared in analogy to Example 1, but using in step 3 (S)-2-[(S)-2-amino-3(4-benzyloxypheny1)-propionylamino]-6- R4-methoxy-pheny1)-diphenyl-methyThamino I -hex anoic acid (4CA 2976966 2017-08-21 -28hydroxymethyl-pheny1)-amide instead of (S)-24(S)-2-amino-3-(4nitrophenyl)propanamido)-N(4-(hydroxymethyl)pheny1)-6-44-methoxyphenyl)diphenyl-methylaminolhexanamide as coupling partner. The former was prepared from (S )-2-(((9H-fluoren-9yl)methoxy)carbonylamino)-3-(4-(benzyloxy)phenyl)propanoic acid as described above in steps a] - c].

MS: expected mass: 1466.8182, found mass: 1466.8136.

Example 3 N14-(13-[(3Beta)-cholest-5-en-3-yloxy]propyll amino)-4-oxobutanoy1]-4-cyano-Lphenylalany1-1\1-6---[(4-methoxyphenyl)(diphenyl)methyl]-N-[4-(1 [(4 nitrophenoxy)carbonyl]oxy I meth yl)phen yl] -L-ly sinamide Chiral N 116 fa, H 1101 " ONH=rN 0 00 0 Nr/, 110/0 0¨ 0 140 -41+, 0 '0 Was prepared in analogy to Example 1, but using in step 3 (S)-24(S)-2-amino-3(4-cyanophenyl)-propionylamino]-6- { [(4-methoxy-phenyl)-diphenyl-methyl]aminolhexanoic acid (4hydroxymethyl-phenyl)-amide instead of (S)-24(S)-2-amino-3-(4-nitropheny1)propanamido)-N(4-(hydroxymethyl)pheny1)-6-((4-methoxyphenyl)diphenyl-methylamino)hexanamide as CA 2976966 2017-08-21 -29coupling partner.

The former was prepared from (S)-2-(((9H-fluoren-9yl)methoxy)carbonylamino)-3-(4-cyanophenyl)propanoic acid as described above in steps a] ¨ c].

MS: expected mass: 1385.7716, found mass: 1385.7696.

Example 4 3,4-Dichloro-N-1-4-({3-1(3beta)-cholest-5-en-3-yloxy]propyl}amino)-4oxobutanoy1]-Lphenylalanyl-N-6--[(4-methoxyphenyl)(diphenyl)methyl]-N-[4-({ [(4nitrophenoxy)carbonyl]oxy }methyl)pheny1]-L-lysinamide Chiral CI H.., CI 0 4.

H 00 0 0 0¨ '01 0 oo ,o Was prepared in analogy to Example 1, but using in step 3 (S)-2-[(S)-2-amino-3(3,4-dichlorophenyl)-propion ylami no] -6- { [(4-methox y-phen y1)-diphenyl-meth yl] -amino } -hex anoic acid (4hydroxymethyl-pheny1)-amide instead of (S )-2-((S )-2 -amino-3 -(4 nitropheny1)-propanamido)-N(4-(hydroxymethyl)pheny1)-6-((4-methoxyphenyl)diphenyl-methylamino)hexanamide as coupling partner.

The former was prepared from (S)-2-(((9H-fluoren-9yl)methoxy)carbonylamino)-3-(3,4-dichlorophenyl)propanoic acid as described above in steps a] ¨ c].

MS: expected mass: 1428.6984, found mass: 1428.695.

Example 5 CA 2976966 2017-08-21 -304-Chloro-N-[4-({ 3- [(3beta)-cholest-5-en-3-yloxy] propyl } amino)-4oxobutanoyll-Lphenylalany1-1\1-6---[(4-methoxyphenyl)(diphenyl)methyl]-N-[4-(1 [(4nitrophenoxy)carbonyl]oxy}methyl)pheny1]-L-lysinamide Chiral fik, Cl 0 '''ONHN 0 0 0 0¨ = 0--LO 101111 0 ' 0 Was prepared in analogy to Example 1, but using in step 3 (S)-2-((S)-2-amino-3(4chlorophenyl)propanamido)-N-(4-(hydroxymethyl)pheny1)-6-((4methoxyphenyl)diphenylmethylamino)hexanamide instead of (S)-24(S)-2-amino-3-(4-nitropheny1)propanamido)-N-(4(hydroxymethyl)pheny1)-6-((4-methoxyphenyl)diphenyl-methylamino)hexanamide as coupling partner.

The former was prepared from (S)-2-(((9H-fluoren-9-yl)methoxy)carbonylamino)-3-(4chlorophenyl)propanoic acid as described above in steps a] - c].

MS: expected mass: 1394.7373, found mass: 1394.7342.

Example 6 4-1[(2S)-2-{R2S)-2-[(4-{ [3-({ (3S,8S,9S,10R,13R,14S,17R)-10,13-dimethy1-17[(2R)-6 methylheptan-2-yl] -2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradec ah ydro-1Hcyclopenta[a]phenanthren-3-y1} oxy)propyl] amino } -4-oxobutanoyl)amino]-3(naphthalen-1yl)propano yl] amino } -6- { [(4-methoxyphenyl)(diphenyl)methyl] amino }hexanoyl]amino }benzyl 4-nitrophenyl carbonate (non-preferred name) CA 2976966 2017-08-21 -31_ H Chiral 1111111IP Fi 0 = 0 0 0 0-0 0 '0 Was prepared in analogy to Example 1, but using in step 3 (S)-24(S)-2-amino-3naphthalen-1yl-propionylamino)-6- [(4-methoxy-phenyl)-diphenyl-methyl] -amino } -hexanoic acid (4hydroxymethyl-phenyl)-amide instead of (S)-24(S)-2-amino-3-(4-nitropheny1)propanamido)-N(4-(h ydroxymeth yl)pheny1)-6-((4-methoxyphen yl)diphenyl-meth ylamino)hex anamide as coupling partner.

The former was prepared from (S)-2-(((9H-fluoren-9yl)methoxy)carbonylamino)-3-(naphthalen- 1 -yl)propanoic acid as described above in steps al - c].

MS: expected mass: 1410.792, found mass: 1410.7918.

Example 7 N-[4-({ 3 -R3Beta)-cho1es t-5-en-3-yloxy]propyl } amino)-4-oxobutanoy1}-4fluoro-Lphenylalanyl-N--6¨}(4-methoxypheny1)(dipheny1)methy1] -N-I4-({ }(4nitrophenoxy)carbonyl}oxy}methyl)phenyli-L-lysinamide CA 2976966 2017-08-21 -32Chiral H ie.

0 ,_, 00 0 0 0 0 ,o Was prepared in analogy to Example 1, but using in step 3 (S)-2-[(S)-2-amino-3(4-fluorophen yl) -propionylamino] -6- { [(4-methox y-pheny1)-diphenyl-methyl] -amino } -hex anoic acid (4hydroxymethyl-pheny1)-amide instead of (S)-24(S)-2-amino-3-(4-nitropheny1)propanamido)-N(4-(hydroxymethyl)pheny1)-6-((4-methoxyphenyl)diphenyl-methylamino)-hexanamide as coupling partner. The former was prepared from (S )-2-(((9H-fl uoren-9yl)metho x y)carbonylamino)-3 -(4 -fluorophen yl )propanoic acid as described above in steps a] ¨ c].

MS: expected mass: 1378.7669, found mass: 1378.7609.

Example 8 N-[4-( 3-[(3Beta)-cholest-5-en-3-yloxy]propyl } amino)-4-oxobutanoy1]-2-fluoroLphenylalanyl-N-6---[(4-methoxyphenyl)(diphenyOmethyl]-N-[44 [(4nitrophenoxy)carbonyl]oxy } methyl)pherty1]-L-lysinamide CA 2976966 2017-08-21 -33Chiral H -E FO 0 H 0111111 0 '"10N)HrN 0 0 0101 0 00 410 0 '0 Was prepared in analogy to Example 1, but using in step 3 (S)-2-[(S)-2-Amino-3(2-fluoropheny1)-propionylamino]-6- I [(4-methoxy-phenyl)-diphenyl-methyl]-amino)hexanoic acid (4hydroxymethyl-pheny1)-amide instead of (S)-2-((S)-2-amino-3-(4-nitropheny1)propanamido)-N(4-(hydroxymethyl)pheny1)-64(4-methoxyphenyl)diphenyl-methylamino)-hexanamide as coupling partner.

The former was prepared from (S)-2-(((9H-fluoren-9yl)methoxy)carbonylamino)-3-(2-fluorophenyepropanoic acid as described above in steps a] ¨ MS: expected mass: 1378.7669, found mass: 1378.7689.

Example 9 N-[4-( { 3 -R3Beta)-cholest-5-en-3-yloxy]propyl} amino)-4-oxobutanoy1]-3fluoro-Lphenylalanyl-N--6--[(4-methoxyphenyl)(diphenyl)methyl j-N-[4-({ [(4nitrophenoxy)carbonyl] oxy } methyl)phenyl] -L-lysinamide CA 2976966 2017-08-21 -34Chiral H 101 411, 4101 0 0 0 0 00 0 0 ,o Was prepared in analogy to Example 1, but using in step 3 (S)-2-[(S)-2-amino-3(3-fluoropheny1)-propionylamino] -6- { [(4-metho xy-pheny1)-diphenyl-meth yl] -amino } hex anoic acid (4hydroxymethyl-phenyl)-amide instead of (S)-24(S)-2-amino-3-(4-nitropheny1)propanamido)-N(4-(hydroxymethyl)pheny1)-644-methoxyphenyl)diphenyl-meth yl amino)-hex an amide as coupling partner.

The former was prepared from (S)-2-(((9H-fluoren-9yl)methoxy)carbonylamino)-3-(3-fluorophenyl)propanoic acid as described above in steps a] ¨ cl.

MS: expected mass: 1378.7669, found mass: 1378.7659.

Example 10 Step 1: N1-(3-((3S,8S,9S,10R,13R,14S ,17R)-10,13-dimethy1-174(R)-6methylheptan-2- y1)2,3,4,7,8,9,10,11,12,13 ,14,15,16,17-tetradec ahydro-1Hcyclopenta[a]phenanthren-3yloxy )prop y1)-N44(S)-1-(4-fluorophenyl)-4-(( S)-1 -(4(hydroxymethyl)phenylamino)-6-((4methoxyphen yl)diphenylmeth ylamino)-1 -oxohexan-2-ylamino)-4-oxobu tan-2- yl)s uccinamide CA 2976966 2017-08-21 -35Chiral H 0 "ON)HrN N 0 0 In a 10 mL round-bottomed flask, the above prepared 4-(3((3S,8S,9S,10R,13R,14S,17R)-10,13dimethy1-174(R)-6-methylheptan-2- y1)-2,3,4,7,8,9,10,11,12,13,14,15,16,17tetradecahydro-1Hcyclopentala]phenanthren-3-yloxy)propylamino)-4-oxobutanoic acid (109 mg, 188 ma Eq:

1.00), (S)-2-(S)-3-amino-4-(4-fluoro-phen y1)-butyrylamino] -6- {1(4-methoxypheny1)-diphenylmethyli-amino I -hexanoic acid (4-hydroxymethyl-phenyl)-amide (132 mg, 188 ilmol, Eq: 1.00), HOAt (25.6 mg, 188 mol, Eq: 1.00) and EDC hydrochloride (36.1 mg, 188 mol, Eq: 1.00) were mixed together in CH2C12 (2 ml) to give a yellow solution.

Huenig'sBase (48.7 mg, 64.1 I, 377 mol, Eq: 2.00) was added and the reaction stirred at ambient temperature over night. TLC indicated the consumption of starting material.

All volatiles were removed i.

V. and the crude product purified by flash chromatography Si02 / 5% Me0H / 0.1% NEt3 in CH2C12 to yield 197 mg of the title compound as off-white solid.

MS: expected mass: 1227.7763, found mass: 1227.7714.

Step 2: 44(S)-24(S)-3-(4-(3-43S,8S,9S,10R,13R,14S,17R)-10,13-dimethyl-17-((R)6methylheptan-2-y1)-2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahydro-1Hcyclopent a lalphenanthren-3 -yloxy)prop ylamino)-4-oxobu tanamido)-4-(4fluorophenyl)butanamido)-64(4-methoxyphenyl)diphenylmethylamino)hexanamido)benzyl 4nitrophenyl carbonate CA 2976966 2017-08-21 -36Chiral H = * 00 H 0 "ON)HrN 0 11110 0 0 0 '0 In a 10 mL round-bottomed flask, the above prepared N1-(343S,8S,9S,10R,13R,14S,17R)10,13-dimethy1-174(R)-6-methylheptan-2-y1)-2,3,4,7,8,9,10,11,12,13,14,15,16,17 tetradecahydro-114-cyclopentgalphenanthren-3- yloxy)propy1)-N4-((S )-1-(4-fl uoropheny1)-4((S )-1-(4-(hydroxymethyl)phenylamino)-6-((4-methoxyphenyl)diphenylmethyl amino)-1oxohex an-2-y1 amino)-4-oxobutan-2-yl)succinamide (196 mg, 160 mmol, Eq: 1.00) and Huenig's base (61.9 mg, 81.4 1, 479 [tmol, Eq: 3.00) were combined with CH2C12 (1.6 ml) and DMF (0.8 ml) to give a yellow suspension; bis(4-nitrophenyl) carbonate (72.8 mg, 239 pmol, Eq: 1.50) was added and the reaction allowed to proceed at ambient temperature over night.

The mixture was poured onto crashed ice / NH4C1 (pH ¨6), extracted 2 x with AcOEt, washed with H20 and brine, dried over Na2SO4, and evaporated to dryness.

After trituration with AcOEt / heptane one obtained 123 mg of the title compound as light yellow solid.

MS: expected mass: 1392.7825, found mass: 1392.7819.

The necessary dipeptidic building block for step 1 was prepared as follows:

Step a: (S)-2-{(S)-3-(9H-Fluoren-9-ylmethoxycarbonylamino)-4-(4-fluoropheny1)butyrylaminol -6- { {(4-methoxy-phenyl)-diphenyl-methyl{ -amino } -hexanoic acid CA 2976966 2017-08-21 -37Chiral , O * (30 F 1101 Ny.0 o 01/0 In a 25 mL round-bottomed flask, (S)-2-amino-6-((4methoxyphenyl)diphenylmethylamino)hexanoic acid (Bioconjugate Chem. 2002, 13, 855-869, 1040 mg, 2.48 mmol, Eq: 1.00) was dissolved in CH2C12 (12.5 ml) to give a pale yellow solution.

Huenig's base (961 mg, 1.27 ml, 7.44 mmol, Eq: 3.00) and trimethylchlorosilane (566 mg, 621 i1, 5.21 mmol, Eq: 2.10) were added and the reaction mixture was stirred at ambient temperature for 20 min.

In a second 50 mL round-bottomed flask, (S)-3-(((9H-fluoren-9yemethoxy)carbonyl-amino)-4(4-fluorophenyl)butanoic acid (1040 mg, 2.48 mmol, Eq: 1.00) was dissolved in DMF (12.5 ml) to give a colorless solution.

Huenig's base (385 mg, 506 iI, 2.98 mmol, Eq:

1.20) and TPTU [125700-71-2] (737 mg, 2.48 mmol, Eq: 1.00) were added and the reaction mixture was stirred for 15 min.

The solution from the first flask containing the corresponding silyl ester monosilylamine was added and the reaction was stirred for another 3 hours at ambient temperature.

The mixture was poured onto crashed ice / NH4C1, extracted 2 x with AcOEt, washed with H20 and brine, dried over Na2SO4, and evaporated to dryness.

Flash chromatography Si02 / 5% Me0H / 0.1% NEt3 in CH2C12 afforded 2.10 g of the title compound as yellow foam.

MS (ISP): (M-1-1-1) 820.6.

CA 2976966 2017-08-21 -38Step b: (4-Fluoro-pheny1)- 1- R(S)-1-(4-hydro xymeth yl-phenylc arbamo y1)-5-11(4methoxy-pheny1)-diphenyl-methyl] -aminol-pentylc arb amoy1)-methyl} -ethyl } carbamic acid 9Hfluoren-9-ylmethyl ester Chiral 0 N Ny0 0 0 In a 250 mL pear-shaped flask, the above synthesized (S)-2-(4-fluoro-pheny1)-1[((S)-1-(4hydroxymethyl-phenylcarbamoy1)-5-1 [(4-methoxy-phenyl)-diphenyl-methyl] aminolpent ylcarbamoy1)-methyl]ethyl}-carbamic acid 9H-fluoren-9-ylmethyl ester (2.10 g, 2.56 mmol, Eq: 1.00), (4-aminophenyl)methanol (315 mg, 2.55 mmol, Eq: 1.00), HOAt (349 mg, 2.56 mmol, Eq: 1.00) and EDC hydrochloride (491 mg, 2.56 mmol, Eq: 1.00) were dissolved in CH2C12 (12.5 m1).

Huenig's base (662 mg, 871 111, 5.21 mmol, Eq: 2.00) was added and the reaction allowed to proceed over night.

The mixture was poured onto crashed ice / NH4C1 (pH ¨7), extracted 2 x with AcOEt, washed with H20 and brine, dried over Na2SO4, and evaporated to dryness.

The crude product was triturated with diethyl ether (1 x 50 ml); the resultant solid was filtered off and dryed to yield 0.796 g of the title compound as light-brown solid.

MS (ISP): (M+H) 925.6.

Step c: (S)-2-[(S)-3-Amino-4-(4-fluoro-pheny1)-butyrylamino]-6-1[(4methoxy-pheny1)diphenyl-methyl]-aminol-hexanoic acid (4-hydroxymethyl-phenyl)-amide CA 2976966 2017-08-21 -39Chiral * 1110 N o In a 50 mL round-bottomcd flask, the above prepared {(S)-2-(4-fluoro-pheny1)-1R(S)-1-(4hydroxymethyl-phen ylcarbamo y1)-5-11 (4-methoxy-phenyl)-diphen yl-methyl] amino)pentylcarbamoy1)-methyThethyl}-carbamic acid 9H-fluoren-9-ylmethyl ester (793 mg, 857 mol, Eq: 1.001) was combined with THF (12 ml) to give a brownish solution.

At 0 , diethylamine (1.13 g, 1.59 ml, 15.4 mmol, Eq: 18) was added.

The reaction was stirred at ambient temperature over night.

The mixture was poured onto crashed ice / NFI4C1 (pH ¨7), extracted 2 x with AcOEt, washed with H20 and brine, dried over Na2SO4, and evaporated to dryness.

Flash chromatography Si02 / 10% Me0H / 0.1% NEt3 in CH2C12 yielded 500 mg of the title compound as off-white solid.

MS: expected mass: 702.3581, found mass: 702.3578.

Example 11 4-((S )-2-((S )-344434(3 S ,8S,9S ,10R,13R,14S ,17R)-10,13-dimethy1-174(R)-6methylheptan-2y1)-2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahydro-1Hcyclopenta[a]phenanthren-3yloxy)propylamino)-4-oxobutanamido)-4-phenylbutanamido)-6-((4methoxyphenyl)diphenylmethylamino)hexanamido)benzyl 4-nitrophenyl carbonate CA 2976966 2017-08-21 -40Chiral * * H 0 H O NO 0 o 0-.N,.0 Was prepared in analogy to Example 10, but using in step 1 (S)-24(S)-3-amino-4 phenylbutanamido)-N-(4-(hydroxymethyl)pheny1)-64(4-methoxyphenyl)diphenylmethylamino)hexanamide instead of (S)-2-RS)-3-amino-4-(4-fluoro-pheny1)butyrylamino]-6{ }(4-methoxy-phenyl)-diphenyl-methyl]-aminol-hexanoic acid (4-hydroxymethylphenyeamide as coupling partner.

The former was prepared from (S)-3-(((9H-fluoren-9yOmethoxy)carbonylamino)-4-phenylbutanoic acid as described above in steps a] c].

MS: expected mass: 1374.792, found mass: 1374.7877.

Example 12 4-(11\1-2-4(3S)-4-(4-chlorophenyl)-3- [4-({34(3beta)-cholest-5-en-3yloxy]propyl } amino)-4oxobutanoyl] amino } bu tanoyl] -N-6¨[(4-methoxyphenyl) (diphenyemeth yl] -Llysyl lamino)benzyl 4-nitrophenyl carbonate CA 2976966 2017-08-21 -4 1 Chiral CI * * oe 141111 0 H 0 "oN)Hr N N7C-r-d N 0 0 0 NO2 Was prepared in analogy to example 10, but using in step 1 (S)-2-((S)-3-amino4-(4chlorophenyl)butanamido)-N-(4-(hydroxymethyl)pheny1)-6-((4-methoxypheny1)diphenylmethylamino)hexanamide instead of (S)-2-[(S)-3-amino-4-(4-fluoropheny1)butyrylamino1-6-1 [(4-methoxy-phenyl)-diphenyl-meth yl] -amino } -hex anoic acid (4hydroxymethyl-pheny1)-amide as coupling partner.

The former was prepared from (S)-3-(((9Hfluoren-9-yl)methoxy)carbonylamino)-4-(4-chloropheny1)-butanoic acidas described above in steps a] ¨ c].

MS (ISP): (M+H) 1409.9.

Example 13 N44-({3-[(3Beta)-cholest-5-en-3-yloxy]propyllamino)-4-oxobutanoy1]-0-methyl-LtyrosylN-6--[(4-methoxyphenyl)(diphenyl)methyl]-N-[4-({ [(4nitrophenoxy)carbonyl]oxylmethyl)pheny1]-L-lysinamide CA 2976966 2017-08-21 -42Chiral oI H !FP * H 0 N 0 0110 0 0 00 0-0 110 4.sf:

0 0 Was prepared in analogy to Example 1, but using in step 3 (S)-2,-((S)-2-amino3-(4methoxyphenyl)propanamido)-N-(4-(hydroxymethyl)pheny1)-64(4-methoxypheny1)diphenylmethylamino)hexanamide instead of (S)-24(S)-2-amino-3-(4-nitropheny1)propanamido)-N-(4-(hydroxymethyl)phenyl)-6-((4-methoxyphenyediphenylmethylamino)hexanamide as coupling partner.

The former was prepared from (S)-2(((9Hfluoren-9-yl)methoxy)carbonylamino)-3-(4-methoxyphenyl)propanoic acid as described above in steps a] - c] of example 1.

MS (ISP): (M+H) 1391.9.

Example 14 N-[4-( 3-[(3Beta)-cho1est-5-en-3-yloxy]propyllamino)-4-oxobutanoy1]-Dphenylalanyl-N-6-[(4-methoxyphenyl)(diphenyl)methyll -1\144-(1 [(4-nitrophenoxy)carbonyl]oxy }methyl)phenyllD-lysinamide CA 2976966 2017-08-21 -439* Chiral H 0 I 0 00 N, 0 mp o o H 0 0 N 0 0 N * 0 411 Was prepared in analogy to example 1, but using in step 3 (R)-2-((R)-2-amino-3phenylpropanamido)-N-(4-(hydroxymethyepheny1)-6-((4methoxyphenyl)diphenylmethylamino) hexanamide instead of (S)-24(S)-2-amino-3-(4-nitropheny1)-propanamido)-N-(4(hydroxymethyl)pheny1)-6-((4-methoxyphenyl)diphenyl-methylamino)hexanamide as coupling partner.

This building block was synthesized from (R)-2-(((9H-fluoren-9yl)methoxy)carbonylamino)-6-aminohexanoic acid and (R)-2-amino-6-((4methoxypheny1)diphenylmethylamino)hexanoic acid (see Bioconjugate Chem. 2002, 13, 885-869) as described above in steps a] - c].

MS: expected mass: 1360.7763, found mass: 1360.7774.

Example 15 4-(1N-2---[(3 S)-3 - 1 [4-( 3- [(3Beta)-cholest-5-en-3-yloxy]propyl }amino)-4oxobutanoyl] amino } - 4-(4-cyanophenyl)butanoy1]-N-6--[(4-methoxyphenyl)(diphenyl)methyll-Llysyl}amino)benzyl 4-nitrophenyl CA 2976966 2017-08-21 -44carbonate Chiral H.., = 00 411 H 0 0 0111111 N.f7-1 N 0 0 41111 -1 04\ ':

0 Was prepared in analogy to example 10, but using in step 1 (S)-2-((S)-3-amino4-(4cyanophenyl)butanamido)-N-(4-(hydroxymethyl)pheny1)-6-((4methoxyphenyl)diphenylmethylamino)hexanamide instead of (S)-2-[(S)-3-amino-4-(4-fluoro-phenyebutyrylamino]-6{ [(4-methoxy-pheny1)-diphenyl-methyl]-amino I -hexanoic acid (4-hydroxymethyl-pheny1)amide as coupling partner.

The former was prepared from (S)-3-(((9H-fluoren-9yl)methoxy)carbonylamino)-4-(4-cyanophenyebutanoic acid as described above in steps a] ¨ c] .

MS: expected mass: 1399.7872, found mass: 1399.7857.

Example 16 N-[4-( 3-[(3beta)-cholest-5-en-3-yloxy]propyllamino)-4-oxobutanoyll-Lphenylalanyl-N-6-[(4-methoxyphenyl)(diphenyl)methyl]-N44-(1[(4nitrophenoxy)carbonyl]oxy}methyl)phenyl]L-lysinamide CA 2976966 2017-08-21 -45Q.

