Compositions and methods for modulating PKK expression

General Method for the Preparation of Phosphoramidites, Compounds 1, 1a and 2

Bx is a heterocyclic base;

Compounds 1, 1a and 2 were prepared as per the procedures well known in the art as described in the specification herein (see Seth et al., Bioorg