Chiral H WI RIP 0 W 0 0 H Njt 11111 alb N 0 r--0 Step 1:

(S)-24(S)-2-Amino-3-phenyl-propionylamino)-6-{ [(4-methoxy-phenye-diphenylmethy11aminol-hexanoic acid (4-hydroxymethyl-phenyl)-amide --0 Chiral * )1 = 0 = N NO(N 1.1 0 The building block (S)-2-((S)-2-amino-3-phenyl-propionylamino)-6- [(4-methoxypheny1)diphenyl-methyli-amino 1 -hexanoie acid (4-hydroxymethyl-pheny1)-amide was prepared in analogy to the procedure described in Bioconjugate Chem., Vol. 13, No.4, 2002, 855-869 MS (ISP): (M+H) 671.5 Step 2:

N1-(34(3S,8S,9SJOR,13R,14S,17R)-10,13-dimethyl-174(R)-6-methylheptan-2-y1)2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahydro-1H-cyclopenta[a]phenanthren3yloxy)propy1)-N44(S)-1-((S)-1-(4-(hydroxymethyl)phenylamino)-6-((4methoxyphenyl)diphenylmethylamino)-1-oxohexan-2-ylamino)-1-oxo-3-phenylpropan2yl)succinamide CA 2976966 2017-08-21 -46Chiral H Ow 0 H 1010 N 01111 0 0 11.

TPTU (125700-71-21(233 mg, 784 mol, Eq: 1.00) was added to a solution of N.-P S R3S ,8S ,9S ,10R,13R,14S,17R)-174(R)-1,5-dimethyl-hexyl)-10.13-dimethyl2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahydro-1H-cyclopenta[a]phenanthren3-yloxy} propy1}-succinamic acid (see example 1, step 2) (426 mg, 0.784 mmol, Eq: 1.00) and Huenig' s base (304 mg, 411111, 2.35 mmol, Eq: 3) in DMF (10 m1).

After 3 minutes (S)-2((S)-2-amino-3phenyl-propionylamino)-6-( [(4-methoxy-pheny1)-diphenyl-methyl]-amino}hexanoic acid (4hydroxymethyl-pheny1)-amide (step 1) was added TLC at t = 1 h showed the reaction was complete.

The solvent was removed under reduced pressure.

The remaining residue was taken up in ethyl acetate and extracted with NaHCO3 half sat. solution (1 X), potassium hydrogen phthalate solution 0.05M (2 X), water (1 X) and brine (1 X).

The organic extract was dried over MgSO4 and concentrated under reduced pressure.

The crude material was purified by flash chromatography to obtain the titled product (682 mg,513, iimol) as a light brown solid.

MS (ISP): (M+H) 1196.8 Step 3:

Hiinig's base (465 mg, 629 i1, 3.6 mmol, Eq: 6) was added to a solution of the previous alcohol (718 mg, 600 pmol, Eq: 1.00) and bis(4-nitrophenyl) carbonate (548 mg, 1.8 mmol, Eq: 3) in THF (20 ml).

The yellow solution was stirred overnight at room temperature.

The solvent was removed under reduced pressure.

The remaining residue was triturated with diethyl ether.

The CA 2976966 2017-08-21 -47solid was collected by filtration, washed with ether and dried under reduced pressure to obtain the title compound (800 mg, 529 mop as a light brown solid.

MS (ISP): (M+H) 1361.9 Example 17 Step 1 (S)-2-RS )-2-(9H-Fluoren-9- ylmethox ye arbonylamino)-3 -phenylpropion ylamino] hexanoic acid o o = 0 N yO 0 do* Commercially available L-Fmoc-Phe-OSu (0.969 g, 2.00 mmol, Eq: 1.00) was suspended in a 1:1 v/v mixture of 1,2-dimethoxyethane and water (17 ml) and treated at 0 C with L-norleucine (0.275 g, 2.10 mmoll, Eq: 1.05) and NaHCO3 (0.185 g, 2.20 mmol, Eq: 1.10).

The cooling bath was removed and the reaction allowed to proceed at ambient temperature for 14 h.

The mixture was poured onto crashed ice / citric acid (pH -3), extracted 2x with ethyl acetate, washed with H20 and brine, dried over Na2SO4, and evaporated to dryness.

Flash chromatography Si02 / AcOEt yielded 0.870 mg of the title compound as white solid.

MS (ISP): (M+H) 501.2.

Step 2: {(S)-11(S)-1-(4-Hydroxymethyl-phenylcarbamoy1)-pentylcarbamoy1.1-2phenyl-ethyllcarbamic acid 9H-fluoren-9-ylmethyl ester CA 2976966 2017-08-21 -480 N N 11101 N y0 0 *** In a pear-shaped flask, the above synthesized (S)-2-{(S)-2-(9H-fluoren-9ylmethoxycarbonylamino)-3-phenyl-propionylamino[-hexanoic acid (10.72 g, 21 mmol, Eq:

1.00), (4aminophenyl)methanol (2.717 g, 22 mmol, Eq: 1.03), and 2-ethoxy-1ethoxycarbony1-1,2dihydroquinoline (EEDQ) (7.994 g, 32 mmol, Eq: 1.50) were dissolved in CH2C12 (320 ml) and stirred over night under a balloon of Ar.

The mixture was poured onto crashed ice / NR4C1, extracted 2 x with AcOEt, washed with H20, dried over Na2SO4, and the volume reduced to ¨ 300 ml.

The precipitate was filtered off and dryed to give 5.25 g of the title compound as light brown solid.

MS (ISP): (M+H ) 606.3.

Step 3: (S)-2-((S)-2-Amino-3-phenyl-propionylamino)-hexanoic acid (4hydroxymethylpheny1)-amide o 0 ip N N N In a round-bottomed flask, the above prepared { (S)-1-[(S )-1-(4hydroxymethylphenyl carbamoye-pentyl carbamoyl] -2-phenyl-ethyl -carbamic acid 9H-fluoren-9ylmethyl ester (4.738 g, 7.822 mmol, Eq: 1.0) was dissolved in CH2C12 (28 ml).

At 0 , diethylamine (28 ml, 19.80 g, 271 mmol, Eq: 35) was added and the reaction mixture stirred at ambient temperature CA 2976966 2017-08-21 -49over night.

All volatiles were evaporated i. V.; ensuing flash chromatography Si02 / CH2C12 / 10% Me0H, followed by crystallization from AcOEt, yielded 2.116 g of the title compound as light brown crystals.

MS (ISP): (M+H ) 384.2.

Step 4 N1-(3-43S,8S,9S,10R,13R,14S,17R)-10,13-dimethy1-174(R)-6-methylheptan-2-y1)2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahydro-1H-cyclopentala}phenanthren3yloxy)propy1)-N44(S)-1-((S)-1-(4-(hydroxymethyl)phenylamino)-1-oxohexan-2ylamino)-1oxo-3-phenylpropan-2-y1)succinamide H 40 0 , 40*0 0 0 was prepared therewith in analogy to example 16 step 2 MS (ISP): (M+H) 909.7 (M+Na) 931.8.

Step 5 N-[4-( f 3-R3beta)-cholest-5-en-3-yloxy}propyl } amino)-4-oxobutanoyll-Lphenylalanyl-N44( [(4-nitrophenoxy)carbonyl]oxy}methyl)pheny1FL-norleucinamide 9 Chiral Hain 01. 40 N: 0 HCji r N,D(N 40 0 0 CA 2976966 2017-08-21 -50was prepared therewith in analogy to example 16 step 3 MS expected mass: 1073.6453, found mass 1073.642 Example 18 N44-(134(3beta)-cholest-5-en-3-yloxy]propyl }amino)-4-oxobutanoy1]-L-alanylN44-({ [(4nitrophenoxy)carbonyl]oxy}methyl)phenyl]glycinamide 9 Chiral Ahrio010 4P H ,0,-,NJHrN-rNJN 0 Step 1:

Addition of FM0C-4-Aminobenzylalcohol to the 2-chlorotrityl resin =0I OÖO0 Pol N 0 *SI 2-Chlorotrityl chloride resin (Novabiochem 01-64-0114, 100-200 mesh), 1%DVB (18 g, 21.6 mmol, Eq: 1.00) was swollen in DCM/DMF=1/1 (300 mL) for ten minutes.

The resin was drained and a solution of FM0C-4-aminobenzyl alcohol (14.9 g, 43.2 mmol, Eq:

2) and pyridine (6.83 g, 6.99 ml, 86.4 mmol, Eq: 4) in DCM/DMF=1/1 (300 mL) was added.

The mixture was shaken over night.

The resin was drained and capped with a solution of 10% Hilnig's Base in methanol (300 mL).

The resin was washed with DMF and DCM and dried over night with HV to obtain CA 2976966 2017-08-21 -5121.7 g resin.

Determination of the load resulted in 0.41 mmoL/g.

Step 2:

N1-(34(3S,8S,9S,10R,13R,14S ,17R)-10,13-dimethy1-174(R)-6-methylheptan-2-y1)2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradec ahydro-1H-c yclopenta[a] phenanthren-3yloxy)propy1)-N4-((S )-1-(2-(4-(hydroxymethyl)phenylamino)-2-oxoethylamino)-1oxopropan-2yl)succinamide Chiral H ;

H 0 N 0 0 The resin from step 1(1 g, 410 mol, Eq: 1.00) was prewashed with DMF (2 X) and treated with piperidine/DMF=1/4 (10 mL) for 5 and 10 minutes.

The resin was washed alternately with DMF and WA (3 X 10 mL). A solution of Fmoc-Gly-OH (488 mg, 1.64 mmol, Eq: 4), TPTU (487 mg, 1.64 mmol, Eq: 4) and Huenig's base (636 mg, 859 I, 4.92 mmol, Eq: 12) in DMF (10 mL) was stirred for 5 minutes and then shaken with the resin for one hour.

The resin was washed alternately with DMF and isopropyl alcohol (3X).

The following Fmoc cleavages and subsequent couplings of Fmoc-Ala-OH (511 mg, 1.64 mmol, Eq: 4) and N-{ 3-R3S ,8S,9S ,10R,13R,14S ,17R)- 17-((R)-1,5-Dimethyl-hexyl) 10,13-dimethyl2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahydro-1H-cyclopenta[a]phenanthren3-yloxy}propyll-succinamic acid (example 1, step 2) (892 mg, 1.64 mmol, Eq: 4) were performed accordingly.

The dried peptide resin was stirred for about 2 X 30 min in TFA 1%/DCM (2 X 20 mL).

The reaction mixture was filtered and the resin was washed with DCM.

The filtrates were pooled and the solvents evaporated under vacuum.

The crude material was triturated with diethyl ether (2 x ).

After purification by flash chromatography, the product (84 mg, 97.3 mop was obtained as a white solid.

CA 2976966 2017-08-21 -52MS expected mass:776.5452, found mass 776.5455 Step 3:

The above prepared alcohol N1-(3-43S,8S,9S,10R,13R,14S,17R)-10,13-dimethy1174(R)-6methylheptan-2-y1)-2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradec ah ydro-1Hcyclopenta1alphenanthren-3-yloxy)propy1)-N4-((S)-1-(2-(4(hydroxymethyl)phenylamino)-2oxoethylamino)-1-oxopropan-2-y1)succinamide [R05545270] (70 mg, 90.1 mol, Eq:

1.00) and bis(4-nitrophenyl) carbonate (137 mg, 450 mol, Eq: 5) under Argon at room temperature were dissolved in DMF (4 ml) and treated with Huenig's base (34.9 mg, 47.2 1.11, 270 lamol, Eq: 3).

and the mixture was alloiwed to react overnight.

The solvent was removed in vacuo.

The resulting solid was triturated with diethylether.

The solid was collected by filtration and washed with diethyl ether.

The product was dried in vacuo to obtain the title compound (84 mg, 80.2 mol) as a light brown solid.

MS expected mass:941.5514, found mass 941.5518 Example 19 N44-(13-[(3beta)-cholest-5-en-3-yloxy]propyl I amino)-4-oxobutanoyli-L-leucylN44-( { [(4nitrophenoxy)carbonylloxy}methyl)phenyl] -L-methioninamide : Chiral H o oo = 0 H - A 0 Step 1:

Addition of FM0C-4-aminobenzylalcohol to the 2-chlorotrityl resin CA 2976966 2017-08-21 -53,c' =ÖO0 N 2() .00 Poi 1101 was prepared in analogy to example 18, step 1 Step 2 N1-(34(3S,8S,9S,10R,13R,145,17R)-10,13-dimethy1-174(R)-6-methylheptan-2-y1)2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahydro-1H-cyc1openta[a]phenanthren3yloxy)propy1)-N44(S)-1-((S)-1-(4-(hydroxymethyl)phenylamino)-4-(methylthio)-1oxobutan-2ylamino)-4-methyl-1-oxopentan-2-y1)succinamide Chiral H 0 la 0 H No e 0 0 was prepared in analogy to example 18, step 2, using Finoc-Met-OH (609 mg, 1.64 mmol, Eq: 4) and Fmoc-Leu-OH (580 mg, 1.64 mmol, Eq: 4) as amino acids.

The product (208 mg, 210 timol) was obtained as a light yellow solid.

MS (ISP): (M+H) 893.6183 Step 3 was prepared in analogy to example 18, step 3.

After purification on silica gel, the title compound (161 mg, 137 pmol) was obtained as light brown solid.

MS expected mass:1057.6174, found mass 1057.6184 CA 2976966 2017-08-21 -54Example 20 N44-({ 3-1(3beta)-cholest-5-en-3-yloxy]propyl I amino)-4-oxobutanoyll-L-leucylN-1-44-(1[(4nitrophenoxy)carbonyl]oxy}methyl)pheny1]-L-aspartamide Q Chiral N H .1. N. 100 0 Step 1:

Addition of FM0C-4-aminobenzylalcohol to the 2-chlorotrityl resin * o 40, Pol 111" N 0 ir 10 was performed in analogy to example 18, step 1 Step 2 (S)-24(S)-2-(4-(34(3S,8S,9S,10R,13R,14S,17R)-10,13-dimethy1-17-((R)-6methylheptan-2-y1)15 2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahydro-1Hcyclopenta[a]phenanthren-3yloxy)propylamino)-4-oxobutanamido)-4-methylpentanamido)-N1-(4(hydroxymethyl)phenyl)succinamide CA 2976966 2017-08-21 -55Chiral H 0 0 H 0-0 0 Was prepared in analogy to example 18, step 2, using Fmoc-Asn-OH (581 mg, 1.64 mmol, Eq:

4) and Fmoc-Leu-OH (580 mg, 1.64 mmol, Eq: 4) as amino acids.

The product (87 mg, 89.4 mol) was obtained as a light yellow solid.

MS expected mass:875.6136, found mass 875.6133 Step 3 The titled compound was prepared in analogy to example 18, step 3.

After purification on silica gel (87 mg, 89.4 mol) the title compound was obtained as light brown solid.

MS expected mass:1040.6198, found mass 1040.6188 Example 21 N-[4-( 31(3beta)-cholest-5-en-3-yloxylpropyllamino)-4-oxobutanoyli-L-alanyl-N1--[4-(1[(4nitrophenoxy)carbonyl]oxy lmethyl)phenylj-L-aspartamide 9 Chiral H 111101P gh 0 SOO '0-Nr1\1(NN 0 0 ---isr0 Step 1:

Addition of FM0C-4-aminobenzylalcohol to the 2-chlorotrityl resin CA 2976966 2017-08-21 -56Cl =o Pol 411 N was performed in analogy to example 18, step 1 Step 2 (S)-24(S)-2-(4-(34(3S,8S,9S,10R,13R,14S,17R)-10,13-dimethy1-174(R)-6methylheptan-2-y1)2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahydro-1H-eyelopenta[a]phenanthren3yloxy)propylamino)-4-oxobutanamido)propanamido)-N1-(4(hydroxymethyl)phenyesuccinamide Chiral H 0 0 H 0 Was prepared in analogy to example 18, step 2, using Fmoc-Asn-OH (581 mg, 1.64 mmol, Eq:

4) and Fmoc-Ala-OH (511 mg, 1.64 mmol, Eq: 4) as amino acids.

The product (140 mg, 159 mol) was obtained as light yellow solid.

MS (ISP): (M+H) 834.8 (M+Na) 856.7 CA 2976966 2017-08-21 -57Step 3 The title compound was prepared in analogy to example 18, step 3.

After purification on silica gel (169 mg, 152 umol) it was obtained as light brown solid.

MS expected mass:998.5729, found mass 998.5739 Example 22 N-2-44-(13-[(3beta)-cholest-5-en-3-yloxy]propyll amino)-4-oxobutanoy1J-Lasparaginyl-N44({[(4-nitrophenoxy)carbonyl]oxy}methyl)phenyllglycinamide 9, Chiral H 010 = 111-11111111 9 Step 1:

Addition of FM0C-4-Aminobenzylalcohol to the 2-chlorotrityl resin 41 1.1 Pol .41 N 0 Op 1110 Was performed in analogy to example 18, step 1 Step 2 (S)-2-(4-(34(3S.8S,9S,10R,13R,14S,17R)-10,13-dimethy1-174(R)-6-methylheptan-2y1)2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahydro-1H-cyclopenta[a]phenanthren3yloxy)propylamino)-4-oxobutanamido)-N1-(2-(4-(hydroxymethyl)phenylamino)-2 oxoethyl)succinamide CA 2976966 2017-08-21 -58Chiral 0 0 0 H 0 0 Was prepared in analogy to example 18, step 2, using Fmoc-Gly-OH (488 mg, 1.64 mmol, Eq: 4) and Fmoc-Asn-OH (581 mg, 1.64 mmol, Eq: 4) as amino acids.

The product (140 mg, 162 [tmol) was obtained as white solid.

MS expected mass:819.551, found mass 819.5503 Step 3: The title compound was obtained in analogy to example 18, step 3 (176 mg, 161 mop as light brown solid.

MS expected mass:984.5572, found mass 984.5489 Example 23 N-[ 4-( { 34(3beta)-cholest-5-en-3-yloxy]propyl} amino)-4-oxobutanoyl] -Lphenylalanyl-N- [4({1(4-nitrophenoxy)carbonylioxyl methyl)phenyl]glycinamide 9 Chiral H AiLdh 0 NjA01 0 0 Step 1:

Addition of FM0C-4-aminobenzylalcohol to the 2-chlorotrityl resin CA 2976966 2017-08-21 -59=c' 0 Nio 100 Pol Was performed in analogy to example 18, step 1 Step 2:

N1-(34(3S,8S,9S,10R,13R,14S,17R)-10,13-dimethy1-174(R)-6-methylheptan-2-y1)2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahydro-1H-cyclopentatalphenanthren3yloxy)propy1)-N4-((S)-1-(2-(4-(hydroxymethyl)phenylamino)-2-oxoethylamino)-1oxo-3phenylpropan-2-y1)succinamide Chiral H 00 0 0 H 0 0 Was prepared in analogy to example 18, step 2, using Fmoc-Gly-OH (488 mg, 1.64 mmol, Eq: 4) and Fmoc-Phe-OH (635 mg, 1.64 mmol, Eq: 4) as amino acids.

The product (259 mg, 288 vinol) was obtained as white solid.

MS expected mass:852.5765, found mass 852.5754 Step 3 The title compound was obtained in analogy to example 18, step 3. (280 mg, 247 mop as light brown solid.

CA 2976966 2017-08-21 -60MS expected mass:1017.5827, found mass 1017.5775 Example 24 N44-(131(3beta)-cholest-5-en-3-yloxy]propyl I amino)-4-oxobutanoy11-L-leucylN44-({[(4nitrophenoxy)carbonyl]oxylmethyl)phenyliglycinamide 9 Chiral afh, H 010 'WI 0 00 0 0 Step 1:

Addition of FM0C-4-aminobenzylalcohol to the 2-chlorotrityl resin =a 401 .0 Ni) 0 Or Pol Was performed in analogy to example 18, step 1 Step 2 N1-(3-43S,8S,9S,10R,13R,14S,17R)-10,13-dimethy1-17-((R)-6-methylheptan-2-y1)2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahydro-lH-cyclopenta[a]phenanthren3yloxy)propyl)-N4-((S)-1-(2-(4-(hydroxymethyl)phenylamino)-2-oxoethylamino)-4methyl-1oxopentan-2-y1)succinamide CA 2976966 2017-08-21 -61Chral H.. 0 H 0 0 Was prepared in analogy to example 18, step 2, using Fmoc-Gly-OH (488 mg, 1.64 mmol, Eq: 4) and Fmoc-Leu-OH (580 mg, 1.64 mmol, Eq: 4) as amino acids.

The product (240 mg, 278 p.mol) was obtained as a light yellow solid.

MS expected mass:818.5921, found mass 818.5921 Step 3 The title compound was prepared in analogy to example 18, step 3.

After purification on silica gel, it (194 mg, 177 pmol) was obtained as light yellow solid.

MS expected mass:983.5983 found mass 983.6004 Example 25 N-[4-( 3-[(3beta)-cholest-5-en-3-yloxy]propyl amino)-4-oxobutanoyll-L-leucylN{4-( { [(4 nitrophenoxy)carbonyl]oxylmethyl)phenylj-L-phenylalaninamide 9 Chiral H 010 = 0 H N 0 0 Step 1:

CA 2976966 2017-08-21 -62Addition of FM0C-4-aminobenzylalcohol to the 2-chlorotrityl resin =CI 0 ei I N 0 Opel Pol Was performed in analogy to example 18, step 1 Step 2 N1-(34(3S,85,9SJOR,13R,14S,17R)-10,13-dimethyl-17-((R)-6-methylheptan-2-y1)2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahydro-1H-cyclopenta[a]phenanthren3yloxy)propy1)-N4-((S)-14(S)-1-(4-(hydroxymethyl)phenylamino)-1-oxo-3phenylpropan-2 ylamino)-4-methyl-l-oxopentan-2-yl)succinamide Chral 0-0 0 0 0 Was prepared in analogy to example 18, step 2, using Fmoc-Phe-OH (635 mg, 1.64 mmol, Eq: 4) and Fmoc-Leu-OH (580 mg, 1.64 mmol, Eq: 4) as amino acids.

The product (153 mg, 151 mol) was obtained as light yellow solid.

MS expected mass: 908.6391 found mass 908.637 Step 3:

CA 2976966 2017-08-21 -63The title compound was prepared in analogy to example 18, step 3.

After purification on silica gel, it (117 mg, 98 limol) was obtained as white solid.

MS expected mass:1073.6453 found mass 1073.646 Example 26 N-(4-({3-[(3beta)-cholest-5-en-3-yloxy]propyl}amino)-4-oxobutanoy1}-Lphenylalanyl-N-{4({ [(4-nitrophenoxy)carbonyl]oxy}methyl)pheny1}-L-phenylalaninamide 9, Chral N. 40 I.= yL,,, 00 0 0 Step 1:

Addition of FM0C-4-aminobenzylalcohol to the 2-chlorotrityl resin 111 CI 0 el N 0 Op.

Po I Was performed in analogy to example 18, step 1 Step 2 N1-(3-((3S,8S.9S,10R,13R,14S,17R)-10,13-dimethy1-174(R)-6-methylheptan-2-y1)2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahydro-1H-cyclopenta[a]phenanthren3yloxy)propyl)-N4-((S)-1-((S)-1-(4-(hydroxymethyl)phenylamino)-1-oxo-3phenylpropan-2ylamino)-1-oxo-3-phenylpropan-2-y1)succinamide CA 2976966 2017-08-21 -64Chiral H g 0 NN ap 0 H N 0 Was prepared in analogy to example 18, step 2, with Fmoc-Phe-OH (635 mg, 1.64 mmol, Eq: 4) as amino acid.

The product (240 mg, 204 pmol) was obtained as light yellow solid.

MS expected mass:942.6234 found mass 942.6218 Step 3:

The title compound was prepared analogously to example 18, step 3.

After purification on silica gel, it (190 mg, 154 p mol) was obtained as white solid.

MS expected mass:1107.6296 found mass 1107.6287 Example 27 N-[4-( t 3-[(3beta)-cholest-5-en-3-yloxy]propyl} amino)-4-oxobutanoyll-Lleucyl-N-[44 ( [(4 nitrophenoxy)carbonyl]oxy I methyl)phenyli-L-leucinamide 9.

Chiral N.

H 1 0 010 0 0 H 0O, 0 Step 1:

CA 2976966 2017-08-21 -65Addition of FM0C-4-aminobenzylalcohol to the 2-chlorotrityl resin =c' 0 OONÄOle) Pol Was performed analogously to example 18, step 1 Step 2 N1-(34(3S,8S,9S,10R,13R,14S,17R)-10,13-dimethy1-174(R)-6-methylheptan-2-y1)2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahydro-1H-cyclopenta[a]phenanthren3yloxy)propyl)-N4-((S)-1-((S)-1-(4-(hydroxymethyl)phenylamino)-4-methyl-1oxopentan-2ylamino)-4-methyl-1-oxopentan-2-y1)succinamide Chiral H N 110 w/iciN24N 0 ====.õ--o Was prepared in analogy to example 18, step 2, with Fmoc-Leu-OH (1.59 g, 4.5 mmol, Eq: 3) as amino acid.

The product (254 mg, 284 mop was obtained as white solid MS expected mass:874.6547 found mass 874.6527 Step 3 The title compound was prepared in analogy to example 18, step 3.

After purification on silica gel it was obtained as white solid (178 mg, 168 mop.

CA 2976966 2017-08-21 -66MS expected mass:1039.6609 found mass 1039.6588 Example 28 1\144-(13-[(3beta)-cholest-5-en-3-yloxy]propyllamino)-4-oxobutanoy1FL-alanylN44-({ [(4nitrophenoxy)carbonylloxy}methyl)pheny1]-L-alaninamide Chiral H Alimh 770-0 N)(N,AN 411 0 s 8 Step 1 {(S)-1-RS)-1-(4-Hydroxymethyl-phenylcarbamoy1)-ethylcarbamoy1}-ethyll-carbamic acid 9Hfluoren-9-ylmethyl ester Chiral 11010 OIN'ITNA N N 110 0 0 41 A solution of Fmoc-Ala-Ala-OH (1 g, 2.61 mmol, Eq: 1.00) and (4aminophenyl)methanol (483 mg, 3.92 mmol, Eq: 1.5) in THF (20 ml) was treated with EEDQ (970 mg, 3.92 mmol, Eq: 1.5).

The solution was stirred over night at room temperature.

The mixture was diluted with 10% 2propanol/ethyl acetate (100 mL) and the solution was washed with KHSO4 5%/K2SO4 10% (2 X), water (1X) and brine (1X), dried over MgSO4 and evaporated in vacuo.

The residue was sonicated in diethyl ether for several minutes and the solid was collected by filtration to obtain the product (1.27 g, 1.2 mmol) as light brown solid.

CA 2976966 2017-08-21 -67MS (ISP): (M+H) 488.3 Step 2:

(S)-2-Amino-N-(S)-1-(4-hydroxymethyl-phenylcarbamoy1)-ethyl]-propionamide N 0 oChiral - N 0 The compound was prepared in analogy to example 1 step c to obtain the product (245 mg, 877 mmol) as light yellow solid.

MS (ISP): (M+H) 266.3, (M+Na) 288.2 (2M+H) 531.3 Step 3:

N1-(34(3S,8S,9SJOR,13R,14S,17R)-10,13-dimethyl-17-((R)-6-methylheptan-2-y1)2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahydro-1H-cyclopenta[a]phenanthren3yloxy)propy1)-N4-((S)-1-((S)-1-(4-(hydroxymethyl)phenylamino)-1-oxopropan-2ylamino)-1oxopropan-2-y1)succinamide Chiral H 40 H 0 0 N 0 The compound was prepared in analogy to example 16 step 2 (165 mg, 198 mol) as light brown solid.

MS expected mass: 790.5608, found mass 790.5587 Step 4 CA 2976966 2017-08-21 -68The title compound was prepared in analogy to example 18, step 3.

After purification on silica gel it was obtained as white solid (99 mg, 98.4 mop.

MS expected mass: 955.567, found mass 955.5651 Example 29 N-44-(13-[ (3beta)-eholest-5-en-3-yloxy]propyl amino)-4-oxobutanoyll-Lisoleucyl-N-1-44(11(4-nitrophenoxy)carbonylloxy} meth yl)phen yl] -L-aspartamide 9, Chiral H Hl ele N0 0 Fl 0 FNJN Step 1 (S)-21(2S,3S)-2-(9H-Fluoren-9-ylmethoxycarbonylamino)-3-methyl-pentanoylamino] succinamic acid Chiral 0 0 111õ, 0 N rrl\1 NA 0 VAL 0 N 0 2-Chlorotrityl chloride resin (5 g, 7.5 mmol, Eq: 1.00) was swollen in DCM and then treated with a solution of Fmoc-Asn(Trt)-OH (8.95 g, 15.0 mmol.

Eq: 2) and Huenig's base (3.88 g, 5.1 ml, 30.0 mmol, Eq: 4) in DCM overnight.

The resin was washed with DCM and capped with a solution of 10% Huenig's base in methanol.

Coupling of Fmoc-Ile-OH (5.3 g, 15.0 mmol, Eq: 2) with TPTU (4.46 g, 15.0 mmol, Eq: 2) and Huenig's base (3.88 g, 5.1 ml, 30.0 mmol, Eq: 4) according to standard solid phase peptide synthesis.

The product was cleaved from the resin with CA 2976966 2017-08-21 -69a cocktail of TFA/Water/triisopropylsilane (95/2.5/2.5 v/v/v) for two hours at room temperature.

The resin was filtered and the filtrate was concentrated under reduced pressure to a small volume.

After trituration with diethyl ether, the product was filtered and dried in vacuum to obtain the product (2.85 g, 5.79 mmol) as white solid.

MS expected mass: 467.2056, found mass 467.2056 Step 2 1(1S,2S)-142-Carbamoy1-14(S)-4-hydroxymethyl-phenylcarbamoy1)-ethylcarbamoy112methyl-butyll-carbamic acid 9H-fluoren-9-ylmethyl ester Chiral 0 0 0 *Ai 0 N71)1-1\1N)LN Via/ 0 N 0 The compound was prepared in analogy to example 28 step 1 (620 mg, 336 pmol) as light yellow solid.

Step 3 (S)-2-((2S,3S)-2-Amino-3-methyl-pentanoylamino)-N*1*-(4-hydroxymethy1-pheny1)succinamide Chiral 0 0 0 The compound was prepared in analogy to example 1 step c (100 mg, 228 pmol) as light yellow solid.

CA 2976966 2017-08-21 -70Step 4 (S)-2-((2S,3S)-2-(4-(3-((3S,8S,9S,10R,13R,14S,17R)-10,13-dimethy1-17-((R)-6methylheptan2-y1)-2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahydro-1Hcyclopenta[a]phenanthren-3yloxy)propylamino)-4-oxobutanamido)-3-methylpentanamido)-N1-(4 (hydroxymethyl)phenyl)succinamide Chiral H ;

0 [2 0 0 H H NisAN 0 0 =IrN 0 The compound was prepared in analogy to example 16 step 2 (89 mg, 91.4 mmol) as light yellow solid.

Step 5 The compound from the previous step was reacted to the title compound analogously to example 18, step 3.

After purification on silica gel, it (42 mg, 36.3 1,1mo1) was obtained as a light brown solid.

MS expected mass: 1040.6198, found mass 1040.6177 Example 30 N44-({3-[(3beta)-cholest-5-en-3-yloxy]propyllamino)-4-oxobutanoy1FLphenylalanyl-N-6-R4-methoxyphenyl)(diphenyl)methyl]-N44-(1[(4nitrophenoxy)carbonyl]oxylmethyl)phenyllD-lysinamide 9, Chiral H jt), 40 N 0 0 H 0-0 0 N 4111 CA 2976966 2017-08-21 -71The compound was prepared in analogy to example 16 step 1, starting with FmocD-Lys(Boc)OH, (158 mg, 116 pmol) as light brown solid.

MS (ISP): (M+H ) 1362.8 (M+Na) 1383.8 Example 31 N- 15-[(3beta)-cholest-5-en-3-yloxy]-4,15-dioxo-8,11-dioxa-5,14diazapentadecan-1-oy1l-Lphenylalanyl-N-6--[(4-methoxyphenyl)(diphenyl)methyli-N-[4-({ [(4nitrophenoxy)carbonyl]oxylmethyl)phenyl] -L-lysinamide (i? Chiral H I 0I0 N 40 =-0 N,JN 40 (:\)1 0 0 ON \\O N * * 0 The title compound was prepared analogous to example 16 using a cholesterololigo-PEG derivative in step 2 of the synthesis.

MS (ISP): (M+H) 1479.8 The necessary cholesterol-PEG intermediate N-[2-(2-{ 2[(35,85,9S,10R,13R,14S,17R)-17-((R)1,5-Dimethyl-hexyl)-10,13-dimethy1-2,3,4,7,8,9,10,11,12,13,14,15,16,17tetradecahydro-1Hcyclopenta[a]phenanthren-3-yloxycarbonylaminoj-ethoxyl-ethoxy)-ethyllsuccinamic acid for step 2 was prepared as follows:

Step a: {242-(2-Amino-ethoxy)-ethoxyi-ethyll-carbamic acid (3S,8S,9S,10R,13R,14S,17R)-17((R)-1,5-dimethyl-hexyl)-10,13-dimethy1-2,3,4,7,8,9,10,11,12,13,14,15,16,17tetradecahydro 1H-cyclopenta[a]phenantlu-en-3-y1 ester CA 2976966 2017-08-21 -72Chiral H H IP* 0 -===== 0 A solution of cholesteryl chloroformate (1 g, 2.23 mmol) in 25 mL dichloromethane was added dropwise under stirring to a solution of 2,2'-(ethylenedioxy)bis-(ethylamine) (495 mg, 3.34 mmol) in 75 mL dichloromethane.

The reaction was stirred overnight at room temperature.

The reaction was diluted with dichloromethane and extracted with water.

The organic extract was dried over anhydrous MgSO4 dihydrate, filtered and evaporated.

After purification on aminomodified silica gel (eluent: MeC12 -> MeC12/Me0H=975:25 v/v) the product (615 mg) was obtained as a white, waxy solid.

MS (ISP): (M+H ) 561.5 Step b: N42-(24 2-[(3S,8S,9S,10R,13R,14S ,17R)- 174(R)-1,5-Dimethyl-hexyl)10,13-dimethyl2,3,4,7,8,9,10,11,12,13,14,15 ,16,17-tetradecahydro- 1Hcyclopenta[a]phenanthren-3yloxycarbonylamino]-ethoxy -ethoxy)-ethy1]-succinamic acid Chiral H 0 0 H 1110111111 )L/-( 0 N 0 CA 2976966 2017-08-21 -73The amine from step a (480 mg, 0.856 mmol) and triethylamine (0.13 mL, 0.94 mmol) were dissolved in 5 mL dichloromethane.

After adding succinic anhydride (90 mg, 0.9 mmol) the solution was stirred overnight at room temperature. TLC check showed still some starting material.

More succinic anhydride (20 mg, 0.2 mmol) was added.

Aftcr stirring the reaction for another 3 hours at room temperature, it was diluted with dichloromethane and washed with a 5%KHSO4/ 10%K2SO4 mixture.

The organic extract was dried over anhydrous MgSO4dihydrate, filtered and evaporated in vacuo to obtain the desired acid (490 mg, 0.667 mmol).

MS (ISP): (M+H ) 661.5 Example 32 N- { 30-[(3beta)-cho1est-5 -en- 3-yloxy]-4,30-dioxo-8,11,14,17,20,23 ,26heptaoxa-5 ,29diaz atriacontan-l-o y1}-L-phenylalanyl-N-6-4(4methoxyphenyl)(diphenyl)methyl] -N-[4-( { [(4nitrophenoxy)carbonyl]oxy }methyl)pheny1]-L-Iysinamide Ch4 o on o 'PO t) The title compound was prepared analogous to example 16 using a cholesterol PEG derivative in step 2 of the synthesis.

MS (ISP): (M+H) 1699.9 The necessary cholesterol-PEG intermediate 1-[(3beta)-cholest-5-en-3-yloxy[1,27-dioxo5,8,11,14,17,20,23-heptaoxa-2,26-diazatriacontan-30-oic acid for step 2 was prepared as follows:

Step a: tert-butyl [25-({(3S,8S,9S,10R,13R,14S,17R)-10,13-dimethy1-17-[(2R)-6methylheptan2-y1]-2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahydro-1Hcyclopenta[a]phenanthren-3y1 oxy)-25-oxo-3,6,9,12,15,18,21-heptaoxa-24- azapentacos-1- yl] carbamate CA 2976966 2017-08-21 -74H, NO H 00111111''', 0 0 O N 0 Cholesteryl chloroformate (476 mg, 1.06 mmol) and triethylamine (155 uL, 1.113 mmol) were dissolved in 5 mL dichloromethane.

Then a solution of alpha-amino-omega-bocaminoocta(ethylene glycol) (497 mg, 1.06 mmol) dissolved in 1 mL dichloromethane was added.

The solution was stirred over night at room temperature and diluted with dichloromethane and extracted with a KHSO4 5%/K2SO4 10% aquieous mixture.

The organic extract was dried over anhydrous MgSO4, filtered and evaporated in vacuo.

After purification on silica gel (eluent:

MeC12/Me0H=975:25 -> 95:5 v/v) the product (530 mg, 0.571 mmol) was obtained as a colorless oil.

MS (ISP): (M+NH4 ) 898.7 Step b: 1-[(3beta)-cholest-5-en-3-yloxy]-1,27-dioxo-5,8,11,14,17,20,23heptaoxa-2,26diazatriacontan-30-oic acid Chiral Ole FL:' H .0 0 0 0 The previous Boc derivative (450 mg, 0.511 mmol) was dissolved in HC1 4M in dioxane (10.2 mL, 40.9 mmol).

The solution was stirred at room temperature for 40 min.

The solvent was CA 2976966 2017-08-21 -75removed in vacuo and the remaining white solid was dissolved in 5 mL dichloromethane and treated with triethylamine (32 uL, 0.229 mmol) and succinic anhydride (11.5 mg, 0.114 mmol) overnight.

More succinic anhydride (11 mg, 0.11 mmol, 0.2 equiv.) was added and after 60 min the reaction was diluted with dichloromethane and washed with KHSO4 5%/K2SO4 10% buffer.

The organic extract was dried over MgSO4 anhydrous, filtered and evaporated to obtain 390 mg of the desired product.

MS (ISP): (M+H ) 881.7 Example 33 N-I 664(3beta)-cholest-5-en-3-yloxy]-4,66-dioxo 8,11,14,17,20,23,26,29,32,35,38,41,44,47,50,53,56,59,62-nonadecaoxa-5,65diazahexahexacontan-1-oy1 } -L-phenylalanyl-N-6--[(4methoxyphenyl)(diphenyl)methyli -N-I4({ [(4-nitrophenoxy)carbonyl]oxylmethyl)phenyThL-lysin amide 0 41111 CVvral 0 0 00Y 0 ..

H N0 Igit or .1,, _ 0 H 0 ON gab \*/ Mgr The title compound was prepared analogous to example 16 using a cholesterol- PEG derivative in step 2 of the synthesis.

MS (ISP): (M+H) 2228.1 The necessary cholesterol-PEG intermediate 1-[(3beta)-cholest-5-en-3-yloxy]1,63-dioxo 5,8,11.14,17,20,23,26,29,32,35,38,41,44,47,50,53,56,59-nonadecaoxa-2,62diazahexahexacontan-66-oic acid for step 2 was prepared as follows:

CA 2976966 2017-08-21 -76Step a: (3beta)-cholest-5-en-3-y1 (59amino3,6,9,12,15,18,21,24,27,30,33,36,39,42,45,48,51,54,57 nonadecaoxanonapentacont-1yl)carbamate Chiral oroolH OHO A oOoo 0000 0000 O Alpha, omega-bis-amino 20(ethylene glycol) (538 mg, 0.6 mmol) and triethylamine (92 uL, 0.66 mmol) were dissolved in 15 mL dry dichloromethane. A solution of cholesteryl chloroformate (270 mg, 0.6 mmol) in 2 mL dry dichloromethane was added dropwise at room temperature.

The solution was stirred overnight, then concentrated in vacuo to a small volume and purified directly on silica gel (eluent: MeC12/Me0H=95:5 -> 9:4 -> 4:1 v/v) to obtain the product (350 mg, 0.254 mmol) as a waxy solid.

MS (ISP): (M+H) 1309.9 Step b: 1[(3beta)-cholest-5-en-3 -yloxy] 1,63-dioxo 5,8,11,14,17,20,23,26,29,32,35,38,41,44,47,50,53,56,59-nonadecaoxa-2,62diazahexahexacontan-66-oic acid CA 2976966 2017-08-21 -77Chiral oricH 00 0 oOoo ooOo O 0 The amine from step a (329 mg, 0.251 mmol), succinic anhydride (26.4 mg, 0.264 mmol) and triethylamine (40 uL, 0.286 mmol) wered dissolved in 5 mL dry dichloromethane.

After adding more triethylamine (40 uL, 0.286 mmol), the solution (pH>8) was stirred overnight at room temperature.

The reaction was diluted with dichloromethane and washed twice with a KHSO4 5%/K2SO4 10% aqueous mixture.

The organic extract was dried over anhydrous MgSO4, filtered and evaporated to obtain the product (260 mg, 0.175 mmol) as a colorless, waxy solid.

MS (ISP): (M+NH4 ) 1408.9 The following working examples illustrate the invention:

Example 34: General procedure for the preparation of RNA conjugates Materials Dimethyl sulfoxide (DMSO), N,N-Diisopropylethylamine (DIPEA) and sodium acetate solution (3 M, pH 5.2) were purchased from Sigma Aldrich Chenne GmbH (Traufkirchen, Germany).

Triethylammonium acetate (TEAA) (2.0 M, pH 7.0) and Acetonitrile (ACN, HPLC quality) for RP-HPLC were purchased from Biosolve (Valkenswaard, Netherlands).

Ethanol (Et0H, p.a.) was purchased from Merck (Darmstadt, Germany).

Purified water from a Optilab HF (Membra Pure, Germany) system was used.

Resource RPC 3 mL column (10 x 0,64 cm; 15 pm particle size) was purchased from GE Healthcare (Freiburg, Germany).

CA 2976966 2017-08-21 -78HPLC purification was accomplished using an AKTA Explorer 100 system (GE Healthcare).

Synthesis of amino-modified RNA RNA equipped with a hexylaminolinker at the 5'-end of the sense strand was produced by standard phosphoramidite chemistry on solid phase at a scale of 1215 p.mol using an AKTA Oligopilot 100 (GE Healthcare, Freiburg, Germany) and controlled pore glass as solid support (Prime Synthesis, Aston, PA, USA). RNA containing 2'49-methyl nucleotides were generated employing the corresponding phosphoramidites, 2'49-methyl phosphoramidites and TFAhexylaminolinker amidite (Sigma-Aldrich, SAFC, Hamburg, Germany).

Cleavage and deprotection as well as purification was achieved by methods known in the field (Wincott F., et al, NAR 1995, 23,14, 2677-84).

The amino-modified RNA was characterized by anion exchange HPLC (purity:

96.1%) and identity was confirmed by ESI-MS ([M+14] + calculated: 6937.4; [M+H] l+measured: 6939Ø Sequence: 5'-(NH2C6)GGAAUCuuAuAuuuGAUCcAsA-3'; u, c: 2'49-methyl nucleotides of corresponding RNA nucleotides, s: phosphorthioate.

General experimental conjugation procedure The title compounds of examples 1-33 were coupled via the amino-modified RNA according the following procedure:

RNA equipped with a C-6 aminolinker at the 5'-end (16.5 mg, 1 equivalent) is dissolved in 500 4, DMSO and 150 pi, water.

The p-Nitrophenylcarbonate derivative (10 equivalents) dissolved in 1 mL DMSO is added followed by 8 pt DIPEA.

The reaction mixture is shaken at 35 C in the dark and monitored using RP-HPLC (Resource RPC 3 mL, buffer: A:

0.1M TEAA in water, B: 0.1M TEAA in 95% ACN, gradient: 3% B to 100% B in 20 CV).

Once the reaction is gone to completion the RNA conjugate is precipitated using sodium acetate (3 M) in Et0H at ¨20 C.

For examples lacking a MMT protecting group in the dipeptide motif the con-esponding conjugates are purified using the conditions described above.

Pure fractions arc pooled and the material is precipitated using sodium acetate/Et0H to give the desired RNA conjugate.

RNA conjugates containing a MMT protecting group in the dipeptide sequence are further processed according to the procedure given below.

CA 2976966 2017-08-21 -79General procedure for MMT cleavage The crude RNA conjugate pellet is dissolved in 500 P., water and 1.5 mL sodium acetate buffer (3 M, pH 5.2 or 0.1M, pH 4.0).

The solution is shaken for 2 days at 30 C.

The reaction mixture is monitored using RP-HPLC (Resource RPC 3 mL, buffer: A: 0.1M TEAA in water, B: 0.1M TEAA in 95% ACN, gradient: 3% B to 100% B in 20 CV).

After complete cleavage of the MMT protecting group the RNA conjugate is directly purified using the conditions just mentioned above.

Pure fractions are pooled and the desired conjugate is precipitated using sodium acetate/Et0H.

As a control a RNA conjugate lacking the dipeptide motif was synthesized.

For this purpose cholesterol was attached to the 5'-end via a linker described in the literature (Nature Biotech, 2007, 25, 1149).

This conjugate is referred to as "non-cleavable".

All the RNA conjugates were analyzed by RP HPLC for purity and identity was confirmed by ESI MS (negative mode).

Briefly, RP-HPLC was performed on a Dionex Ultimate system (Dionex, Idstein, Germany) equipped with a XBridge C18 column (2.5 x 50mm, 2.5 pm particle size, Waters.

Eschbom, Germany) at 65 C column temperature.

Gradient elution was performed using 100 mM hexafluoroisopropanol (HFIP) and 16 mM triethylamine in 1% methanol as eluent A and in 95% methanol as eluent B (1% B to 18%B in 30 minutes). UV detection was recorded at 260 nm.

For mass spectrometric analysis a ThermoFinnigan LCQ DecaXP ESI-MS system with micro-spray source and ion trap detector was coupled online to the HPLC system.

Examples of specific compounds of formula (IIa) are disclosed in table 1.

The resulting compounds are referred to "di-peptide containing cholesterol siRNA conjugates" , wherein the specific di-peptide containing cholesterol siRNA conjugates are further referred to as "Title compound example X- (NHC6)-(siRNA sequence)" and "siRNA with Title compound of Example X".

siRNA preparation Antisense sequence: 5 '-uuGGAUcAAAuAuAAGAuUCcscsU-3 u, c: 2'-O-methyl nucleotides of corresponding RNA nucleotides, s:

phosphorthioate The di-peptide containing cholesterol siRNA conjugates directed against the apolipoprotein B mRNA were generated by mixing an equimolar solution of complementary strands in annealing CA 2976966 2017-08-21 -80buffer (20 mM sodium phosphate, pH 6.8; 100 mM sodium chloride), heated in a water bath at 80-85 C for 3 minutes and cooled to room temperature over a period of 3 - 4 hours.

Duplex fon-nation was confirmed by native gel electrophoresis.

All prepared di-peptide containing cholesterol siRNA conjugates are listed in table 2.

Table 1: Di-peptide containing cholesterol siRNA conjugates (5' -3') and analytical data.

Key:

lower case letters a, c, g, u, are 2'-0-Methyl nucleotides; A phosphorothioate linkages is symbolized with a lower case "s". (NHC6) is the aminohexyl linker incorporated at the 5'-end of the sense strand. T ;

' v61.1twa.61 ,.

1 6:616 VA Itan,. MOi IS14.4$ hi1 t'llibs ,:sset3tomi:tRIt. 1 el:(M."5:ty.',60..4.AMI\cus:.,,s.4,ticeveAtoik &103.7 111103.2 173 .

--:-.1 13.1 0'1116 c:6044,..-$6,1 EX, 2'..V:111>1.10Sati.AUCultA,Ak4;.,1,1(zh1.1C;kkgA ').7.'.57..'.' E7.52,4 0' 6 1 32 ri it.16 comp6n.r.d Ex 7.).X1I11COGGAAtiti:.8.6.1:46.AnKaelk1X,YAK A C.',..71.'õ E226. .5 ..

SS 17' el'itI,, ,6$16.664 Ex 1. 'N'11.1-11.>0 ...K.,:6AIR1+,41tA'µO.u64;1,=61*.`1;04PA n"-o.7 w73. R1.6 $ :KØ eritt :t.".t*.=;.:Atti.

Et 2f.Y.V.Hi2,60kAUCts,..4µ4µ41%,4kA4AUCg..,,10.A, 7E=2=4 t .n.n. ' ..a t = =,: M46.6 26. ait.16. v.)1461.166 Eg: Xti.N.Hat.)00AAU1.76u.A.6A663.414;11COAA 77 a.A1 7g411.0 77.4 .9 04 g i 1..õ,-,........,29,03AMMU241MM.V09166gliMeaMVIFLedt,--,,,,,M191-.õõ-,,,õ71221,--õõ,,-,X14 1 ; (Ti116 ..x4r5pm6.<1 11%. 1.X111i.(16,10C1A.AUCk61A,:kl..i:11612:0:1C,.A A ),::(1..., 79:35.2 n2 .:.,.....T.2,_, 1.... õõstgiqtam.FA 4 m4.3tklõliccelPA,:ttriA.A.vatl,u.cs4........õ.õ:1211:::.......õ......12.11.E.,õ õõ_õõõFil ____ .5; (fith7.,;:o.

......... 1 0 4Titu ,:A.-/),A$5.-4 E..x 6YN1irf.N.KVSA:11C4-6A31.6.tuta3-1...1.3C6.$A 7936.:', 77393'5 = ' 93 3 , \'-ritIA' '''''Va4.'al.kg:01Elk:'',:a6;=,':',.õMata',4,ts!Altt:!2:E2:''õõõõla;V%:=1-õ,....õõ.õ..211,1 ...............

SS 'il (Mit comp:mad Ea -141.V.'.',0.'.`K,IA.Ae'.(.1=5r1u.k:õAl.v.rolittC!M ak WC:4...1 7ooa. 7 .... ,...A -- i 10. t.71116 66sr.v.ti Et 1 CVlicIncidA.ALT (1.,At.A.I.k:U6.6$1.61:.u..,%,64A 79 f.2.1 79.21. (I 95 .= 4 11 (.1-at msq.....6:%63c1E.:g: 1 1)3:11110.1)0C.I.A43C6,,a1w4s..4.6.11C4AM. WO:2.6 79;.. 1. ,5 ' 9g .7 .

i1111 <T.i.61,.

Gov.,i)...-...3.4 r,-., .i.'..1.0;11,t.!:t..,',.=...i.,:.2,,,o,.,;.wc....,a.A.A.,...K.:Iikuccahm. 7E'nt4,54 7.?.36..$ , 94.41 gleA1111a291M296ALTCF,2e:Wat :VA,,,,-.21.4Q.

! --,....2179, -55---.....2-11 __________________________________________________ --I CS AA 7721.õ1:

- = - ;I!..M._õ...._....?,P.1..-1 .,õõ,.

'222 (71t/..= - __ . 77Z. /....7 . .._ ....... 7767.9, ....... ,__$V ...

,Nr4:4.kkiA=tiE=.K 261.:111-1C.6)(7(.1,6,t4.1(16i1A.66i.41:WC,,MA nCi,5.., 79117.0 % 4 i Title.,631',01114 Ex 2:',,X141,5000(1..AAU.,:k$6,A.6.66'.1-42.kMA.6 1 M71.9 M73.2 WI 1 _______ 21)1 1Ti1.16 mak:k6612tlit Ex 2:3,X.1<114000:1AA11.6,,ktAuta0M.1::1cAk,..µ 1 _ U41.) 7 . . niM - ... = ===, omom=.1.

Ex :'...%.14.1-10011.:1At.A.P.7u6At0.646:1A,11:CK,; IA .

: "A:: 47i.gt* 1 ,,,,...4=Six,4E ', 1'..',,Y1:411.COOGAAt.k.)Wik,aS.:4!-Ø4CVATCS:Ao:t\ 7141 6 ... 7F41 1 1 21; rrit16 .631...6owt E '21V4Hqope.A.Atigmk,mr,..41%/...1r4ArsA, .-6 ----1 F- ...... 1.51..fautcquwararaliacsaggAto..,,,,,,,A,mawc14.0k i.. *.-v---.--i-.. lizzi- 97.43 Table 2: Di-peptide containing cholesterol siRNA conjugates .

The last entry (SEQ ID NO pair 266/154) represents a siRNA conjugate lacking the di-peptide motif.

Key: lower case letters a, c, g, u, are 2'-0-Methyl nucleotides; A phosphorothioate linkages is symbolized with a lower case "s". (NHC6) is the aminohexyl linker incorporated at the 5'-end of the sense strand.

CA 2976966 2017-08-21 -81SEQ ID SEQ ID No Sense sequence (5'-3') No Antisense sequence (5'-3') (Title compound Ex16)(NHC6) - GGAAUCuuAuAuuuGAUCc As A 154 uuGGAUcAAAuAuAAGAuUCcscsU (Title compound Ex 31)(NHC6) - GGAAUCuuAuAuuuGAUCcAs A 154 uuGGAUcAAAuAuAAGAuUCcscsU (Title compound Ex 33)(NHC6) - GGAAUCuuAuAuuuGAUCc As A 154 uuGGAUcAAAuAuAAGAuUCcscsU (Title compound Ex 32)(NHC6) - GGAAUCuuAuAuuuGAUCcAs A 154 uuGGAUcAAAuAuAAGAuUCcscsU (Title compound Ex 17)(NHC6) - GGAAUCuuAuAuuuGAUCcAs A 154 uuGGAUcAAAuAuAAGAuUCcscsU (Title compound Ex 30)(NHC6) - GGAAUCuuAuAuuuGAUCc As A 154 uuGGAUcAAAuAuAAGAuUCcscsU (Title compound Ex 27)(NHC6) - GGAAUCuuAuAuuuGAUCcAsA 154 uuGGAUcAAAuAuAAGAuUCcscsU (Title compound Ex 28)(NHC6) - GGAAUCtmAu AtiuuGAUCc AsA 154 uuGGAUcAAAuAuAAGAuUCcscsU (Title compound Ex 29)(NHC6) - GG AAUCuuAttAuuuGAUCc As A 154 uuGGAUcAAAuAuAAGAuUCcscsU (Title compound Ex 1)(NHC6) - GGAAUCuuAuAuuuGAUCcAs A 154 uuGGAUcAAAuAuAAGAuUCcscsU (Title compound Ex 2)(NHC6) - GG AAUCuuAuAuuuG AU CcAs A 154 uuGGAUcAAAuAuAAGAuUCcscsU (Title compound Ex 3)(NHC6) - GGAAU CuuAuAuuuGAU CcAs A 154 uuGGAUcAAAuAuAAGAuUCcscsU (Title compound Ex 4)(NHC6) - GGAAUCuuAuAuuuGAUCc As A 154 uuGGAUcAAAuAuAAGAuUCcscsU (Title compound Ex 5)(NHC6) - GGAAUCuuAuAuuuGAU Cc As A 154 uuGGAUcAAAuAuAAGAuUCcscsU (Title compound Ex 6)(NHC6) - GGAAUCuuAuAuuuGAUCc As A 154 uuGGAUcAAAuAuAAGAuUCcscsU (Title compound Ex 7)(NHC6) - GGAAUCuuAuAuuuGAUCcAs A 154 uuGGAUcAAAuAuAAGAuUCcscsU (Title compound Ex 8)(NHC6) - GGAAUCuuAuAuuuGAUCcAs A 154 uuGGAUcAAAuAuAAGAuUCcscsU (Title compound Ex 9)(NHC6) - GG AAUCuuAuAuuuG AUCc As A 154 uuGGAUcAAAuAuAAGAuUCcscsU (Title compound Ex 10)(NHC6) - GGAAUCuu Au Au u uGAU Cc As A 154 uuGGAUcAAAuAuAAGAuUCcscsU (Title compound Ex 11)(NHC6) - GG A AUCuuAu Au uuGAUCcA sA 154 uuGGAUcAAAuAuAAGAuUCcscsU (Title compound Ex 12)(NHC6) - GGAAUCuuAuAuuuGAUCcAs A 154 uuGGAUcAAAuAuAAGAuUCcscsU (Title compound Ex 13)(NHC6) - GGAAUCuuAuAuuuG AUCc A s A 154 uuGGAUcAAAuAuAAGAuUCcscsU (Title compound Ex 14)(NHC6) - GG AAUCuuAuAuuuG AU CcAs A 154 uuGGAUcAAAuAuAAGAuUCcscsU (Title compound Ex 24)(NHC6) GGAAUCuuAuAuuuGAU Cc As A 154 uuGGAUcAAAuAuAAGAuUCcscsU (Title compound Ex 23)(NHC6) - GGAAUCuuAuAuuuGAUCc As A 154 uuGGAUcAAAuAuAAGAuUCcscsU (Title compound Ex 22)(NHC6) - GGAAU CuuAuAuuuGAU CcAs A 154 uuGGAUcAAAuAuAAGAuUCcscsU (Title compound Ex 26)(NHC6) - GGAAUCuuAuAuuuGAUCc As A 154 uuGGAUcAAAuAuAAGAuUCcscsU (Title compound Ex 25)(NHC6) - GGAAUCuuAuAuuuGAUCcAsA 154 uuGGAUc AAAuAuAAGAuUCcsc sU CA 2976966 2017-08-21 -82SEQ ID SEQ ID No Sense sequence (5'-3') No Antisense sequence (5'-3') (Title compound Ex 20)(NHC6) GGAAUCuuAuAuuuGAUCcAsA 154 uuGGAUcAAAuAuAAGAuUCcscsU (Title compound Ex 19)(NHC6) GGAAUCuuAuAuuuGAUCcAsA 154 uuGGAUcAAAuAuAAGAuUCcscsU (Title compound Ex 18)(NHC6) GGAAUCuuAuAuuuGAUCcAsA 154 uuGGAUcAAAuAuAAGAuUCcscsU (Title compound Ex 21)(NHC6) GGAAUCuuAuAuuuGAUCcAsA 154 uuGGAUcAAAuAuAAGAuUCcscsU (Title compound Ex 15)(NHC6) GGAAUCuuAuAuuuGAUCcAsA 154 uuGGAUcAAAuAuAAGAuUCcscsU (Chol)GGAAUCuuAuAuuuGAUC 266 cAsA 154 uuGGAUcAAAuAuAAGAuUCcscsU Example 35: In vivo experiments Co-Administration of di-peptide containing cholesterol siRNA conjugates and delivery polymer in vivo.

Six to eight week old mice (strain C57BL/6 or ICR, ¨18-20 g each) were obtained from Harlan Sprague Dawley (Indianapolis IN).

Mice were housed at least 2 days prior to injection.

Feeding was performed ad libitum with Harlan Teklad Rodent Diet (Harlan, Madison WI).

Mice (n=3 per group) were injected with a mixture of 0.2 mL solution of delivery polymer and 0.2 ml di-peptide containing cholesterol siRNA conjugates .

The injected dose was, unless otherwise stated, 15 mg/kg for the delivery polymer and 0.1 mg/kg with respect to the di-peptide containing cholesterol siRNA conjugates .

Solutions were injected by infusion into the tail vein.

48 hours post injection serum ApoB levels were measured relative to isotonic glucose treated animals according to the procedure below.

Serum ApoB levels determination.

Mice were fasted for 4 h before serum collection by submandibular bleeding.

Serum ApoB protein levels were determined by standard sandwich ELISA methods.

Briefly, a polyclonal goat anti-mouse ApoB antibody and a rabbit anti-mouse ApoB antibody (Biodesign International) were used as capture and detection antibodies respectively.

An HRP-conjugated goat anti-rabbit IgG antibody (Sigma) was applied afterwards to bind the ApoB/antibody complex.

Absorbance of tetramethyl-benzidine (TMB, Sigma) colorimetric development was then measured by a Tecan Safire2 (Austria, Europe) microplate reader at 450 nm.

In figure 1 various di-peptide containing cholesterol siRNA conjugates were benchmarked against the same siRNA conjugated to cholesterol but lacking the cleavable motif elaborated CA 2976966 2017-08-21 -83earlier in this section.

The effect of this siRNA conjugate (SEQ ID NO pair 266/154, "noncleavable control") on serum ApoB levels was set to 1 in order to evaluate the influence of the di-peptide containing conjugates relative to the non-cleavable control.

Substituting the initially used Phe-Lys motif (siRNA with Title compound of Example 16) with the corresponding Damino acids (siRNA with Title compound of Example 14) or just replacing the Lys with the unnatural enantiomer (siRNA with Title compound of Example 30) yielded ApoB reduction less pronounced or equivalent to the non-cleavable control siRNA.

Replacing Lys by Gly (siRNA with Title compound of Example 23) or Phe by p-Methoxyphenylalanine (siRNA with Title compound of Example 13) reduced the potency compared to siRNA with Title compound of Example 16.

Other di-peptide motifs containing siRNA conjugates were shown to be as efficacious as the original Phe-Lys containing conjugate.

Figure 2 summarizes di-peptide containing cholesterol siRNA conjugates that were as efficacious or had improved efficacy compared to siRNA with Title compound of Example 16 consisting of the Phe-Lys motif.

All these conjugates were significantly more active compared to the "non-cleavable" cholesterol siRNA conjugate SEQ ID NO pair 266/154.

The best performing di-peptide containing cholesterol siRNA conjugates had a fluorine modified phenyl ring in the Phy-Lys motif (siRNA with Title compound of Example 8, siRNA with Title compound Example 9) or had the phenylalanine substituted with beta-phenylalanine (siRNA with Title compound of Example 11) or a derivative thereof (siRNA with Title compound of Example 10).

Since di-peptide containing cholesterol siRNA conjugates with di-peptide motifs consisting of D-amino acids are performing equal to the non-cleavable control conjugate it is conceivable that the other di-peptide sequences are indeed cleaved by a protease activity in vivo.

However, given the broad acceptance of different amino acids and derivatives thereof it is likely that more than one enzyme is participating in the cleavage reaction as suggested in the literature (Bioconjugate Chem. 2002,13,855).

As shown in figure 3, the incorporation of a Cathepsin cleavable di-peptide motif (in this case Phe-Lys, siRNA with Title compound of Example 16) between the siRNA and the small molecule ligand cholesterol boosts the potency of the siRNA conjugate compared to the straight cholesterol siRNA conjugate (SEQ ID NO pair 266/154).

Further spacing of the cholesterol ligand from the di-peptide motif by means of PEG based linkers diminishes the potency proportional to the length of the PEG linker.

CA 2976966 2017-08-21 -84In figure 4 the polymer dose was kept constant at 15 mg/kg.

The siRNA dose was titrated and the effect on serum ApoB content was measured.

The Di-peptide containing cholesterol siRNA conjugates containing the Phc-Lys (F-K) motif was significantly more potent compared to thc control conjugate lacking the di-peptide sequence.

Example 36: 2'-modified oligoribonucleotide synthesis Oligoribonucleotides were synthesized according to the phosphoramidite technology on solid phase.

Depending on the scale either an ABI 394 synthesizer (Applied Biosystems) or an AKTA oligopilot 100 (GE Healthcare, Freiburg, Germany) was used.

Syntheses were performed on a solid support made of controlled pore glass (CPG, 520A, with a loading of 75 gmol/g, obtained from Prime Synthesis, Aston, PA, USA).

All 2'-modified RNA phosphoramidites as well as ancillary reagents were purchased from SAFC (Hamburg, Germany).

Specifically, the following 2'-0-Methyl phosphoramidites were used: (5' -0-dimethoxytrityl-N6-(benzoy1)-2' -0-methyladenosine-3' -0-(2-cyanoethyl-N,N-diisopropylamino) phosphoramidite, 5'-0dimethoxytritylN4-(acety1)-2'-0-methyl-cytidine-3'-0-(2-cyanoethyl-N,N-diisopropylamino) phosphoramidite, (5' - 0-dimethoxytrityl-N2-(i so butyry1)-2' - 0-methyl-guano s ine-3 ' -0-(2cyanoethyl-N,Ndiisopropylamino) phosphoramidite, and 5'-0-dimethoxytiity1-2'-0-methyluridine-3'-0-(2cyanoethyl-N,N-diisopropylamino) phosphoramidite.

The 2' -Deoxy-2' -fluorophosphoramidites carried the same protecting groups as the 2'-0-methyl RNA amidites.

All amidites were dissolved in anhydrous acetonitrile (100 mM) and molecular sieves (3A) were added.

To generate the 5'-phosphate the 2-12-(4,4'Dimethoxytrityloxy)ethylsulfonyl]ethyl-(2-cyanoethyl)(N,N-diisopropy1)-phosphoramidite from Glen Research (Sterling, Virginia, USA) was used.

In order to introduce the C-6 aminolinker at the 5'-end of the oligomers the 6 (Trifluoroacetylamino)-hexyl-(2-cyanoethyl)-(N,N-diisopropy1)phosphoramidite from Thermo Fisher Scientific (Milwaukee, Wisconsin, USA) was employed.

The 5' -modifications were introduced without any modification of the synthesis cycle. 5-Ethyl thiotetrazole (ETT, 500 mM in acetonitrile) was used as activator solution.

Coupling times were 6 minutes.

In order to introduce phosphorothioate linkages a 50 mM solution of 3((Dimethylamino-methylidene)amino)-3H-1,2,4-dithiazole-3-thione (DDTT, obtained from AM Chemicals, Oceanside, CA, USA) in anhydrous Acetonitrile/pyridine (1:1 v/v) was employed.

CA 2976966 2017-08-21 -85Example 37: Cleavage and deprotection of support bound oligomer.

After finalization of the solid phase synthesis, the dried solid support was transferred to a 15 mL tube and treated with concentrated aqueous ammonia (Aldrich) for 18 hours at 40 C.

After centrifugation the supernatant was transferred to a new tube and the CPG was washed with aqueous ammonia.

The combined solutions were evaporated and the solid residue was reconstituted in buffcr A (see below).

Example 38: Purification of oligoribonucleotides Crude oligomers were purified by anionic exchange HPLC using a column packed with Source Q15 (GE Helthcare) and an AKTA Explorer system (GE Helthcare).

Buffer A was 10 mM sodium perchlorate, 20 mM Tris, 1 mM EDTA, pH 7.4 (Fluka, Buchs, Switzerland) and contained 20% Acetonitrile and buffer B was the same as buffer A with the exception of 500 mM sodium perchlorate. A gradient of 22%B to 42%B within 32 column volumes (CV) was employed. UV traces at 280 nm were recorded Appropriate fractions were pooled and precipitated with 3M Na0Ac, pH=5.2 and 70% Ethanol.

Finally, the pellet was washed with 70% Ethanol.

Example 39: Annealing of oligoribonucleotides to generate siRNA Complementary strands were mixed by combining equimolar RNA solutions.

The mixture was lyophilized and reconstituted with an appropriate volume of annealing buffer (100 mM NaC1, 20 mM sodium phosphate, pH 6.8) to achieve the desired concentration.

This solution was placed into a water bath at 85 C which was cooled to rt within 3h.

Example 40: In vitro activity of siRNAs devoid of 2'-OH residues In order to investigate if siRNAs lacking any 2'-OH residues show potent in vitro knock down activity, we tested a panel of EGFP mRNA-targeted siRNAs with different 2'-modification chemistries (SEQ ID pairs 31/32 to 149/150, and see Table 3 for examples).

The siRNAs were screened for sense and antisense activity with the Dual-Glo Luciferase Assay System (Promega) using the psiCHECK2 vector (Promega) in COS7 cells (DSMZ, Braunschweig, Germany, cat.

No. ACC-60).

To address the silencing activity conferred by sense and antisense strand we cloned each corresponding 19mer target site sequence as separate psiCHECK2 construct (psiCHECK2-AT for antisense activity, psiCHECK2-ST for sense activity) into the multiple cloning region located 3' to the translational stop codon of the synthetic Renilla luciferase.

By CA 2976966 2017-08-21 -86using Lipofectamine 2000 (Invitrogen GmbH, Karlsruhe, Germany, cat.

No. 11668019) COS7 cells were co-transfected with vector construct and 3nM of the corresponding siRNA complementary to the cloned target site.

Successful siRNA-mediated silencing was determined 24 hours after transfection via the activity of the renilla luciferasc nortnalized to firefly luciferase levels to take transfcction efficiency into account (see Figure 5a for antisense activity and Figure 5b for sense activity).

Table 3: Exemplary siRNA sequences and chemical modifications used for determination of in vitro knock down activity dependent on 2' -modifications.

Reference dupliccs and selected examples of corresponding modification variants used in this study.

Xf indicates a 2'fluoro modification of the nucleotide X, small letters indicate a 2'-0-methyl modification, underlined letters indicate a DNA nucleotide, all other capital letters indicate ribonucleotides.

The letter "p" indicates a 5'-phosphate.

RNA duplices unmod 5'- UGCCCAUCCUGGUCGAGCUTT -3' 3'- ICACGGGUAGGACCAGCUCGAp -5' F/OMe 5'- UfgaEcCfaUfcCfuGfoUfcGfaGfoUfTsT -3' 3'- TsTaCfgGfgUfaCfgAfcCfaGfctifcGtap F/DNA 5'- UfCCfCCfAUfCCfUGfGUfCGfAGfCilfTsT -3' 3' TsTTGEGfalfTGfaLE00T0f-c-Uf-c-GfTp DNA/OMe 5'- UgGcCaUcCuGgUcGaGcUTsT -3' 3'- TsTaCgGgUaGgAcCaGcUcCap It was found that the 5 most potent modified siRNAs (?60% knock-down) were designed in an alternating 2'-fluoro/2'-0-methyls (2'F/2'-0Me) pattern.

While conferring antisense activity, this chemistry fully eliminated the activity of the corresponding sense strands, as shown by lack or minimal renilla luciferase activity for all tested 2'F/2'-0Me variants.

We concluded that such 2'F/2'-0Me pattern is promoting the siRNA's intended antisense strand activity while undesired off-target effects coming from the sense strand are fully suppressed.

CA 2976966 2017-08-21 -87Example 41: Detection of DNAse 11-sensitive sites by in vitro assay An ion pairing (JP) reversed phase (RP) high performance liquid chromatography (HPLC) coupled to an electrospray ionization (ESI) mass spectrometry (MS) or an anion exchange (AEX)-HPLC based method was established to test the in vitro stability of selected single and double stranded RNAs.

Method description: For stability analysis a 10 M solution of either single stranded or double stranded RNA was incubated at 37 C in 5mM sodium acetate buffer solution (pH 4.5) containing 0.8 or 8 units DNase II (from bovine spleen, Type V, Sigma Aldrich).

The incubation reaction was stopped by adding a 100 mM triethyl ammonium acetate (TEAA) solution, shifting the pH to 7 and inactivating the DNase II enzyme.

Analysis was done by either LC/MS combined with UV-detection or by AEX-HPLC with UV-detection. UV-detection traces at 260 nm were used for quantitative analysis, MS data served for cleavage site identification within the RNA sequence.

A. IP-RP-HPLC was done employing a Waters XBridge C18 column (2.5 x 50 mm, 2.5 grn particle size) at 65 C column temperature.

Gradient elution was performed using 100 mM hexafluoroisopropanol (HFIP) and 16 mM triethylamine in 1% methanol as eluent A and composition A in 95% methanol as eluent B. A gradient from 1% B to 18% B in 30 minutes was employed.

B. AEX-HPLC was performed on a Dionex DNA Pac200 column (4 x 250 mm) at 50 C using a 20 mM phosphate buffer containing 10% ACN at pH=11.

Eluent B contained 1 M NaBr in eluent A. . A gradient from 25 to 62% B in 18 minutes was employed.

Table 4: Duplexes and the remaining intact strandsevaluated for their stability against DNase II.

Key:

lower case letters a, c, g, u, arc 2'-0-Methyl nucleotides; Upper case letters A, C, G, U followed by "f "indicates a 2'-fluoro nucleotide.

Lower case "p" indicates a 5'-phosphate.

(invdT) represents an inverted deoxythimidine (3'-3'-linked). A phosphorothioate linkages is symbolized with a lower case "s". dT is deoxythimidine. (NHC6) is the aminohexyl linker incorporated at the 5'-end of the sense strand.

% Intact SEQ strand SEQ % Intact ID after 6 ID Antisense strand sequence (5'- strand after NO Sense strand sequence (5'-3') hours NO 3') 6 hours CA 2976966 2017-08-21 -88GGAuGAAGuGGAGAuuAGud ACuAAUCUCcACUUcAUCCd 157 TsdT 0 158 TsdT 0,1 (NH2C6)GfgAfuGfaAfgUfgGfa pasCfuAfaUfcUfcCfaCfuUfcAf 160 GfaUfuAfgUf(invdT) 101 159 uCfc(invdT) 97 (NH2C6)GfcAfaAfgGfcGfuGfc puGfaGfuUfgGfcAfcGfcCfulifu 165 CfaAfcUfcAf(invdT) 103 166 Gfc(invdT) 103 (NH2C6)GcAAAGGcGuGccAA UGAGUUGGcACGCCUUUGC 167 cucAdTsdT 56 168 dTsdT 49 (NH2C6)GGAUfCfAUfCfUfCf GUfAAGACfUfUfGAGAUfGA 169 AAGUfCfUfUfACfdTsdT 64 170 UfCfCfdTsdT 54 GGAAUCuuAuAuuuGAUCcAs uuGGAUcAAALtAuAAGAuUC 153 A 0,1 154 cscsU 0,1 (NH2C6)UfgAfcCfaCfaGfuCfg pusUfuAfaUfcCfgAfcUfgUfgGf 173 GfaUfuAfaAf(invdT) 102 174 uCfa(invdT) 102 (NH2C6)uGAccAcAGucGGAu puUuAAUCCGACUGUGGucA 175 uAAAdTsdT 0,4 176 dTsdT 0,3 (N1-12C6)uGAccAcAGucGGAu UUuAAUCCGACUGUGGUcA 175 uAAAdTsdT 6 177 dTsdT 3 Conclusions:

A. RNA strands containing at least one 2' -OH nucleotide (e.g. both strands of SEQ JD NO piar 157/158) are rapidly degraded via a cyclic pentavalent intermediate, leading to 2'-3' cyclic phosphates at the 5'-cleavage product.

The formation of the pentavalent intermediate can be inhibited using nucleotides lacking a 2'-OH group, like e.g. 2'-deoxy, 2'-0Me or 2'-F.

B.

Additionally, RNA is degraded via a 5' -exonucleolytic pathway, that is independent from the 2'-modification on the 5'-terminal nucleotides.

This degradation pathway can be inhibited using 5'-terminal non-nucleotide moieties, like e.g. a C6-aminolinker (e.g. SEQ ID NO 160 in SEQ ID NO pair 160/159 or SEQ ID NO 165 in SEQ ID NO pair 165/166) or a phosphorothioate at the first internucleotide linkage (e.g. SEQ ID NO 160 in SEQ ID NO pair 160/159).

C. A 5'-phosphate group slows down the exonucleolytic cleavage kinetics, but can not fully block the degradation starting at this end (e.g. SEQ ID NO 160 in SEQ ID NO pair 160/159).

This is most probably due to the cleavage of the 5'-phosphate by either phosphatases or by an inherent phosphatase activity of the DNase II enzyme.

D.

The best protection for RNA strands was achieved with oligonucicotides containing no 2'-OH nucleotide, starting with a 2'-0Me nucleotide at the 5'-end connected by a phosphorothioate linkage to the second nucleotide (e.g. SEQ ID NO 173 in SEQ ID NO pair 173/174).

Other terminal non-2'-OH nucleotides also protect against the 5'-exo degradation, but to a lower extent compared to the 2'-0Me modification (refer to Table 9) CA 2976966 2017-08-21 -89Example 42: In vivo knock down activity of siRNAs devoid of 2'-OH residues In vivo experiments were conducted with mice injected with Factor VII (FVII)targeting siRNAs (SEQ ID NO pairs 1 79/1 66 and 180/168, see Table 5) co-administered with DPCGalNac.

Table 5 a: Sequences of siRNAs for in vivo experiment.

Key: lower case letters a, c, g, u, are 2'-0Methyl nucleotides; Upper case letters A, C, G. U followed by "f" indicatcs a 2'-fluoro nucleotide.

Lower case "p" indicates a 5'-phosphate. (invdT) represents an inverted deoxythimidine (3'-3'-linked). A phosphorothioate linkages is symbolized with a lower case "s". dT is deoxythimidine. (NHC6) is the aminohexyl linker incorporated at the 5'-end of the sense strand.

GaINAc refers to the structure in formula (IV).

SEQ ID SEQ ID Sequence 5'->3' NO pair NOs 179 Ga1NAc-(NHC6)- GfcAfaAfgGfcGfuGfcCfaAfcUfcAf(invdT) 179/166 166 puGfaGfuUfgGfcAfcGfcCfuUfuGfc(invdT) 180 GaINAc-(NHC6)-GcAAAGGcGuGccAAcucAdTsdT 180/168 168 UGAGUUGGc AC GCCUUUGCdTsdT A FVII siRNA with an alternating 2'-0Me/2'-F pattern on sense and antisense strand was generated with a 5' -terminal 2' -0Me nucleotide on the antisense and a 5'terminal 2'-F strand on the sense strand.

Both strands are protected by an inv(dT) at the 3'-terminal overhang.

The antisense strand was bearing a 5'-phosphate group to maintain activity of the siRNA.

To the 5'end of the sense strand a GaINAc-palmitoyl ligand was conjugated to enable targeting to hepatocytes by the asialyloglycoprotein-receptor expressed on these cells.

siRNA (2.5 mg/kg) was co-administered with GalNAc-targeted PBAVE delivery polymer (15 mg/kg) in mice.

FVII mRNA measurements were done from liver homogenates using QuantiGene 1.0 branched DNA (bDNA) Assay Kit (Panomics, Fremont, Calif., USA,Cat-No: QG0004).

At necropsy 1-2 g liver tissue was snap frozen in liquid nitrogen.

Frozen tissue was powderizedwith mortar and pistil on dry ice. 15-25 mg of tissue was transferred to a chilled 1,5 mL reaction tube, 1 mL 1:3 Lysis Mixture prediluted in MilliQ water and 3,3 pL Proteinase K(50pg/pL) was added and tissue was lysed by several seconds ultrasound sonication at 30-50% power (HD2070, Bandelin, Berlin, Germany).

Lysates were stored at -80 C until analysis.

For CA 2976966 2017-08-21 -90mRNA analysis lysate was thawed and digested with Proteinase K for 15min at 1000 rpm in a thermomixer at 65 C (Thermomixer comfort, Eppendorf, Hamburg, Germany). FVII and GAPDH mRNA levels were determined using QuantiGene 1.0 bDNA Assay Kit reagents according to the manufacturer's recommendations. FVII mRNA expression was analyzed using 20 I. lysate and a tnouse FVII probe set. GAPDH mRNA expression was analysed using 40 1.1I.

lysatc and mints norwegicus probe sets shown to be cross-react with mice (sequences of probesets see above).

As assay readout the chemiluminescence signal at end of the assay was measured in a Victor 2 Light luminescence counter (Perkin Elmer, Wiesbaden, Germany) as relative light units (RLU).

The signal for FVII mRNA was divided by signal for GAPDH mRNA from the same lysate Values are reported as FVII mRNA expression normalized to GAPDH.

Results demonstrate a 79% FVII mRNA knock down at 48 hours post dosing after administration of SEQ ID NO pair 179/166.

In contrast,the 2' -OH nucleotide bearing siRNA SEQ ID NO pair 180/168 showed no significant knock down (< 25 %), as shown in Table 5.

Table 5 b: Results of in vivo knockdown studies SEQ ID NO SEQ ID NO pair 179/166 pair 180/168 Time Remaining Remaining [hour] mRNA [%] mRNA [%] 1 84 92 6 83 88 24 53 100 48 21 76 Example 43: Tissue distribution of siRNAs lacking 2'-OH residues The siRNA concentration in the liver tissue samples was determined using a proprietary oligonucleotide detection method as described in W02010043512.

Briefly, the siRNA quantification is based on the hybridization of a fluorescently (Atto-425) labeled PNA-probe CA 2976966 2017-08-21 -91(Atto425-00-GCAAAGGCGTGCCAACT, obtained from Panagene Inc, Korea) complementary to the antisense strand of the siRNA duplex, followed by AEX-HPLC based separation.

Quantification was done by fluorescence detection against an external calibration curve that was generated from a dilution series of the two FVII siRNA used in the in vivo experiment (see example 42).

For plasma samples between 0.2 to 2 1.t1_, and for tissue ¨ lmg aliquots were injected onto the HPLC system.

Liver tissue analysis of the stabilized siRNA lacking 2'-OH nucleotide showed high concentrations of intact antisense strand in the liver in the ug/g range, but ¨ 95% was present in the 5'-dephosphorylated inactive form (see table 6).

The resulting RNA with a terminal 2'-0Me nucleotide is not prone for rephosphorylation in the cytoplasm by the phosphokinase hC1p1 (see below).

In contrast, the antisense strand of the 2'-OH containing siRNA was completely degraded in the tissue within the first 6 hours post dosing.

Table 6: Liver tissuc analysis of the stabilized siRNA containing no 2'-OH nucleotide SEQ ID NO pair 181/186 in SEQ ID NO pair 181/185 in Liver [ng/g] Liver [ng/g] Time [hour] - 5'-Phosphat + 5'-Phosphat - 5'-Phosphat + 5'-Phosphat 1 873 171 9 BDL 6 1351 106 BDL* BDL 24 1043 65 BDL BDL 48 1062 66 BDL BDL * BDL = below detection limit Example 44: In vitro knock down activity of siRNAs with optimized 5'-ends An additional in vitro screen for FVII siRNAs was conducted in order to identify siRNAs that may be intracellularly (re-)phosporylated at the antisense's 5'-end to result in the RNAicompetent species.

All siRNAs from this screen are shown in Table 7.

The alternating 2'-0Me / 2'-F modification pattern was identical to the 1S1 generationdesign (without any 2'-OH residues) CA 2976966 2017-08-21 -92with exception of various modifications at the first two nucleotides at the 5'end of the antisense strand.

The two 5'-terminal nucleotides of the antisense strand were generated as 2'-F or 2'deoxy modified nucleotides in various combinations with and without an additional 5'-phosphate or 5'-phosphothioate.

All siRNAs were screened in dose response (24 nM to 0.00037 nM in 4fold dilutions) for knock down activity after transfection of primary mouse hepatocytes (30000 cells per well; 96 well plate formate) using Lipofectamine 2000 according to the manufacturer's instructions.

Two siRNAs were comparable active to the parental duplex (SEQ ID NO pair 182/168); comparable active siRNAs: SEQ ID NO pairs 181/186 and 181/185) in terms of IC50 values, one with a 5'-terminal 2'-F and a phosphate group and one with two 5'terminal 2'deoxy nucleotides and a 5'-phosphorothioate (see Table 7 for 1050 values).

Both of them are ¨56-fold more active compared to the siRNA (SEQ ID NO pair 181/166) used in the first animal experiment with the terminal 2'-0Me nucleotide.

Example 45: In vitro 5'-phosphorylation of siRNAs with optimized 5'-termini All siRNAs without a 5'-phosphate or 5'-phosphorothioate listed in Table 7 were assessed for phosphorylation by hC1p1 in a HeLa S100 cell extract.

5'-phosphorylation s was analyzed from S100 HeLa extracts as described by Weitzer and Martinez (S.

Weitzer and J.

Martinez. hClpl: a novel kinase revitalizes RNA metabolism.

Cell Cycle 6 (17):2133-2137, 2007).

Directly after incubation of 1 [tM siRNAs in the S100 HeLa extract containing 5mM ATP, the solution was analyzed by either 1P-RP-HPLC or AEX-HPLC under denaturing conditions by injection of 5 [IL sample solution:

A. IP-RP-HPLC was done employing a Waters XBridge C18 column (2.5 x 50 mm, 2.5 [im particle size) at 65 C column temperature.

Gradient elution was performed using 100 mM hexafluoroisopropanol (HFIP) and 16 mM triethylamine in 1% methanol as eluent A and composition A in 95% methanol as eluent B. A gradient from 1% B to 18% B in 30 minutes was employed.

B. AEX-HPLC was performed on a Dionex DNA Pac200 column (4 x 250 mm) at 50 C using a 20 mM phosphate buffer containing 10% ACN at pH=11.

Eluent B contained 1 M NaBr in eluent A. A gradient from 25 to 62% B in 18 minutes was employed.

CA 2976966 2017-08-21 0 vi Lit cn (A n.) to -4 1-3 01 tu to a' cn r ol n.) o *Cri 1-. t..n.

--.1 o O <

P CO = I CD 1-.

SEQ-ID SEQ ID IC53 NO Sense strand sequence (5%3') : NO AniIsense strand sequence (5'-3') (OM 18 1 .0fcAfaMEG* CINGfeC faAfcU fc Af(itivc1T) 185 , p WS GfaGfuli fgGfcAfcCrfc C MI faGfc (itytc1T) 0;17 182 .0 cAAAGG cOuGccAAcucA ciTs cIT _ 168 U GA GIRT GO cA cOCCUUIJ GCSr scIT 0,228 , 181 OfcAfaAfgGfc GftiGfcC faAfcLT fc Al( invdT)_ 186. pscrrciGaGfulI fgGfcAfc Gfceft_LIGfc (itiv err) o, 228 1S1 I Gfc AfaAfgGfc Ci fuG fcC faAfcii fc.A4itivcrr) 187 .. 0 s cirsGfaGU fgGfcAfc Gfc C fttUftiGfc(invcrr) 0,554 18 1 = GfcAfaAfgGf's GfuGfcC faAfcU fcAfOnvcrlp 188 _FidT scIG aG ii./U %RIC Aft Gfc C fuU fuOf e (invdT) 0,631 181 = GfcAfaAfg0fCGfuGfcC faAfell fcAfinvd1) 189 p dT sGfaGfuUfgGfcAfc0fc CfaU fu G fc(inv di) 0702 181 Qfc4faAfgGfcCifuGfcCfa.AfclIfcAkinvcip k 190 pufaGfuLTfgGfcAfc Gfc CfuLT5iGfc(irridT) 0,749 18 1 G fcAfaAfg Oft! G fu G fcC faAfeU ft' Aginvdt 166 pi_lriaGfLiUfgGfc Afc Gfc CfuU fuGfc(inirgiT) 1,Q02 181 GfcAfa.MaGfc GfuGfcCfaAfcU fc .e.k invert) 191 to sUfGfaGfuT.3 fgGfcAfcGfeC ful.3 fa Gfc Cinvap 1,12 131 GfcAfaAfgOfc G fuGfcC faA fc= Er fcAiT inv SI) 192 Ufsr,.1faGfullfgC'TfcAfeGfc CfuLT fuGfc(invd1) 2,257 18 1 OfSAfaAfgGfc GfuGfcCfaAfcLT fc.Aff,iir.,,d1) 193 _ pleTjfsGfaGfui-TfgGfcAfc Gfc cfuLt fuGfc(invdT) 2,42 8 18 1 G ft: AfaAfgG fc GfuGfcC f aAfc LT fcA kinvd1) 194 OS clTsdGaGfuU fgGfcAfcGfcCfaU fi,l0fc(inn7di) 3,208 131 GfcAfaAfgGfc GfuGfc C faA fell fcA tiriv Sr) 195 usGfaGfuLT fgGfcAfc GfcC faUfuG fc (lay di) 3,974 , 181 GfcAfaAfgGfcGfuGfcC faAfcU fcAginvdT) 198 u0faGfaUfgGfcAfcGfcCfuUfuGfc inv 4T) 4,235 18 1 GfcAfaAfgGfc GfuCifcC faAfc13 fc.4;_nvcri.) 197 (IT sCrt¨ aGfulJ fgGfcAfc GfcCfuLlfuCifc(invdT) 18 1 Gf cAfaAf?, Gfc G fuGfcC faAfcti feA itiv4 198 ; s cITCsfa.GfeJf GfcAfeGfc Cfu LT fu.0 fe(itivciT) 4,704 18 1 GfcAfa.AfgGfc GfuGfc.0 faAfcTifcAkinvdT) 199 crr s clGaGfuLlfgGfc AFSGfcCMJ fuGfDriv Tr) 5,341 183 (Choi) C,,:fc AfaAfgGfa OfuGfcCfaAfcljfcekkuWcIT) 190 ious(3f.

Oftillfg Oft Afc 0 fc CfupfuOffi(iriviiT) 9, 77 1 -94The ratio of 5' -phosphorylation is calculated for each strand of a siRNA from the UV trace at 260nm using the following equitation (PA is peak area):

% (5'-phophorylation) = 100 * PA [5'-phosphorylated strand] / (PA [5'phosphorylaied strand] PA [parent strand]) In Table 8 is shown, that the antisense strand of an siRNA cannot be 5'phosphorylated, when a 2'-0Me nucleotide is located at the 5'-terminus (SEQ ID NO pair 181/196 and SEQ ID NO pair 181/195).

In contrast the antisense strand is susceptible to 5'phosphorylation, when a 2'-F, 2'deoxy or 2'-OH nucleotide is incorporated at the 5'-terminus (SEQ ID NO pair 181/195, SEQ ID NO pair 181/192, SEQ ID NO pair 181/197, SEQ ID NO pair 181/199 and SEQ ID NO pair 182/168).

The two siRNAs, that were comparably active in the in vitro assay as the parental SEQ ID NO pair 182/168 (SEQ ID NO pair 181/186 and 181/185), are susceptible to 5' phosphorylation once the synthetically introduced 5'-phosphate / 5'-PTO group is cleaved in vivo, eg. by phosphatases.

Table 8: Percentage of 5'-phosphorylated strand after 4 hours incubation in S100 HeLa cell extract.

Key:

lower case letters a, c, g, u, arc 2'-0-Methyl nucleotides; Upper case letters A, C, G, U followed by "f "indicates a 2'-fluoro nucleotide.(invdT) represents an inverted deoxythimidine (3'-3'-linked). A phosphorothioate linkages is symbolized with a lower case "s". dT is deoxythimidine.

sense SEQ antisen SEQ ID 5'P ID se 5'P NO Sense strand sequence (5'-3') [c701 NO Antisense strand sequence (5'-3') [To] GfcAfaAfgGfcGfuGfcCfaAfcUfc uGfaGfuUfgGfcAfcGfcCfulifuGfc 181 Af(invdT) 52 196 (invdT) 0 GfcAfaAfgGfcGfuGfcCfaAfcUfc usGfaGfuUfgGfcAfcGfcCfuUfuGf 181 Af(invdT) 53 195 c(invdT) 0 GfcAfaAfgGfcGfuGfcCfaAfcUfc UfsGfaGfuUfgGfcAfcGfcCfutlfuG 181 Af(invdT) 44 192 fc(invdT) 22 GfcAfaAfgGfcGfuGfcCfaAfcUfc dTsGfaGfuUfgGfcAfcGfcCfnUfuG 181 Af(invdT) 42 197 fc(invdT) 22 GfcAfaAfgGfcGfuGfcCfaAfcUfc dTsdGaGfuUfgGfcAfcGfcCfuUfu 181 Af(invdT) 47 199 Gfc(invdT) 13 GcAAAGGcGuGccAAcucAdTsd UGAGUUGGcACGCCUUUGCdT 182 T 31 168 sdT 42 GfcAfaAfgGfcGfuGfcCfaAfcUfc UGAGUUGGcACGCCU UU GCdT 184 A 22 168 sdT 100 CA 2976966 2017-08-21 -95Example 46: In vitro DNAse II-stability of siRNAs with optimized 5' ends All antisense strands were screened for DNAse II stability as described in example 41.

The two antisense strands present in the siRNAs that were comparable active to the parental duplex (SEQ ID NO 186 and SEQ ID NO pair 185 one with a 5' -terminal 2'-F and a phosphate group and one with two 5'-terminal 2'-deoxy nucleotides and a 5'-phosphorthioate are stable towards DNAse II cleavage II (> 70% intact strand after 20 hr incubation).

Table 9: In vitro stability of siRNAs towards DNase II after 20 hours incubation Sense Antisense SEQ ID SEQ ID NO NO Sequence (5'-3') % intact strand 181 192 UfsGfaGfuUfgGfcAfcGfcCfuUfuGfc(invdT) 11 181 197 dTsGfaGfuUfgGfcAfcGfcCfuUfuGfc(invdT) 0 181 199 dTsdGaGfuUfgGfcAfcGfcCfuUfuGfc(invdT) 0 181 193 psUfsGfaGfuUfgGfcAfcGfcCfuUfuGfc(invdT) 106 181 187 psdTsGfaGfuUfgGfcAfcGfcCfuUfuGfc(invdT) 96 181 194 psdTsdGaGfuUfgGfcAfcGfcCfuUfuGfc(invdT) 101 181 191 psUfGfaGfuUfgGfcAfcGfcCfuUfuGfc(invdT) 100 181 198 psdTGfaGfuUfgGfcAfcGfcCfuUfuGfc(invdT) 95 181 186 psdTdGaGfuUfgGfcAfcGfcCfuUfuGfc(invdT) 99 181 185 pUfsGfaGfuUfgGfcAfcGfcCfuUfuGfc(invdT) 71 181 189 pdTsGfaGfuUfgGfcAfcGfcCfuUfuGfc(invdT) 74 181 188 pdTsdGaGfuUfgGfcAfcGfcCfuUfuGfc(invdT) 64 Example 47: In vivo knock down activity of siRNAs with optimized 5' ends In order to evaluate if the in vitro improvement by optimized 5'-ends transfers to the in vivo situation, we conducted further mouse experiments with GaINAc-palmitoyl conjugates of selected siRNAs (see Table 10).

SiRNAs were administred as under identical conditions as described for the first mouse experiment (example 42, this patent application).

For measurement of FVII levels, plasma samples from mice were prepared by collecting blood (9 volumes) by submandibular bleeding into microcentrifuge tubes containing 0.109 mol/L sodium citrate anticoagulant (1 volume) following standard procedures. FVII activity in plasma was measured with a chromogenic method using a BIOPHEN VII kit (Hyphen BioMed/Aniara, Mason, OH) CA 2976966 2017-08-21 -96following manufacturer's recommendations.

Absorbance of colorimetric development was measured using a Tecan Safire2 microplate reader at 405 nm.

The siRNAs under investigation showed improved in vivo activity, fully correlating with the in vitro screening results. FVII activity in serum was reduced by more than 80% for both siRNAs 48 hours post dosing, compared to 49% using the first generation siRNA design (see Table 10).

This result clearly underscores the importance of a 5'-terminal nucleotide on the antisense strand that can be effectively phosphorylated, in case phosphatases in vivo cleave the synthetically generated 5'-phosphate or 5'-phosphothioate group.

In case of a 5' -terminal 2'0Me nucleotide as used in the initial design or described in the literature as a more potent siRNA design based on in vitro comparison with canonical siRNAs (Allerson et al. J.

Med Chem. 2005, 48, 901-904), the cleavage of the synthetic phosphate in vivo would lead to a strong reduction in potency of the corresponding siRNA.

Table 10: In vivo knockdown activity of siRNAs with optimized 5'ends.

Key:

lower case letters a, c, g, u, are 2'-0-Methyl nucleotides; Upper case letters A, C, G, U followed by "f" indicates a 2'41uoro nucleotide.

Lower case "p" indicates a 5'-phosphate. (invdT) represents an inverted deoxythimidine (3'3'-linked). A phosphorothioate linkages is symbolized with a lower case "s".(NHC6) is the aminohexyl linker incorporated at the 5'-end of the sense strand.GaINAc refers to the structure in formula (IV).

SEQ SEQ remaining ID ID FVII in NO Sense strand sequence (5'-3') NO Antisense strand sequence (5'3') serum GalNAc-(NHC6)179 pGfcAfaAfgGfcGfuGfcCfaAfcUfcAf(invdT) 166 puGfaGfuUfgGfcAfcGfcCfulifuGfc(invdT) 27 GaLNAc-(NHC6)179 pGfcAfaAfgGfcGfuGfcCfaAfcUfcAf(invdT) 190 pusGfaGfulifgGfcAfcGfcCfulAuGfc(invdT) 51 G alNAc-(NHC6)179 pGfcAfaAfgGfcGfuGfcCfaAfcUfcAf(invdT) 185 pUfsGfaGfuUfgGfcAfcGfcCfulifuGfc(invdT) 17 GaINAc-(NHC6)179 pGfcAfaAfgGfcGfuGfcCfaAfcUfcAf(invdT) 186 psdTdGfaGfuUfgGfcAfcGfcCfuUfuGfc(invdT) 13 Example 48: In vitro knock down activity of siRNAs with optimized 3' ends CA 2976966 2017-08-21 -97To further increase activity of the DNase II stable siRNAs an SAR study of the 3'-overhang was performed.

Various combinations of invdT, dTinvdT or dTsdT on either the sense or the antisense strand 3'-overhang were applied to Ahal- and EGFP-targeting siRNAs (see Tables 11 and 12, respectively) and were pair wise compared for composition of both 3'ends in most potent siRNAs.

All siRNAs were screened in dose response (24nM to 0.00037 nM in 4fold dilutions) for knock down activity after transfection of primary mouse hepatocytes (30000 cells/well; 96 well plate format) using Lipofectamine2000 according to the manufacturer's instructions.

Table 11: In vitro knock down activity of EGFP-targeting siRNAs with different 3'-ends.

SEQ ID NO Sense strand sequence (5'-3') l SEQ ID NO lAntisense strand sequence (5'3') I 1050 [nM] .45 GCUGGAGUUC G UGACCGCC dTdT 46 .

GGCGGUCACGAACUCCAGCdTdT 1,0490 212 GcuGGAGuucGuGAccGccdTsdT 225 GGCGGUcACGAACUCcAGCdTsdT *VA 201 gcUfgGfaGfuLlfcGfuGfaCfcGfcCf(invdT) 221 dGsGfcGfgUfcAfcGfaAfcLifcCfaGfc(invd1) 0 A377 -201 qc UfgGfa Gfu UfcGfuGfa CfcGfcCf(invdT) 214 dGsGfcGfqUfcAfcGfaAfcUfcCfaGfcdTsdT 0 ,1479 . 211 gc UfgGfa Gfu UfcGfuGfa CfcGfcCfdT(InvdT) 223 dGs GfcGfgUfsAfc GfaAfc UfcCfaGfcdT(i nvdT) 0,5833 203 gc UfgGfa Gfu Ufc GfuGfa CfcGfcCfdTsdT 214 dGsGfcGrgUlcAfcGfaAfclifcCfaGfcdTsdT 0,2166 204 Gfc Ufg Gfa Gfu Ufc Gfu GfaCfcGfc Cf(i nvdT) 224 pGfsGfcGfglifcAfcGfaAfcLifcCfaGfc(invd1) 09100 -204 GfcUfg GfaGfuUfcGfuGfaCfc GfcCf(i nvdT) 215 pGfs Gfc Gig UfcAfcGfaAfclifcCfaGfc dTsdT 0 2241 , 207 GfcUfg Gfa Gfu Ufc Gfu Gfa Cfc Gfc CfdT(invdT) 215 pGfsGfcGfgUtcAfcGfaAfcUtSCfaGfcdT(invdT) 0 3474 ' 206 Gfclifg Gfa Gfullfc Gfu Gfa Cfc Gfc CfdTsdT 215 pCifsGfcGfglIcAfcGfaAfclifcCfaGfcdTsdT 0 2392 .205 GfscUfgGfaGfuUfcGfuGfaCfcGfcCf(invdT) 220 Gfs Gfc GfgUfcAfcGfaAfc Ufc Cfa Gfc OnvdT) 0,4251 205 GfscUfgGfaGfuUfcGfuGfaCfcGfcCf(invd1) 216 GfsGfcGfgUfcAfcGfaAfc UfcCfaGfcdTsdT 0 2349 210 GfscUfgGfaGfuUfcGfuGfaCfcGfcCfdT(jnvdT) 222 GfsGfc GfgUfcAfcGfaAfc Ufc Cfa Gfc dT(i nvdT) 0 5230 = 209 GfscUfqGfaGfuUfcGfuGfaCfcGfcCfdTs dT 218 GfsGfcGfqUfcAfcGfaAfcUfcCfaGfcdTsdT 0 4937 200 .... gs cUfg Gfa Gfullfc Gfu GfaCfc Gfc Cf(invdT) 217 pdGsGfcGfgUfcAfcGfaAfcUfcCfaGfc(invdT) 0,2643 200 gsclifgGfaGfancGfuGfaCfcGfcCf(invdT) 213 pdGsGfcGfglifcAfcGfaAfclifcCfaGfcdTsdT 06936 208 gscUfgGfaGfuUfcGfuGfaCfcGfcCfdT(invdT) 219 pd G sGfcGfgUfcAfcGfaAfclifcCfa Gfc dT(invdT) 0 3776 202 qscUfgGfaGfuUfcGfuGfaCfcGfcCfdTsdT 213 pd GsGfcGfgUfbAfcGfaAfclifcCfaGfcdTsdT 0 ,1569 Table 12: In vitro knock down activity of Aha I-targeting siRNAs with different 3'-ends.

= SEQ ID NO 1 Sense strand sequence (5'-3') i SEQ ID NO Antisense strand sequence (5-3) i 1050 MI - 157 GGAuGAAGuGGAGAuuAGudTs dT 158 AC uAAUCUCcACUU cAUCC dTsdT 0,094 - 234 GfgAfuGfaAfg Ufg Gfa GfaUfuAfg Uf(i nvdT) 246 Afs CluAfaUfcUfcCfaCfuLlfcAfuCfc(invdT) : 0 ,081 = 234 GfgAfuGfaAfq UfgGfaGfaUfuAlqUf(invdT) 240 Afs CfuAfaUfc lifcCfaCfuUfcAfuCfcdTsdT 0,036 = 233 GfgAfuGfaAfg Ufg Gfa GfaUfuAfg UfdT(i nvdT) 239 Afs CfuAfaUfc lifcCfaCfuUfcAfuCfcdT(OvdT) = 0,034 - 236 GfgAfuGfaAfg UfgGfaGfaUfuAfgUfdTsdT 240 Afs CfuAfaUfclifcCfaCfulAcAfuCfcdTsdT .: 0,040 : 231 GfsgAfuGfaAfgUfgGfaGfaUfuAfgUf(invdT) 241 pAfsCfuAfaUfcUfc CfaCfullfcAfu Cfc 0 nvd1) ', 0,037 = 231 Gfs gAfu GfaAfg Ufg GfaGfaUfuAfq WO nvdT) 267 pAfsCfuAfaUfcUfc CfaCfuUfcAfuCfcdTsdT 0,030 229 Gfs gAfu GfaAfg Ufg GfaGfaUfuAfg UfdT(i nvdT) 268 pAfsCfuAfa Ufc UfcCfaCfullfcAfu Cfc dT(invdT) r 0,024 - 228 Gfs gAfu GfaAfq Ufq GfaGfaUfuAfq UfdTs dT 267 pAfsCruAfa UfclifcCfaCfulifcAfuCfcdTsdT := 0,021 , 232 ggAfuGfaAfgUfgGfaGfaUfuAfgUf(invdT) 245 dAsCfuAfaUfcUfc CfaCfuUfcAfuCfc (invdT) 0.060 232 qqAfu GfaAfq Ufq GfaGfaUfuAfg UkinvdT) 238 dAsCfuAfaUfcUfc CfaCfuUfcAfuCfc dTsdT opal .237 ggAfu GfaAfg UfgGfaGfaUfuAfg UNT(invdT) -244 dAsCfuAfaUfc Ufc CfaCfulifcAfuCfc dT(invdT) 0,045 : 230 g gAfu GfaAfq Ufq GfaGfa UfuAfq UfdTsdT 238 dAsCluAfaUfc Ufc Cfa CfulifcAfu Cfc dTsdT 0,025 : 227 gsgAfuGfaAfgUfgGfaGfaUfuAfgUf(invdT) 243 pdAsCfuAfaUfc Ufc Cfa Cfu UfcAfuCfc(invdT) 0,045 227 q sgAfuGfaAfgUfg Gfa Gfa UfuAfgUf(i nvdT) 266 2dAsCfuAfa Ufc Ufc Cfa Cfu LifcAfuCfcciTsdT 0,015 = 235 gsgAfuGfaAfgUfgGfaGfaUfuAfgUfdT(invdT) 242 pdAsCfuAfaUfc UfcCfaCfuLlfcAfuCfcdT(invdT) 0,039 226 gsgAfuGfaAfgUfgGfaGfaUfuAfgUfdTsdT 266 pdAsCfuAfaUfcUfcCfa Cfu UfcAfuCfcdTsdT 0 .014 It was found, that siRNAs with 2 nucleotide dTsdT-overhangs on the antisense strand performed always better than those with a single invdT overhang at the antisense's 3'end (while sense CA 2976966 2017-08-21 -98strands were the same).

Further benefitial was the combination with a sense strand modified with a single invdT-overhang as 3'overhang.

Example 49: In vivo knock down activity of siRNAs in non-human primates Preparation of DPCs and dosing DPCs were prepared by covalently attaching polymer "149 RAFT" to the indicated siRNA targeting coagulation Factor VII (siF7) at 4:1 wt:wt ratio (polymer: siRNA) through a disulfide linkage and then modifying the polymer-siRNA conjugate with a 2:1 wt:wt mixture of CDMPEG:CDM-NAG at a 7x wt:wt ratio (CDM:polymer).

Cynomolgous monkeys were dosed with 1 mg/kg DPC (polymer weight) and 0.25 mg/kg of the indicated siRNA.

One animal received DPC containing siF7 SEQ ID NO pair 151/152, two animals received DPC containing siF7 SEQ ID NO pair 253/254), #1 and #2), and two animals received DPC containing SEQ ID NO pair 251/255, #1 and #2). F7 values were normalized to the average of the two predose values.

Animals receiving DPCs containing SEQ ID NO pair 253/254 or SEQ ID NO pair 251/255 had greater levels of F7 knockdown and longer PT than the animal receiving SEQ ID NO pair 251/252.

DPC Injection Procedure For each injection procedure, animals were given an IM injection containing a combination of ketaminc (up to 7 mg/kg) and dexmedetomidine (up to 0.03 mg/kg) and moved to a procedure room.

In the procedure room, animals were placed on a water jacketed heating pad and the injection site was shaved and prepped with an antiseptic.

An intravenous catheter (20 to 22 gauge) was inserted into a systemic vein (cephalic or small saphenous) and the DPC solution was infused (2 ml/kg) slowly over 1 to 2 minutes. A pulse oximeter was used to monitor the heart rate and oxygen saturation during and immediately following the injection procedure.

Each injection procedure took about 20 minutes to perform.

After injection the catheter was removed and gentle pressure was applied to the venipuncture site.

Animals were taken back to their cages and given an IM injection of the reversal drug atipamezole (antisedan) (0.10 to 0.15 mg/kg).

Animals were monitored until they regained normal activity.

Blood Collection Procedure CA 2976966 2017-08-21 -99Blood samples (1-5 ml) were obtained for the measurement of gene inhibition (F7 activity, coagulation time), blood chemistries, and markers of liver damage (CBC, chemistry panel, ALT, cytokines, complement).

For these blood collection procedures, animals were given an IM injection containing a combination of ketamine (up to 7 mg/kg) and dexmedetomidine (up to 0.03 mg/kg).

Once sedated, animals were moved on to a portable procedure table and a 22 gauge needle and syringe were used to collect blood from the femoral vein.

Immediately after the blood collection, pressure was applied to the venipuncture site and the blood was divided into the appropriate sample tubes for each blood test.

Animals were then given an IM injection of the reversal drug atipamezole (antisedan) (0.10 to 0.15 mg/kg) and returned to their cage.

No more than 20% of total blood volume was drawn in any 30-day period (estimated blood volume=60 ml/kg).

Each blood collection procedure took about 10 minutes to perform.

Factor VII (F7) activity measurements Blood samples from non-human primates were prepared by filling serum separator tubes with whole blood and allowing the blood to clot at room temperature for at least 20 minutes.

After clotting, blood tubes were centrifuged for 3 minutes at 9000 rpm, aliquoted into eppendorf tubes, and stored at -20 C until assayed. F7 activity in serum was measured with a chromogenic method using a BIOPHEN VII kit (Hyphen BioMed/Aniara, Mason, OH) following manufacturer's recommendations.

Absorbance of colorimetric development was measured using a Tecan Safire2 microplate reader at 405 nm.

Coagulation tests (protime, partial protime and fibrinogen) Blood samples from non-human primates were prepared by completely filling sodium citrate tubes (BD Vacutainer) with whole blood and gently mixing to prevent clot formation.

Tubes were transported to a clinical testing lab within one hour and coagulation assays were performed within 4 hours from the time of collection.

Table 13: FVII dsRNAs used for NHP experiment: Key: lower case letters a, c, g, u, are 2' -0-Methyl nucleotides; Upper case letters A, C, G, U followed by "f indicates a 2'fluoro nucleotide.

Lower case "p" indicates a 5'-phosphate. (invdT) represents an inverted deoxythimidine (3'-3'-linked). A phosphorothioate linkages is symbolized with a lower case "s". dT is deoxythimidine. NH2C6 is the aminohexyl linker incorporated at the 5'-end of the sense strand.

CA 2976966 2017-08-21 -100SEQ ID SEQ ID NO Sense strand sequence (5'-3') NO Antisense strand sequence (5'-3') (NH2C6)GfuUfgGfuGfaAfuGfgAfgCfuCf pCfs UfgAfgCfuCfcAfulifcAfcCfaAfc(i 251 aGf(invdT) 252 nvdT) (NH2C6)GfgUfcCfuGfuUfgUfuGfgUfgAf pAfsUfuCfaCfcAfaCfaAfcAfgGfaCfcd 253 aUf(invdT) 254 TsdT (NH2C6)GfuUfgGfuGfaAfuGfgAfgCfuCf pCfsUfgAfgCfuCfcAfta1fcAfcCfaAfcd 251 aGf(invdT) 255 TsdT Changing from an single nucleotide (invdT)-3'-overhang on both strands to an asymmetric siRNA design with a 3'-(invdT) overhang on the sense strand and a dTsdT overhang on the antisense strand, but otherwise constant modification pattern lead to a more pronounced serum FVII reduction and a significantly prolonged duration of this effect in nonhuman primates (see figure 6a).

This observation is supported by an expected biologic consequence, namely a more pronounced effect on the prothrombin time corresponding to the extent of Factor 7 reduction (see figure 6b).

Example 50: In vivo knock down activity of siRNAs with cleavable RNA linkers In Table 14 the in vivo efficacy based on FVII protein inhibition in serum was compared using cholesterol or the GalNAc-palmitoyl siRNA conjugate in the same sequence context in mice.

The in vivo experiment was conducted as described in example 42. FVII inhibition was strongly decreased for the cholesterol conjugated siRNAs containing no 2'-OH nucleotide compared to the GalNAc-palmitoyl conjugated counterparts (SEQ ID NO pair 179/166 vs.

179/190, SEQ ID NO pair 257/264 vs. SEQ ID NO pair 179/262, SEQ ID NO pair 257/263 vs. SEQ ID NO pair 179/163 and SEQ ID NO pair 257/166 vs. (SEQ ID NO pair 179/166).

In contrast for a 2'-OH containing siRNA the cholesterol conjugate lead to higher FVII inhibition compared to the GalNAc-palmitoyl derivative (SEQ ID NO pair 180/168 vs. SEQ ID NO pair 258/168).

The small molecule ligands GalNAc-palmitoyl and cholesterol used in the described in vivo experiment are connected to the siRNA via a non-cleavable linker to the 5'-end of the sense strand.

In case the sense strand exhibit 2'-OH nucleotides the ligand is still cleavable by nucleases (e.g. DNase II in the endosomal or lysosomal compartment).

The cleavage reaction releases the free siRNA that is then released into the cytoplasm by the endosomal perturbing activity of the delivery polymer.

CA 2976966 2017-08-21 -101For siRNAs lacking a 2'-OH nucleotide in the sense strand, the ligands are stably connected to the duplex, as no enzymatic (nuclease/protease/esterase etc.) or chemical mechanism triggers the cleavage of the ligand.

Therefore, fully stable cholesterol conjugated siRNA can be trapped in cell tnembranes due to the membrane interaction of the lipophilic cholesterol ligand.

Even high concentrations of the siRNA in the tissue is not sufficient for effective release of the siRNA into in the cytoplasm.

In contrast, the less lipophilic GaINAc-palmitoyl conjugated siRNA can be released into the cytoplasm, due to a less pronounced interaction with cell membranes.

For this reason a stable, non-cleavable GalNAc-palmitoyl siRNA conjugate is more efficacious compared to a cholesterol conjugated to the same siRNA.

Developing cleavable linker constructs would help to circumvent the issue of membrane trapping of stably cholesterol conjugated siRNA.

Using disulfide linker chemistry is described as an attractive possibility to introduce a defined cleavage site.

However, cleavage is restricted to reducing organelles within the cell (PNAS, 2006, 103, 13872).

As cleavage is expected to be slow in the endosomal/lysosomal compartment most of the cholesterol-disulfide conjugated siRNA can still be trapped in membranes as described for the non-cleavable cholesterol conjugates.

Table 14 % FVII SEQ ID SEQ ID activity in Conjugate NO pair NO Sequence 5'->3' serum Ga1NAc-NHC6179 pGfcAfaAfgGfcGfuGfcCfaAfcUfcAf(invdT) 27 1 79/1 66 166 puGfaGfuUfgGfcAfcGfcCfuUfuGfc(invdT) GalNAc-NHC6179 pGfcAfaAfgGfcGfuGfcCfaAfcUfcAf(invdT) 51 179/190 190 pusGfaGfuUfgGfcAfcGfcCfuLffuGfc(invdT) GaINAc-NHC6GalNAc 179 pGfcAfaAfgGfcGfuGfcCfaAfcUfcAf(invdT) 17 179/262 262 pUfsGfaGfuUfgGfcAfcGfcCfuUrfuGfc(invdT) GalNAc-NHC6179 pGfcAfaAfgGfcGfuGfcCfaAfcUfcAf(invdT) 13 179/263 263 psdTdGfaGfuUfgGfcAfcGfcCfulifuGfc(invdT) GaLNAc-NHC6180 pGcAAAGGcGuGccAAcucAdTsdT 86 180/168 168 UGAGUUGGcACGCCUUUGCdTsdT Cholesterol Chol257 pGfcAfaAfgGfcGfuGfcCfaAfcUfcAf(invdT) 48 257/166 166 puGfaGfuUfgGfcAfcGfcCfuUfuGfc(invdT) Chol257/190 257 pGfcAfaAfgGfcGfuGfcCfaAfcUfcAf(invdT) 93 CA 2976966 2017-08-21 -102% FVII SEQ ID SEQ ID activity in Conjugate NO pair NO Sequence 5'->3' serum 190 pusGfaGfuUfgGfcAfcGfcCfut1fuGfc(invdT) Chol257 pGfcAfaAfgGfcGfuGfcCfaAfcUfcAf(invdT) 63 257/264 264 pUfsGfaGfuUfgGfcAfcGfcCfuL1fuGfc(invdT) Chol257 pGfcAfaAfgGfcGfuGfcCfaAfcUfcAf(invdT) 41 257/263 263 psdTdGfaGfuUfgGfcAfcGfcCfuUfuGfc(invdT) 258 Chol-pGcAAAGGcGuGccAAcucAdTsdT 258/168 168 UGAGUUGGcACGCCUUUGCSfsdT In addition to the well-established disulfide cleavable linker chemistry another possibility is the generation of defined cleavage sites by using 2'-OH nucleotides at certain positions.

Introduction of 2'-OH nucleotides at selective positions is a new approach to achieve cleavage of the 5 conjugates from RNA strands.

The 2'-OH nucleotides can either be implemented by adding single stranded overhangs with at least one 2'-OH-nucleotide at the 3'- or 5'end of the RNA strand or by using 2'-OH nucleotides within the duplex region of an siRNA.

The enzymatic activity of nucleases present in the endosome / lysosome cleaves selectively at this positions.

In a first design the cholesterol was connected to the sense strand via a single stranded overhang 10 containing 3 2' -OH nucleotides (AUC) at the 5'-terminus.

Cholesterol conjugated siRNAs comparing various cleavable linker chemistries are shown in Table 15.

All siRNAs have the identical sequence context, just the linker chemistry was altered.

Cholesterol was connected to the sense strand via single stranded overhang comprised of a three 15 2'-OH nucleotides (AUC)to the 5'-terminus.

When co-administered with a delivery polymer this siRNA (SEQ ID NO pair 260/263) lead to 77% FVII down modulation in serum in mice, compared to only 60% when using the identical siRNA with a stably attached cholesterol (SEQ ID NO pair 257/263).

The same siRNA with a cholesterol conjugated via a linker according to formula Ia to the 5'-terminus of the sense strand (SEQ ID NO pair 261/263) lead to 93% FVII 20 activity reduction in serum.

All results were achieved by coadministration of 15 mg/kg of a delivery polymer with 2.5 mg/kg of the cholesterol conjugated siRNA in mice.

These results indicate, that the use of a cleavable linker improves the in vivo potency of siRNAs containing no 2'-OH nucleotide.

The cleavable linker can either comprised of 2'-OH containing nucleotides, a di-peptide cleavage motive or a disulfide linker chemistry.

All cleavable linker CA 2976966 2017-08-21 -103constructs improve the in vivo potency in a co-administration setup of a cholesterol conjugated siRNAs with a slow endosomal release delivery polymer.

Table 15 In vivo comparison of various linker chemistries for cholesterol conjugated siRNAs SE SE % FVII ID ID activity in NO Sense strand sequence (5'-3') NO Antisense strand sequence (5'-3') serum CholpGfcAfaAfgGfcGfuGfcCfaAfeUfc A psdIdGfaGfuUfgGfcAfcGfcCfuLlfuGfe 257 f(invdT) 263 (invdT) 40 Chol-C6SSC6pGfcAfaAfgGfcGfuGfcCfaAfcUfeA psdTdGfaGfuUfgGfcAfcGfeCfuUfuGfc 259 f(invdT) 263 (invdT) 22 Chol-AUCpGfcAfaAfgGfcGfuGfcCfaAfcUfc A psdTdGfaGfuUfgGfcAfcGfcCfuUfuGfc 260 f(invdT) 263 (invdT) 23 Chol-Cathepsin-(NHC6)pGfcAfaAfgGfcGfuGfcCfaAfcUfcA psdTdGfaGfuUfgGfcAfcGfcCfaUfuGfc 261 f(invdT) 263 (invdT) 7 Example 51: In vitro serum stability of siRNAs with cleavable linkers The stability of the cleavable linker was evaluated in an in vitro stability assay.

The cholesterol conjugated sense strands were incubated in 90% mouse serum at 37 C for various time points.

The incubation reaction was stopped by addition of proteinase K in a sodium dodecyl sulfate (SDS) containing buffer- The treatment degrades all proteins and enzymes without interfering with the RNA strand integrity. 25 pL of this solution was directly injected onto a AEX-HPLC system connected to a UV detector at 260nm. AEX-HPLC was performed on a Dionex DNA Pac200 column (4x250mm) at 75 C using a 20mM Tris buffer containing 50% ACN at pH=8.

800 mM NaBr in eluent B serves as eluent salt. A gradient from 25 to 62% B in 18 minutes was employed.

The cholesterol containing single stranded RNA elutes from the HPLC column as a broad peak at 260 nm.

After cleavage of the cholesterol sharp symmetric peaks is observed at lower retention time.

Cleavage rate of cholesterol was determined by the following equitation (PA = Peak Area):

CA 2976966 2017-08-21 -104% (free RNA) = 100 * PA [free RNA] / (PA [free RNA] + PA [cholesterol conjugated RNA]) In vitro it was shown, that the 3nt nucleotide (AUC)-overhang is quantitatively cleaved in less than 1 hour in 90% mouse serum.

The cleavage occurs 3' to the two pyrimidine nucleotides in the overhang, leading to two distinct cleavage metabolites (peak areas of metabolites were summarized for data evaluation).

In contrast, the di-peptide containing linker according to formula la, the disulfide and the stably linked cholesterol are fully stable in mouse serum.

Example 52: Tissue distribution of siRNAs with cleavable linkers These results indicate, that the use of a cleavable linker improves the in vivo potency of siRNAs containing no 2'-OH nucleotide.

The cleavable linker can either comprised of 2'-OH containing nucleotides, a di-peptide cleavage motive or a disulfide linker chemistry.

All cleavable linker constructs improve the in vivo potency in a co-administration setup of a cholesterol conjugated siRNAs with a slow endosomal release delivery polymer.

.

Briefly, the siRNA quantification is based on the hybridization of a fluorescently (Atto425) labeled PNA-probe (A1to425-00-TGAGTTGGCACGCCTTT obtained from Panagene Inc, Korea) complementary to the sense strand of the siRNA duplex, followed by AEXHPLC based separation.

Quantification was done by fluorescence detection against an external calibration curve that was generated from a dilution series of the two FVII siRNA used in the in vivo experiment (see example 42).

For plasma samples between 0.2 to 2 1_, and for tissue ¨ lmg aliquots were injected onto the HPLC system.

In Table 16 results from liver tissue analysis are shown.

When analyzing the siRNA content it was found, that the sense strand that is present in liver tissue, is quantitatively cleaved from cholesterol when using either the di-peptide linker motive or the 3 nt 5'overhang with the unmodified linker sequence AUC.

In contrast, only 15% of the disulfide linked siRNA that is present in the liver is cleaved from cholesterol within the first 48 hours post dosing and nothing of the stably attached cholesterol is cleaved from the siRNA.

CA 2976966 2017-08-21 -105When comparing the absolute amounts of cholesterol-free siRNA in liver tissue similar amounts were found for the disulfide linker and for the RNA AUC-linker, nicely correlating with equal FVII serum activity 48 hours post dosing The lower FV1I activity achieved with the di-peptide linked cholesterol siRNA fully correlates with the higher absolute amount of the cleaved cholesterol-free siRNA.

The total amount of cholesterol siRNA conjugate equipped with an (AUC)-linker on the sense strand delivered into the liver is ¨ 6-fold lower as compared to the stably or the disulfide attached cholesterol and ¨ 3-fold lower compared to the di-peptide conjugated cholesterol siRNA.

The reduced tissue presence can be attributed to the fact that the AUC-linker is not only a substrate for intracellular nucleases, but also for nucleases present in circulation as shown in the in vitro incubation with mouse serum.

When the cholesterol ligand is cleaved from the siRNA already in circulation the resulting siRNA is prone to renal clearance and is rapidly excreted into urine without delivery into tissue.

Table 16:

SE SE Total % sense siRNA cleaved ID ID in Liver ligand NO Sense strand sequence (5'-3') NO Antisense strand sequence (5'-3') [ng/g] in liver CholpGfcAfaAfgGfcGfuGfcCfaAfcU psdTdGfaGfuUfgGfcAfcGfcCfuUfuGfc( 25'7 fcAf(invdT) 263 invdT) 5837 0 Chol-C6SSC6GfcAfaAfgGfcGfuGfcCfaAfclif psdTdGfaGfuUfgGfcAfeGfcCfulffuGfe( 259 cAf(invdT) 263 invdT) 4357 14,8 Chol-AUCGfcAfaAfgGfcGfuGfcCfaAfcUf psdTdGfaGfuUfgGfcAfcGfcCfaUfuGfc( 260 cAf(invdT) 263 invdT) 912 96,1 Chol-Cathepsin-(NHC6)pGfeAfaAfgGfcGfuGfcCfaAfcU psdTdGfaGfuUfgGfcAfcGfcCfulifuGfc( 261 fcAf(invdT) 263 invdT) 2760 99,8 Example 53: In vivo knock down activity of siRNAs with cleavable RNA linkers The in vivo experiment was conducted as described in example 50 in mice using cholesterol siRNA conjugates.

In example 50 the cholesterol was connected to the sense strand via a single stranded overhang containing 3 2'-OH nucleotides (AUC) at the 5'-terminus (SEQ ID NO pair 260/263), that showed low serum stability as described in example 51.

This lead to a clearly CA 2976966 2017-08-21 -106reduced tissue concentration compared to serum stable linker chemistries as described in example 52.

Combination of only one or two selected 2'-OH nucleotides together with 2'-0Me nucleotides within the linker lead to higher serum stability, but maintain sensitivity against nucleases present in the endosome / lysosome.

The enzymatic activity of nucleases present in the endosome / lysosome cleaves selectively at the positions of the 2'-OH nucleotides.

The cholesterol conjugated siRNAs comparing various cleavable nucleotide linker motives are summarized in Table 17.

All siRNAs have the identical sequence context, just the linker chemistry was altered.

Cholesterol was connected to the 5'-terminus of the sense strand via single stranded overhang comprised of three or four nucleotides to the, with variable numbers of 2'-OH and 2'-0Me nucleotides.

When co-administered with a delivery polymer all siRNAs lead to FVII down modulation in serum in mice.

The siRNA (SEQ ID NO pair 276/282) lead to 87% FVII activity reduction in serum 48 hours and 95% 168 hours post dosing.

The siRNA (SEQ ID NO pair 277/282) lead to 79% FVII activity reduction in serum 48 hours and 97% 168 hours post dosing.

All results were achieved by co-administration of 15 mg/kg of a delivery polymer with 2.5 mg/kg of the cholesterol conjugated siRNA in mice.

Table 17 In vivo comparison of various nucleotide linker motives for cholesterol conjugated siRNAs SE SE % FVII activity in ID ID serum (48 and NO Sense strand sequence (5'-3') NO Antisense strand sequence (5'-3') 168 hr p.d.) Chol-uAuGfcAfaAfgGfcGfuGfcCfaAfcUfcAf( pdTsGfaGfuUfgGfcAfcGfcCfuUfuGfdT 276 invdT) 282 _sdT 13 and 5 Chol-uGuGfcAfaAfgGfcGfuGfcCfaAfcUfcAf( pdTsGfaGfuUfgGfcAfcGfcCfuLifuGfdT 277 invdT) 282 sdT 21 and 3 Chol-uAuuGfcAfaAfgGfcGfuGfcCfaAfcUfcAf( pdTsGfaGfuUfgGfcAfcGfcCfuUfuGfdT 278 invdT) 282 sdT 19 and 9 Chol-uGuuGfcAfaAfgGfcGfuGfcCfaAfeUfcAf( pdTsGfaGfuUfgGfcAfcGfcCfuUfuGfdT 279 invdT) 282 sdT 20 and 6 Chol-uuAGGfcAfaAfgGfcGfuGfcCfaAfcUfcAf( pdTsGfaGfuUfgGfcAfcGfcCfuUfuGfdT 280 invdT) 282 sdT 46 and 18 Chol-uAGuGfcAfaAfgGfcGfuGfeCfaAfclifeAf( pdTsGfaGfuUfgGfcAfcGfcCfuUfuGfdT 281 invdT) 282 sdT 21 and 6 CA 2976966 2017-08-21 -107These results indicate, that the use of a nucleotide linker that are stable in serum and cleavable in endosome / lysosome further improve the in vivo potency of siRNAs compared to siRNAs with a serum labile linker.

All cleavable linker constructs improve the in vivo potency in a coadministration setup of a cholesterol conjugated siRNAs with a slow endosomal release delivery polymer.

In the following tables siRNAs used in the examples are summarized:

Table 18: Core sequences SEQ ID NO Sense strand sequence (5'-3') SEQ ID NO Antisense strand sequence (5'-3') 1 ACAUGAAGCAGCACGACUU 2 AAGUCGUGCUGCUUCAUGU 3 GCCCGACAACCACUACCUG 4 CAGGUAGUGGUUGUCGGGC 5 CGAGAAGCGCGAUCACAUG 6 CAUGUGAUCGCGCUUCUCG 7 AUAUCAUGGCCGACAAGCA 8 UGCUUGUCGGCCAUGAUAU 9 ACAAGCUGGAGUACAACUA 10 UAGUUGUACUCCAGCUUGU 11 GCAGCUCGCCGACCACUAC 12 GUAGUGGUCGGCGAGCUGC 13 CGUCCAGGAGCGCACCAUC 14 GAUGGUGCGCUCCUGGACG GCUGGAGUUCGUGACCGCC 16 GGCGGUCACGAACUCCAGC 17 CCACCCUGACCUACGGCGU 18 ACGCCGUAGGUCAGGGUGG 19 CGACUUCAAGGAGGACGGC 20 GCCGUCCUCCUUGAAGUCG 71 UUCAAGAUCCGCCACAACA 2/ UGUUGUGGCGGAUCUUGAA 23 GGCAACUACAAGACCCGCG 24 CGCGGGUCUUGUAGUUGCC ./5 CCGGCAAGCUGCCCGUGCC 26 GGCACGGGCAGCUUGCCGG 27 UGCCCAUCCUGGUCGAGCU /8 AGCUCGACCAGGAUGGGCA 29 CAAGUUCAGCGUGUCCGGC 30 GCCGGACACGCUGAACUUG 151 GGAAUCUUAUAUUUGAUCCAA 152 UUGGAUCAAAUAUAAGAUUCCCU 155 GGAUGAAGUGGAGAUUAGU 156 ACUAAUCUCCACUUCAUCC 161 GCAAAGGCGUGCCAACUCA 162 UGAGUUGGCACGCCUUUGC 161 GCA A AGGCGUGCCAACUCA 178 TGAGUUGGCACGCCUUUGC 163 GGAUCAUCUCAAGUCUUAC 164 GUAAGACUUGAGAUGAUCC 171 UGACCACAGUCGGAUUAAA 172 UUUAAUCCGACUGUGGUCA 247 GUUGGUGAAUGGAGCUCAG 248 CUGAGCUCCAUUCACCAAC 249 GGUCCUGUUGUUGGUGAAU 250 AUUCACCAACAACAGGACC 256 UAUGCAAAGGCGUGCCAACUCA 178 TGAGUUGGCACGCCUUUGC 271 UGUGCAAAGGCGUGCCAACUCA 178 TGAGUUGGCACGCCUUUGC 272 UAUUGCAAAGGCGUGCCAACUCA 178 TGAGUUGGCACGCCUUUGC 273 UGUUGCAAAGGCGUGCCAACUCA 178 TGAGUUGGCACGCCUUUGC 274 UUAGGCAAAGGCGUGCCAACUCA 178 TGAGUUGGCACGCCUUUGC _ 275 UAGUGCAAAGGCGUGCCAACUCA 178 TGAGUUGGCACGCCUUUGC CA 2976966 2017-08-21 -108Table 19: Mapping of core sequences and modified sequence Core sequences Modified sequences SEQ Sense strand SEQ SEQ SEQ ID sequence ID Antisense strand ID Sense strand ID Antisense strand NO (5'-3') NO sequence (5'-3') NO sequence (5'-3') NO sequence (5'-3') ACAUGAA AC AUGAAGC GC AGC AC AAGUCGUGCUG AGCACGACU A AGUCGUGCUGC 1 GACUU 2 CUUCAUGU 31 UdTdT 32 UUCAUGUdTdT ACAUGAA AfcAfuGfaAfgC GCAGCAC AAGUCGUGCUG faGfcAfcGfaCfu paAfgU fcGfuGfcUfg 1 GACUU 2 CUUCAUGU 61 UfdTsdT 62 CfuLifc AfuGfudTsdT ACAUGAA dAcdAudGadAg padAgdTcdGudGcdT GC AGC AC AAGUCGUGCUG dCadGcdAcdGa gdCudTcdAudGudTs 1 GACUU 2 CUUCAUGU 91 dCudTdTsdT 92 dT AfdCAfdTGfdA ACAUG AA AfdGCfdAGfdC pdAAfdGUfdCGfdT GC AGCAC AAGUCGUGCUG AfdCGfdACfdT GfdCUfdGCfdTUfd 1 GACUU 2 CUUCAUGU , 121 UfdTsdT 122 CAfdTGfdTdTsdT GCCCG AC GCCCGACA AC AACCACU CAGGUAGUGGU CACUACCUGd CAGGUAGUGGUU 3 ACCUG 4 UGUCGGGC 33 IdT 34 GUCGGGCdTdT GCCCG AC GfcCfcGfaCfaAf AACCACU CAGGUAGUGGU cCfaCfuAfcCfu pcAfgGfuAfgUfgGfu 3 ACCUG 4 UGUCGGGC 63 GfdTsdT 64 UfgUfcGfgGfcdTsdT GCCCGAC dGcdCcdGadCa pcdAgdGudAgdTgd AACCACU CAGGUAGUGGU dAcdCadCudAc GudTgdTcdGgdGcd 3 ACCUG 4 UGUCGGGC 93 dCudGdTsdT 94 TsdT GfdCCfdCGfdA GCCCG AC CfdAAfdCCfdA pdCAfdGGfdTAfdG AACCACU CAGGUAGUGGU CfdTAfdCCfdT UfdGGfdTUfdGUfd 3 ACCUG 4 , UGUCGGGC 123 GfdTsdT 124 CGfdGGfdCdTsdT CGAGAAG CGAGAAGCG CGCGAUC CAUGUGAUCGC CGAUCACAU CAUGUGAUCGCG ACAUG 6 GCUUCUCG 35 GdTdT 36 CUUCUCGdTdT CGAGAAG CfgAfgAfaGfcG CGCGAUC CAUGUGAUCGC fcGfaUfcAfcAfu pcAfuGfuGfaUfcGfc 5 ACAUG 6 GCUUCUCG 65 GfdTsdT 66 GfcUfuCfuCfgdTsdT CGAGAAG dCgdAgdAadGc pcdAudGudGadTcd CGCGAUC CAUGUGAUCGC dGcc1GadTcdAc GcdGcdTudCudCgd 5 ACAUG 6 , GCUUCUCG 95 dAudGdTsdT 96 TsdT CfdGAfdGAfdA CGAGAAG GfdCGfdCGfdA pdCAfdTGfdTGfdA CGCGAUC CAUGUGAUCGC UfdCAfdCAfdT UfdCGfdCGfdCUfd 5 ACAUG 6 GCUUCUCG 125 GfdTsdT 126 TCfdTCfdGdTsdT AUAUCAU AUAUCAUGG GGCCGAC UGCUUGUCGGC CCGACAAGCA UGCUUGUCGGCC 7 AAGCA 8 CAUGAUAU 37 dTdT 38 AUGAUAUdTdT AUAUCAU AfuAfuCfaUfgG GGCCGAC UGCUUGUCGGC fcCfgAfcAfaGfc puGfcilfuGfuCfgGfc 7 AAGCA 8 CAUGAUAU , 67 AfdTsdT 68 CfaUfgAfuAfudTsdT AUAUCAU dAudAudCadTg pudGcdTudGudCgd GGCCGAC UGCU UGUCGGC dGcdCgdAcdAa GcdCadTgdAudAud 7 AAGCA 8 CAUGAUAU 97 dGcdAdTsdT 98 TsdT AUAUCAU UGCUUGUCGGC AfdTAfdTCfdA pdTGfdCUfdTGfdT 7 GGCCGAC 8 CAUGAUAU 127 UfdGGfdCCfdG 128 CfdGGfdCCfdAUfd CA 2976966 2017-08-21 -109Core sequences Modified sequences SEQ Sense strand SEQ SEQ SEQ ID sequence ID Antisense strand ID Sense strand ID Antisense strand NO (5'-3') NO sequence (5'-3') NO sequence (5'-3') NO sequence (5'-3') AAGCA AfdCAfdAGfdC GAfdTAfdTdTsdT AfdTsdT ACAAGCU ACAAGCUGG GGAGUAC UAGUUGUACUC AGUAC A ACU UAGUUGUACUCC 9 AACUA 10 CAGCUUGU 39 , AdIdT 40 AGCUUGUdTdI ACAAGCU AfcAfaGfcUfgG GGAGUAC UAGUUGUACUC faGfuAfcAfaCfu puAfg UfuGfuAfcUfc 9 AACUA 10 CAGCUUGU 69 AfdTsdT 70 CfaGfcUfuGfudTsdT ACAAGCU dAcdAadGcdTg pudAgdTudGudAcd GGAGUAC UAGUUGUACUC dGadGudAcdAa TcdCadGcdTudGudT 9 AACUA 10 CAGCUUGU 99 dCudAdTsdT 100 sdT AfdCAfdAGfdC ACAAGCU UfdGGfdAGfdT pdTAfdGUfdTGfdT GGAGUAC UAGUUGUACUC AfdCAfdACfdT AfdCUfdCCfdAGfd 9 AACUA 10 CAGCUUGU 129 AfdTsdT 130 CUfdTGfdTdTsdT GCAGCUC GC AGCUCGCC GCCGACC GUAGUGGUCGG GACCACUACd GUAGUGGUCGGC 11 ACUAC 12 CGAGCUGC 41 , TdT 42 GAGCUGCdTdT GCAGCUC GfcAfgCfuCfgC GCCGACC GUAGUGGUCGG fcGfaCfcAfcUfa pgUfaGfuGfgUfcGfg 11 ACUAC 12 CGAGCUGC 71 CfdTsdT 72 Cfg A fgCfuGfcdTsdT GCAGCUC dGcdAgdCudCg pgdTadGudGgdTcdG GCCGACC GUAGUGGUCGG dCcdGadCcdAc gdCgdAgdCudGcdTs 11 ACUAC 12 CGAGCUGC 101 dIadCdTsdT 102 dT GfdCAfdGCfdT GCAGCUC CfdGCfdCGfdA pdGUfdAGfdTGfdG GCCGACC GUAGUGGUCGG CfdCAfdCUfdA UfdCGfdGCfdGAfd 11 ACUAC 12 CGAGCUGC 131 CfdTsdT 132 GCfdTGfdCdTsdT CGUCCAG CGUCCA GG A GAGCGCA GAUGGUGCGCU GCGCACCAUC GAUGGUGCGCUC 13 CCAUC 14 CCUGGACG 43 dTdT 44 CUGGACGdTdT CGUCCAG CfgUfcCfaGfgA GAGCGCA GAUGGUGCGCU fgCfgCfaCfcAfu pgAfuGfgUfgCfgCfu 13 CCAUC 14 CCUGGACG 73 CfdTsdT 74 CfcUfgGfaCfgdTsdT CGUCCAG dCgdTcdCadGg pgdAudGgdTgdCgd GAGCGCA GAUGGUGCGCU dAgdCgdCadCc CudCcdTgdGadCgd 13 CCAUC 14 CCUGGACG 103 , dAudCdTsdT 104 TsdT CfdGUfdCCfdA CGUCCAG GfdGAfdGCfdG pdGAfdTGfdGUfdG GAGCGCA GAUGGUGCGCU CfdACfdCAfdT CfdGCfdTCfdCUfdG 13 CCAUC 14 CCUGGACG 133 CfdTsdT 134 GfdACfdGdTsdT GCUGGAG GCUGGAGUU UUCGUGA GGCGGUCACGA CGUGACCGCC GGCGGUCACGAA 15 CCGCC 16 ACUCCAGC 45 dTdT 46 CUCCAGCdTdT GCUGGAG GfcUfgGfaGfuU UUCGUGA GGCGGUCACGA fcGfuGfaCfcGfc pgGfcGfgUfc A fcGfa 15 CCGCC 16 ACUCCAGC 75 CfdTsdT 76 AfcUfcCfaGfcdTsdT GCUGGAG dGcdTgdGadGu pgdGcdGgdTcdAcd UUCGUGA GGCGGUCACGA dTcdGudGadCc GadAcdTcdCadGcdT 15 CCGCC 16 ACUCCAGC 105 , dGcdCdTsdT 106 sdT GCUGGAG GfdCUfdGGfdA pdGGfdCGfdGUfdC UUCGUGA GGCGGUCACGA GfdTUfdCGfdT AfdCGfdAAfdCUfd 15 CCGCC 16 ACUCCAGC 135 GfdACfdCGfdC 136 CCfdAGfdCdTsdT CA 2976966 2017-08-21 - 1 1 0Core sequences Modified sequences SEQ Sense strand SEQ SEQ SEQ ID sequence ID Antisense strand ID Sense strand ID Antisense strand NO (5'-3') NO sequence (5'-3') NO sequence (5'-3') NO sequence (5'-3') CfdTsdT GCUGGAG gscUfgGfaGfuUf pdGsGfcGfgUfcAfc UUCGUGA GGCGGUCACGA cGfuGfaCfcGfc GfaAfcUfcCfaGfcdT 15 CCGCC 16 ACUCCAGC 200 Cf(invdT) 213 sdT GCUGGAG gcUfgGfaGfuUf dGsGfcGfgUfcAfcGf UUCGUGA GGCGGUCACGA cGfuGfaCfcGfc aAfcUfcCfaGfcdTsd 15 CCGCC 16 ACUCCAGC 201 Cf(invdT) 214 T GCUGGAG gscUfgGfaGfuUf pdGsGfcGfgUfcAfc U UCGUGA GGCGGUCACGA cGfuGfaCfcGfc GfaAfcUfcCfaGfcdT 15 CCGCC 16 ACUCCAGC = 202 CfdTsdT 213 sdT GCUGGAG gcUfgGfaGfuUf dGsGfcGfgUfcAfcGf UUCGUGA GGCGGUCACGA cGfuGfaCfcGfc aAfcUfcCfaGfcdTsd 15 CCGCC 16 ACUCCAGC 203 CfdTsdT 214 T GCUGGAG GfcUfgGfaGfuU pGfsGfcGfgUfcAfcG U UCGUGA GGCGGUCACGA fcGfuGfaCfcGfc faAfcUfcCfaGfcdTsd 15 CCGCC 16 ACUCCAGC 204 Cf(invdT) 215 T GCUGGAG GfscUfgGfaGfu GfsGfcGfgUfcAfcGf UUCGUGA GGCGGUCACGA UfcGfuGfaCfcG aAfcUfcCfaGfcdTsd 15 CCGCC 16 ACUCCAGC 205 fcCf(invdT) 216 T GCUGGAG GfcUfgGfaGfuU pGfsGfcGfgUfcAfcG UUCGUGA GGCGGUCACGA fcGfuGfaCfcGfc faAfcUfcCfaGfcdTsd 15 CCGCC 16 ACUCCAGC 206 CfdTsdT 215 T GCUGGAG gscUfgGfaGfuUf pdGsGfcGfgUfcAfc UUCGUGA GGCGGUCACGA cGfuGfaCfcGfc GfaAfcUfcCfaGfc(in 15 CCGCC 16 ACUCCAGC 200 Cf(invdT) 217 vdT) GCUGGAG GfcUfgGfaGfuU pGfsGfcGfgUfcAfcG UUCGUGA GGCGGUCACGA fcGfuGfaCfcGfc faAfcUfcCfaGfcdT(i 15 CCGCC 16 ACUCCAGC 207 CfdT(invdT) 218 nvdT) GCUGGAG gscUfgGfaGfuUf pdGsGfcGfgUfcAfc UUCGUGA GGCGGUCACGA cGfuGfaCfcGfc GfaAfcUfcCfaGfcdT 15 CCGCC 16 ACUCCAGC 208 CfdT(invdT) 219 (invdT) GCUGGAG GfscUfgGfaGfu GfsGfcGfgUfcAfcGf UUCGUGA GGCGGUCACGA UfcGfuGfaCfcG aAfcUfcCfaGfc(invd 15 CCGCC 16 ACUCCAGC 205 fcCf(invdT) 220 T) GCUGGAG gcUfgGfaGfuUf dGsGfcGfgUfcAfcGf UUCGUGA GGCGGUCACGA cGfuGfaCfcGfc aAfcUfcCfaGfc(invd 15 CCGCC 16 ACUCCAGC 201 Cf(invdT) 221 T) GCUGGAG GfscUfgGfaGfu GfsGfcGfgUfcAfcGf UUCGUGA GGCGGUCACGA UfcGfuGfaCfcG aAfcUfcCfaGfcdTsd 15 CCGCC 16 ACUCCAGC 209 fcCfdTsdT 216 T GCUGGAG GfscUfgGfaGfu GfsGfcGfgUfcAfcGf UUCG UGA GGCGGUCACGA UfcGfuGfaCfcG aAfcUfcCfaGfcdT(in 15 CCGCC 16 ACUCCAGC 210 fcCfdT(invdT) 222 vdT) GCUGGAG gcUfgGfaGfuUf dGsGfcGfgUfcAfcGf UUCGUGA GGCGGUCACGA cGfuGfaCfcGfc aAfcUfcCfaGfcdT(in 15 CCGCC 16 ACUCCAGC 211 CfdT(invdT) 223 vdT) GCUGGAG GfeUfgGfaGfuU pGfsGfcGfgUfcAfcG UUCGUGA GGCGGUCACGA fcGfuGfaCfcGfc faAfcUfcCfaGfc(inv 15 CCGCC 16 ACUCCAGC 204 Cf(i nvd T) 224 dT) GCUGGAG GCUGGAGUU UUCGUGA GGCGGUCACGA CGUGACCGCC GGCGGUCACGAA 15 CCGCC 16 ACUCCAGC 45 dTdT 46 CUCCAGCdTdT CA 2976966 2017-08-21 -111Core sequences Modified sequences SEQ Sense strand SEQ SEQ SEQ ID sequence ID Antisense strand ID Sense strand ID Antisense strand NO (5'-3') NO sequence (5'-3') NO , sequence (5'3') NO sequence (5'-3') GCUGGAG GcuGGAGuucG UUCGUGA GGCGGUCACGA uGAccGccdTsd GGCGGUc ACGAA 15 CCGCC 16 ACUCCAGC 212 T 225 CUCcAGCdTsdT CCACCCU CC ACCCUGAC GACCUAC ACGCCGUAGGU CUACGGCGUd ACGCCGUAGGUC 17 GGCGU 18 CAGGGUGG 47 TdT 48 AGGGUGGdTdT CCACCCU CfcAfcCfcUfgAf GACCUAC ACGCCGUAGGU cCfuAfcGfgCfg paCfgCfcGfuAfgGfu 17 GGCGU 18 CAGGGUGG 77 UfdTsdT 78 CfaGfgGfuGfgdTsdT CCACCCU dCcdAcdCcdTgd padCgdCcdGudAgd GACCUAC ACGCCGUAGGU AcdCudAcdGgd GudCadGgdGudGgd 17 GGCGU 18 CAGGGUGG 107 CgdTdTsdT 108 TsdT CfdCAfdCCfdC CCACCCU UfdGAfdCCfdT pdACfdGCfdCGfdT GACCUAC ACGCCGUAGGU AfdCGfdGCfdG AfdGGfdTCfdAGfd 17 GGCGU 18 CAGGGUGG 137 UfdTsdT 138 GGfdTGfdGdTsdT CGACUUC CGACUUCAA AAGGAGG GCCGUCCUCCU GGAGGACGG GCCGUCCUCCUU 19 ACGGC 20 UGAAGUCG 49 CdTdT 50 GA AGUCGdTdT CGACUUC CfgAfcUfuCfaA AAGGAGG GCCGUCCUCCU fgGfaGfgAfcGfg pgCfcGfuCfcUfcCfu 19 ACGGC 20 UGAAGUCG 79 CfdTsdT 80 , UfgAfaGfuCfgdTsdT CGACUUC dCgdAcdTudCa pgdCcdGudCcdTcdC AAGGAGG GCCGUCCUCCU dAgdGadGgdAc udTgdAadGudCgdTs 19 ACGGC 20 UGAAGUCG 109 dGgdCdTsdT 110 dT CfdGAfdCUfdrf CGACUUC CfdAAfdGGfdA pdGCfdCGfdTCfdC AAGGAGG GCCGUCCUCCU GfdGAfdCGfdG UfdCCfdTUfdGAfd 19 ACGGC 20 UGAAGUCG 139 CfdTsdT 140 AGfdTCfdGdTsdT UUCAAGA UUCAAGAUC UCCGCCA UGUUGUGGCGG CGCCACAACA UGUUGUGGCGGA 21 CAACA 22 AUCUUGAA 51 dTdT 52 UCUUGAAdTdT UUCAAGA UfuCfaAfgAfuC UCCGCCA UGUUGUGGCGG fcGfcCfaCfaAfc puGfuUfgUfgGfcGfg 21 CAACA 22 AUCUUGAA 81 AfdTsdT 82 AfuCfuUfgAfadTsdT UUCAAGA dTudCadAgdAu pudGudTgdTgdGcd UCCGCCA UGUUGUGGCGG dCcdGcdCadCad GgdAudCudTgdAad /1 CAACA 22 AUCUUGAA 111 AcdAdTsdT 112 TsdT UfdTCfdAAfdG UUCAAGA AfdTCfdCGfdC pdTGfdTUfdGUfdG UCCGCCA UG U UGUGGCGG CfdACfdAAfdC GfdCGfdGAfdTCfdT 21 CAACA 22 AUCUUGAA 141 AfdTsdT 142 UfdGAfdAdTsdT GGCAACU GGCAAC U AC ACAAGAC CGCGGGUCUUG AAGACCCGCG CGCGGGUCUUGU 23 CCGCG 24 UAGUUGCC 53 dTdT 54 AGUUGCCdTdT GGCAACU GfgCfaAfcUfaCf ACAAGAC CGCGGGUCUUG aAfgAfcCfcGfc pcGfcGfgGfuCfuUfg 23 CCGCG 24 UAGUUGCC 83 GfdTsdT 84 UfaGfuUfgCfcdTsdT GGCAACU dGgdCadAcdTa pcdGcdGgdGudCud ACAAGAC CGCGGGUCUUG dCadAgdAcdCc TgdTadGudTgdCcdT 23 CCGCG 24 UAGUUGCC 113 dCredGdTsdT 114 sdT GGCAACU CGCGGGUCUUG GfdGCfdAAfdC pdCGfdCGfdGGfdT 23 ACAAGAC 24 UAGUUGCC 143 UfdACfdAAfdG 144 CfdTUfdGUfdAGfd CA 2976966 2017-08-21 -112Core sequences Modified sequences SEQ Sense strand SEQ SEQ SEQ ID sequence ID Antisense strand ID Sense strand ID Antisense strand NO (5'-3') NO sequence (5'-3') NO sequence (5'-3') NO sequence (5'-3') CCGCG AfdCCfdCGfdC TUfdGCfdCdTsd'r GfdTsdT CCGGCAA CCGGCAAGCU GCUGCCC GGCACGGGCAG GCCCGUGCCd GGCACGGGCAGC 25 GUGCC , 26 CUUGCCGG 55 TdT 56 UUGCCGG dTdT CCGGCAA CfcGfgCfaAfgCf GCUGCCC GGCACGGGCAG uGfcCfcGfuGfc pgGfcAfcGfgGfcAfg 25 GUGCC 26 CUUGCCGG 85 CfdTsdT 86 CfuUfgCfcGfgdTsdT CCGGC AA dCcdGgdCadAg pgdGcdAcdGgdGcd GCUGCCC GGC A CGGGC AG dCudGcdCcdGu AgdCudTgdCcdGgd 25 GUGCC 26 CUUGCCGG , 115 dGcdCdTsdT 116 TsdT CfdCGfdGCfdA CCGGCAA AfdGCfdTGfdC pdGGfdCAfdCGfdG GCUGCCC GGCACGGGCAG CfdCGfdTGfdC GfdCAfdGCfdTUfd 25 GUGCC 26 CUUGCCGG 145 CfdTsdT 146 GCfdCGfdGdTsdT UGCCCAU UGCCCAUCCU CCUGGUC AGCUCGACCAG GGUCGAGCUd AGCUCGACCAGG 27 GAGCU 28 GAUGGGCA 57 TdT 58 A UGGGCAdTd'r UGCCCAU UfgCfcCfaUfcCf CCUGGUC AGCUCGACCAG uGfgUfcGfaGfc paGfcUfcGfaCfcAfg 27 GAGCU 28 GAUGGGCA 87 UfdTsdT 88 GfaUfgGfgCfadTsdT UGCCCAU dTgdCcdCadTcd padGcdTcdGadCcdA CCUGGUC AGCUCGACCAG CudGgdTcdGad gdGadTgdGgdCadTs 27 GAGCU , 28 GAUGGGCA 117 GcdTdTsdI 118 dT UfdGCfdCCfdA UGCCCAU UfdCCfd'rGfdG pdAGfdCUfdCGfdA CCUGGUC AGCUCGACCAG UfdCGfdAGfdC CfdCAfdGGfdAUfd 27 GAGCU 28 GAUGGGCA 147 UfdTsdT 148 GGfdGCfdAdTsdT CAAGUUC CAAGUUCAG AGCGUGU GCCGGACACGC CGUGUCCGGC GCCGGACACGCU 29 CCGGC 30 UGAACUUG 59 dTdT 60 GAACUUGdTdT CAAGU UC CfaAfgUfuCfaG AGCGUGU GCCGGACACGC fcGfuGfuCfcGfg pgCfcGfg Afc AfcGfc 29 CCGGC 30 UGAACUUG 89 CfdTsdT 90 UfgAfaCfuUfgdTsdT CAAGUUC dCadAgdTudCa pgdCcdGgdAcdAcd AGCGUGU GCCGGACACGC dGcdGudGudCc GcdTgdAadCu dTgd 29 CCGGC 30 UGAACUUG 119 dGgdCdTsdT 120 TsdT CfdAAfdGUfdT CAAGUUC CfdAGfdCGfdT pdGCfdCGfdGAfdC AGCGUGU GCCGGACACGC GfdTCfdCGfdG AfdCGfdCUfdGAfd 29 CCGGC 30 UGAACUUG 149 CfdTsdT 150 ACfdTUfdGdTsdT GGAAUCU UUGGAUCAAAU GGAAUCuuAu UAUAUUU AUAAGAU UCCC AuunGA U Cc As uuGGAUcAAAuAu 151 GAUCC A A 152 U 153 A 154 A AGAuUCc sc sU CholGGAAUCU UUGGAUCA A AU pGGAAUCuuAu UAUAUUU AUAAGAUUCCC AuuuG AUCc As uuGG AUcAAAuAu 151 GAUCCAA 152 U 265 A 154 AAGAuUCcscsU GGAUGAA GGAuGAAGuG GUGGAGA ACUAAUCUCCA GAGAuuAGudT ACuAAUCUCcACU 155 UUAGU 156 CUUCAUCC 157 sdT 158 UcAUCCdTsdT GGAUGAA ACUAAUCUCCA (NH2C6)GfgAfu pasCfuAfaUfcUfcCfa 155 GUGGAGA 156 CUUCAUCC 160 GfaAfgUfgGfaG 159 Cfu UfcAfuCfc(i nvdT CA 2976966 2017-08-21 -113Core sequences Modified sequences SEQ Sense strand SEQ SEQ SEQ ID sequence ID Antisense strand ID Sense strand ID Antisense strand NO (5'-3') NO sequence (5'-3') NO sequence (5'-3') NO sequence (5'-3') UUAGU fa U fuAfgUf(inv ) dT) GGAUGAA gsgAfuGfaAfgU pdAsCfuAfaUfcUfcC GUGGAGA ACUAAUCUCCA fgGfaGfaUfuAfg faCfuUfcAfuCfcdTsd 155 UUAGU 156 CUUCAUCC 226 UfdTsdT 266 'I' GGAUGAA gsgAfuGfaAfgU pdAsCfuAfaUfcUfcC GUGGAGA ACUAAUCUCCA fgGfaGfaUfuAfg faCfuUfcAfuCfcdTsd 155 UUAGU 156 CUUCAUCC 227 Uf(invdT) 266 T GGAUGAA GfsgAfuGfaAfg pAfsCfuAfaUfcUfcC GUGGAGA ACUAAUCUCCA UfgGfaGfaUfuA faCfuUfcAfuCfcdTsd 155 UUAGU 156 CUUCAUCC 228 , fgUfdTsdT 267 T GGAUGAA GfsgAfuGfaAfg pAfsCfuAfaUfcUfcC GUGGAGA ACUAAUCUCCA UfgGfaGfaUfuA faCfuUfcAfuCfcd'r(i 155 UUAGU 156 CUUCAUCC 229 fgUfdT(invdT) 268 nvdT) GGAUGAA ggAfuGfaAfgUf dAsCfuAfaUfcUfcCf GUGGAGA ACUAAUCUCCA gGfaGfaUfuAfg aCfuUfcAfuCfcdTsd 155 UUAGU 156 CUUCAUCC 230 UfdTsdT 238 T GGAUGAA GfsgAfuGfaAfg pAfsCfuAfaUfcUfcC GUGGAGA ACUAAUCUCCA UfgGfaGfaUfuA faCfuUfcAfuCfcdTsd 155 UUAGU 156 CUUCAUCC 231 fgUf(invdT) 267 T GGAUGAA ggAfuGfaAfgUf dAsCfuAfaUfcUfcCf GUGGAGA ACUAAUCUCCA gGfaGfaUfuAfg aCfuUfcAfuCfcdTsd 155 UUAGU 156 CUUCAUCC 232 Uf(invdT) 238 T GGAUGAA GfgAfuGfaAfgU AfsCfuAfaUfcUfcCf GUGGAGA ACUAAUCUCCA fgGfaGfaUfuAfg aCfuUfcAfuCfcdT(in 155 UUAGU 156 CUUCAUCC 233 UfdT(invdT) 239 vdT) GGAUGAA GfgAfuGfaAfgU AfsCfuAfaUfcUfcCf GUGGAGA ACUAAUCUCCA fgGfaGfaUfuAfg aCfuUfcAfuCfcdTsd 155 UUAGU 156 CUUCAUCC 234 Uf(invdT) 240 T GGAUGAA GfsgAfuGfaAfg pAfsCfuAfaUfcUfcC GUGGAGA ACUAAUCUCCA UfgGfaGfaUfuA faCfuUfcAfuCfc(inv 155 UUAGU 156 CUUCAUCC 231 fg Uf(invdT) 241 dT) GGAUGAA GfgAfuGfaAfgU pasCfuAfaUfcUfcCfa GUGGAGA ACUAAUCUCCA fgGfaGfaUfuAfg CfuUfcAfuCfc(invdT 155 UUAGU 156 CUUCAUCC 234 Uf(invdT) 159 ) GGAUGAA gsgAfuGfaAfg U pdAsCfuAfaUfcUfcC GUGGAGA ACUAAUCUCCA fgGfaGfaUfuAfg faCfuUfcAfuCfcdT(i 155 UUAGU 156 , CUUCAUCC 235 UfdT(invdT) 242 nvdT) GGAUGAA Gfg AfuGfa A fgU AfsCfuAfaUfcUfcCf GUGGAGA ACUAAUCUCCA fgGfaGfaUfuAfg aCfuUfcAfuCfcdTsd 155 UUAGU 156 CUUCAUCC 236 UfdTsdT 240 T GGAUGAA gsgAfuGfaAfgU pdAsCfuAfaUfcUfcC GUGGAGA ACUAAUCUCCA fgGfaGfaUfuAfg faCfuUfcAfuCfc (inv 155 UUAGU 156 CUUCAUCC 227 Uf(invdT) 243 dT) GGAUGAA ggAfuGfaAfgUf dAsCfuAfaUfcUfcCf GUGGAGA ACUAAUCUCCA gGfaGfaUfuAfg aCfuUfcAfuCfcdT(in 155 UUAGU 156 CUUCAUCC 237 _ UfdT(invdT) 244 vdT) GGAUGAA ggAfuGfaAfgUf dAsCfuAfaUfcUfcCf GUGGAGA ACUAAUCUCCA gGfaGfaUfuAfg aCfuUfcAfuCfc(invd . 155 UUAGU 156 CUUCAUCC 232 .

Uf(invdT) 245 T) GGAUGAA GfgAfuGfaAfgU AfsCfuAfaUfcUfcCf GUGGAGA ACUAAUCUCCA fgGfaGfaUfuAfg aCfutifcAfuCfc(invd 155 UUAGU 156 CUUCAUCC 234 Uf(invdT) 246 T) CA 2976966 2017-08-21 -114Core sequences Modified sequences SEQ Sense strand SEQ SEQ SEQ ID sequence ID Antisense strand ID Sense strand ID Antisense strand NO (5'-3') NO sequence (5'-3') NO sequence (5'-3') NO sequence (5'-3') GGA UGAA GGAuGAAGuG GUGG AGA ACUAAUCUCC A GAG AuuAGudT ACuAAUCUCcACU 155 UUAGU 156 CUUCAUCC 157 . sdT _ 158 UcAUCCdTsdT (NH2C6)GfcAfa GCAAAGG AfgG fcGfuG fcC puGfaGfuUfgGfcAfc CGUGCCA UGAGUUGGCAC faAfcUfcAf(invd GfcCfuUfuGfc(invdT 161 ACUCA 162 GCCUUUGC 165 T) 166 ) GCAAAGG (NH2C6)GcAA CGUGCCA UGAGUUGGCAC AGGcGuGccAA UGAGUUGGcACG 161 ACUCA 162 GCCUUUGC 167 cucAdTsdT 168 CCUUUGCdTsdT GCAAAGG UGAGUUGGCAC GalNAcCGUGCCA GCCUUUGC (NHC6)ACUCA pGfcAfaAfgGfc puGfaGfuUfgGfcAfc GfuGfcCfaAfc U GfcCfuUfuGfc(invdT 161 162 179 fcAf(invdT) 166 ) GCAAAGG UGAGU UGGCAC GalNAc- UGAGU UGGcACG CGUGCCA GCCUUUGC (NHC6)- CCUUUGCdTsdT ACUCA pGcAAAGGcGu GccAAcucAdTs 161 162 180 dT 168 GCAAAGG GfcAfaAfgGfcG pUfsGfaGfuUfgGfc A CGUGCCA UG AGIJUOGC AC fuGfcCfaAfctJfc fcGfcCfuUfuGfc(inv 161 ACUCA 162 GCCUUUGC 181 Af(invdT) 185 dT) GCAAAGG GfcAfaAfgGfcG psdTdGaGfuUfgGfc CGUGCCA TGAGUUGGCAC fuGfcCfaAfcUfc AfcGfcCfuUfuGfc(in 161 ACUCA 178 GCCUUUGC 181 Af(invdT) 186 vdT) GCAAAGG GfcAfaAfgGfcG psdTsGfaGfuUfgGfc CGUGCCA TGAGUUGGCAC fuGfcCfaAfcUfc AfcGfcCfuUfuGfc(in 161 ACUCA 178 GCCUUUGC 181 Af(invdT) 187 vdT) GCAAAGG GfcAfaAfgGfcG pdTsdGaGfuUfgGfc CGUGCCA TGAGU UGGCAC fuGfcCfaAfcUfc AfcGfcCfuUfuGfc(in 161 ACUCA 178 GCCUUUGC 181 Af(invdT) 188 vdT) GCAAAGG GfcAfaAfgGfcG pdTsGfaGfuUfgGfc CGUGCCA TGAGUUGGCAC fuGfcCfaAfcUfc AfcGfcCfuUfuGfc(in 161 ACUCA 178 GCCUUUGC 181 Af(invdT) 189 vdT) GCAAAGG GfcAfaAfgGfcG pusGfaGfuUfgGfcAf CGUGCCA UGAGUUGGCAC fuGfcCfaAfcUfc cGfcCfuUfuGfc(invd 161 , ACUCA 162 GCCUUUGC 181 Af(invdT) 190 T) _ GCAAAGG GfcAfaAfgGfcG puGfaGfuUfgGfcAfc CGUGCCA UGAGUUGGCAC fuGfcCfaAfcUfc GfcCfuUfuGfc(invdT 161 ACUCA 162 GCCUUUGC 181 Af(invdT) 166 ) GCAAAGG GfcAfaAfgGfcG psUfGfaGfuUfgGfc A CGUGCCA UGAGUUGGCAC fuGfcCfaAfcUfc fcGfcCfuUfuGfc(inv 161 ACUCA 162 GCCUUUGC 181 Af(invdT) 191 dT) GCAAAGG GfcAfaAfgGfcG UfsGfaGfuUfgGfcAf CGUGCCA UGAGUUGGCAC fuGfcCfaAfcUfc cGfcCfutifuGfc(invd . 161 ACUCA 162 GCCUUUGC 181 Af(invdT) 192 T) GCAAAGG GfcAfaAfgGfcG ps UfsGfaGfuUfgGfc CGUGCCA UGAGUUGGCAC fuGfcCfaAfcUfc AfcGfcCfuUfuGfc(in 161 ACUCA 162 GCCUUUGC 181 Af(invdT) 193 vdT) GCAAAGG GfcAfaAfgGfcG psdTsdGaGfuUfgGfc CGUGCCA TGAGUUGGCAC fuGfcCfaAfc U fc AfcGfcCfuUfuGfc(in 161 ACUCA 178 GCCUUUGC 181 _ Af(invdT) 194 vdT) CA 2976966 2017-08-21 -115Core sequences Modified sequences SEQ Sense strand SEQ SEQ SEQ ID sequence ID Antisense strand ID Sense strand ID Antisense strand NO (5'-3') NO sequence (5'-3') NO sequence (5'-3') NO sequence (5'-3') GCAAAGG GfcAfaAfgGfcG usGfaGfuUfgGfcAfc CGUGCCA UGAGUUGGCAC fuGfcCfaAfcUfc GfcCfuUfuGfc(invdT 161 ACUCA 162 GCCUUUGC 181 Af(invdT) 195 ) GCAAAGG GfcAfaAfgGfcG uGfaGfuUfgGfcAfc CGUGCCA UGAGUUGGCAC fuGfcCfaAfcUfc GfcCfuUfuGfc(invdT 161 ACUCA 162 GCCUUUGC 181 Af(invdT) 196 ) GCAAAGG GfcAfaAfgGfcG dTsGfaGfuUfgGfcAf CGUGCCA TGAGUUGGCAC fuGfcCfaAfcUfc cGfcCfuUfuGfc(invd 161 ACUCA 178 GCCUUUGC 181 Af(invdT) 197 T) GCAAAGG GfcAfaAfgGfcG psdTGfaGfuUfgGfc CGUGCCA TGAGUUGGCAC fuGfcCfaAfcUfc AfcGfcCfuUfuGfc(in 161 ACUCA 178 GCCUUUGC 181 Af(invdT) 198 vdT) GCAAAGG GfcAfaAfgGfcG dTsdGaGfuUfgGfcA CGUGCCA TGAGUUGGCAC fuGfcCfaAfcUfc fcGfcCfuUfuGfc(inv 161 ACUCA 178 GCCUUUGC 181 Af(invdT) 199 dT) CholGCAAAGG pGfcAfaAfgGfc pusGfaGfuUfgGfcAf CGUGCCA UGAGUUGGCAC GfuGfcCfaAfcU cGfcCfuUfuGfc(invd 161 ACUCA 162 GCCUUUGC 183 fcAf(invdT) 190 T) GCAAAGG GfcAfaAfgGfcG uGfaGfuUfgGfcAfc CGUGCCA UGAGUUGGCAC fuGfcCfaAfcUfc GfcCfutifuGfc(invdT 161 ACUCA 162 GCCUUUGC 181 Af(invdT) 196 ) GCAAAGG GfcAfaAfgGfcG usGfaGfuUfgGfcAfc CGUGCCA UGAGUUGGCAC fuGfcCfaAfcUfc GfcCfuUfuGfc(invdT 161 ACUCA 162 GCCUUUGC 181 Af(invdT) 195 ) GCAAAGG GfcAfaAfgGfcG UfsGfaGfuUfgGfcAf CGUGCCA UGAGUUGGCAC fuGfcCfaAfcUfc cGfcCfulffuGfc(invd 161 ACUCA 162 GCCUUUGC 181 Af(invdT) 192 T) GCAAAGG GfcAfaAfgGfcG dTsGfaGfutJfgGfcAf CGUGCCA TGAGUUGGCAC fuGfcCfaAfcUfc cGfcCfuUfuGfc(invd 161 ACUCA 178 GCCUUUGC 181 Af(invdT) 197 T) GCAAAGG GfcAfaAfgGfcG dTsdGaGfuUfgGfcA CGUGCCA TGAGUUGGCAC fuGfcCfaAfcUfc fcGfcCfuUfuGfc(inv 161 ACUCA 178 GCCUUUGC 181 Af(invdT) 199 dT) GCAAAGG GcAAAGGcGu CGUGCCA UGAGUUGGCAC GccAAcucAdTs UGAGUUGGcACG 161 ACUCA 162 GCCUUUGC 182 dT 168 CCUUUGCdTsdT GCAAAGG GfcAfaAfgGfcG CGUGCCA UGAGUUGGCAC fuGfcCfaAfcUfc UGAGUUGGcACG 161 ACUCA 162 GCCUUUGC 184 A 168 CCUUUGCdTsdT GaINAc(NHC6)GCAAAGG pGfcAfaAfgGfc puGfaGfuUfgGfcAfc CGUGCCA UGAGUUGGCAC GfuGfcCfaAfcU GfcCfuUfuGfc(invdT 161 ACUCA 162 GCCUUUGC 179 fcAf(invdT) 166 ) GaINAc(NHC6)GCAAAGG pGfcAfaAfgGfc pusGfaGfuUfgGfcAf CGUGCCA UGAGUUGGCAC GfuGfcCfaAfcU cGfcCfuUfuGfc(invd 161 ACUCA 162 GCCUUUGC 179 fcAf(invdT) 190 T) GCAAAGG GaINAc- pUfsGfaGfuUfgGfcA CGUGCCA UGAGUUGGCAC (NHC6)- fcGfcCfuUfuGfc(inv 161 ACUCA 162 GCCUUUGC 179 pGfcAfaAfgGfc 185 dT) CA 2976966 2017-08-21 -116Core sequences Modified sequences SEQ Sense strand SEQ SEQ SEQ ID sequence ID Antisense strand ID Sense strand ID Antisense strand NO (5'-3') NO sequence (5'-3') NO sequence (5'-3') NO sequence (5'-3') GfuGfcCfaAfcU fcAf(invdT) GalNAc(NHC6)GCAAAGG pGfcAfaAfgGfc psdTdGfaGfuUfgGfc CGUGCCA TGAGUUGGCAC GfuGfcCfaAfcU AfcGfcCfuUfuGfc(in 161 ACUCA 178 GCCUUUGC 179 fcAf(invdT) 186 vdT) GaIN Ac - (NHC6)GC AAAGG pGfcAfaAfgGfc puGfaGfuUfgGfcAfc CGUGCCA UGAGUUGGCAC GfuGfcCfaAfcU GfcCfuUfuGfc(invdT 161 ACUCA 162 GCCUUUGC 179 fcAf(invdT) 166 ) GalNAc(NHC6)GCAAAGG pGfcAfaAfgGfc pusGfaGfuUfgGfcAf CGUGCCA UGAGUUGGCAC GfuGfcCfaAfcU cGfcCfuUfuGfc(invd 161 ACUCA 162 GCCUUUGC 179 fcAf(invdT) 190 T) GaINAc(NHC6)GC AA AGG pGfcAfaAfgGfc pUfsGfaGfuUfgGfc A CGUGCCA UGAGUUGGCAC GfuGfcCfaAfcU fcGfcCfuUfuGfc(inv 161 ACUCA 162 GCCUUUGC 179 fcAf(invdT) 262 dT) GaINAc(NHC6)GCAAAGG pGfcAfaAfgGfc psdTdGfaGfuUfgGfc CGUGCCA TGAGUUGGCAC GfuGfcCfaAfc U AfcGfcCfuUfuGfc(in 161 ACUCA 178 GCCUUUGC 179 fcAf(invdT) 263 vdT) CholGCAAAGG pGfcAfaAfgGfc puGfaGfuUfgGfcAfc CGUGCCA UGAGUUGGCAC GfuGfcCfaAfcU GfcCfuUfuGfc(invdT 161 ACUCA 162 GCCUUUGC 257 fcAf(invdT) 166 ) CholGCAAAGG pGfcAfaAfgGfc pusGfaGfuUfgGfcAf CGUGCCA UGAGUUGGCAC GfuGfcCfaAfcU cGfcCfu UfuGfc (invd 161 ACUCA 162 GCCUUUGC 257 fcAf( invdT) 190 T) CholGCAAAGG pGfcAfaAfgGfc p U fsGfaGfuU fgGfc A CGUGCCA UGAGUUGGCAC GfuGfcCfaAfcU fcGfcCfuUfuGfc(inv 161 ACUCA 162 GCCUUUGC 257 fcAf(invdT) 264 dT) CholGCAAAGG pGfcAfaAfgGfc psdTdGfaGfuUfgGfc CGUGCCA TGAGUUGGCAC GfuGfcCfaAfcU AfcGfcCfuUfuGfc(in 161 ACUCA 178 GCCUUUGC 257 fcAf(invdT) 263 vdT) CholGCAAAGG pGcAAAGGcGu CGUGCCA UGAGUUGGCAC GccAAcucAdTs UGAGUUGGcACG 161 ACUCA 162 GCCUUUGC 258 dT 168 CCUUUGCdTsdT CholGCAAAGG pGfcAfaAfgGfc psdTdGfaGfuUfgGfc CGUGCCA TGAGUUGGCAC GfuGfcCfaAfcU AfcGfcCfuUfuGfc(in 161 ACUCA 178 GCCUUUGC 257 fcAf(invdT) 263 vdT) GCAAAGG TGAGUUGGCAC Chol-C6S SC6- psdTdGfaGfuUfgGfc 161 CGUGCCA 178 GCCUUUGC 259 pGfcAfaAfgGfc 263 AfcGfcCfuUfuGfc(in CA 2976966 2017-08-21 -117Core sequences Modified sequences SEQ Sense strand SEQ SEQ SEQ ID sequence ID Antisense strand ID Sense strand ID Antisense strand NO (5'-3') NO sequence (5'-3') NO sequence (5'-3') NO sequence (5'-3') ACUCA GfuGfcCfaAfcU vdT) fcAf(invdT) Chol-AUCGCA A AGG pGfcAfaAfgGfc psdTdGfaGfuUfgGfc CGUGCCA TGAGU UGGCAC GfuGfcCfaAfcU AfcGfcCfuUfuGfc(in 161 ACUCA 178 GCCUUUGC 260 fcAf(invdT) 263 vdT) Chol-Cathepsin(NHC6)GCAAAGG pGfcAfaAfgGfc psdTdGfaGfuUfgGfc CGUGCCA TGAGUUGGCAC GfuGfcCfaAfcU AfcGfcCfuUfuGfc(in 161 ACUCA 178 GCCUUUGC 261 fcAf(invdT) 263 vdT) CholGCAAAGG pGfcAfaAfgGfc psdTdGfaGfuUfgGfc CGUGCCA TGAGUUGGCAC GfuGfcCfaAfcU AfcGfcCfuUfuGfc(in 161 ACUCA 178 GCCUUUGC 257 fcAf(invdT) 263 vdT) Chol-C6S SC6GCAAAGG pGfcAfaAfgGfc psdTdGfaGfuUfgGfc CGUGCCA TGAGUUGGCAC GfuGfcCfaAfcU AfcGfcCfuUfuGfc(in 161 ACUCA 178 GCCUUUGC 259 fcAf(invdT) 263 vdT) Chol-AUCGCAAAGG pGfcAfaAfgGfc psdTdGfaGfuUfgGfc CGUGCCA TGAGUUGGCAC GfuGfcCfaAfcU AfcGfcCfuUfuGfc(in 161 ACUCA 178 GCCUUUGC 260 fcAf(invdT) 263 vdT) Chol-Cathepsin(NHC6)GCAAAGG pGfcAfaAfgGfc psdTdGfaGfuUfgGfc CGUGCCA TGAGUUGGCAC GfuGfcCfaAfcU AfcGfcCfuUfuGfc(in 161 ACUCA 178 GCCUUUGC 261 fcAf(invdT) 263 vdT) (NH2C6)GGAUf GGAUCAU CfAUfCfUfCfA GUfAAGACtUfUfG CUCA AGU GUAAGACUUGA AGUfCfUfUfAC AGAUfGAUfCfCfd 163 CUUAC 164 GAUGAUCC 169 fdTsdT 170 TsdT UGACCAC UUUAAUCCGAC (NH2C6)Ufg A fc pusUfuAfaUfcCfg A f AGUCGGA UGUGGUCA CfaCfaGfuCfgGf cUfgUfgGfuCfa(invd UUAAA aUfuAfaAf(invd T) 171 172 173 T) 174 UGACCAC (NH2C6)uGAcc AGUCGGA UUUAAUCCGAC AcAGucGGAuu puUuAAUCCGACU 171 UUAAA 172 UGUGGUCA 175 AAAdTsdT 176 GUGGucAdTsdT UGACCAC (NH2C6)uGAcc AGUCGGA UUUAAUCCGAC AcAGucGGAuu UUuAAUCCGACU 171 UUAAA 172 UGUGGUCA 175 AAAdTsdT 177 GUGGUcAdTsdT (NH2C6)GfuUfg GUUGGUG GfuGfaAfuGfgA pCfsUfgAfgCfuCfcA AA UGGAG CUGAGCUCCAU fgCfuCfaGf(invd fullfc AfcCfaAfc(inv 247 CUCAG 248 UCACCA AC 251 T) 252 dT) (NH2C6)GfuUfg GUUGGUG GfuGfaAfuGfg A pCfsUfg AfgCfuCfc A AAUGGAG CUGAGCUCCAU fgCfuCfaGf(invd fuUfcAfcCfaAfcdTsd 247 CUCAG 248 UCACCAAC 251 T) 255 T GGUCCUG (NH2C6)GfgUfc pAfsUfuCfaCfcAfaC UUGUUGG AUUCACCAACA CfuGfuUfgUfuG faAfcAfgGfaCfcdTsd 249 UGAAU 250 ACAGGACC 253 fgUfgAfaUf(inv 254 T CA 2976966 2017-08-21 -118Core sequences Modified sequences SEQ Sense strand SEQ SEQ SEQ ID sequence ID Antisense strand ID Sense strand ID Antisense strand NO (5'-3') NO sequence (5'-3') NO sequence (5'-3') NO sequence (5'-3') dT) UAUGCAA Chol-uAuAGGCGUG GfcAfaAfgGfcG pdTsGfaGfuUfgGfc CCAACUC TGAGUUGGCAC fuGfcCfaAfcUfc AfcGfcCfuUfuGfcdT 256 A 178 GCCUUUGC 276 Af(invdT) 282 sdT UGUGCAA Chol-uGuAGGCGUG GfcAfaAfgGfcG pdTsGfaGfuUfgGfc CCAACUC TGAGUUGGCAC fuGfcCfaAfcUfc AfcGfcCfuUfuGfcdT 271 A 178 GCCUUUGC 277 Af(invdT) 282 sdT UAUUGCA Chol-uAuuAAGGCGU GfcAfaAfgGfcG pdTsGfaGfuUfgGfc GCCAACU TGAGUUGGCAC fuGfcCfaAfcUfc AfcGfcCfuUfuGfcdT 272 CA 178 GCCUUUGC 278 Af(invdT) 282 sdT UGUUGCA Chol-uGuuAAGGCGU GfcAfaAfgGfcG pdTsGfaGfuUfgGfc GCCAACU TGAGUUGGCAC fuGfcCfaAfcUfc AfcGfcCfuUfuGfcdT 273 CA 178 GCCUUUGC 279 Af(invdT) 282 sdT UUAGGCA Chol-uuAGAAGGCGU GfcAfaAfgGfcG pdTsGfaGfuUfgGfc GCCAACU TGAGUUGGCAC fuGfcCfaAfcUfc AfcGfcCfuUfuGfcdT 274 CA 178 GCCUUUGC 280 Af(invdT) 282 sdT UAGUGCA Chol-uAGuAAGGCGU GfcAfaAfgGfcG pdTsGfaGfuUfgGfc GCCAACU TGAGUUGGCAC fuGfcCfaAfcUfc AfcGfcCfuUfuGfcdT 275 CA 178 GCCUUUGC 281 Af(invdT) 282 sdT CA 2976966 2017-08-21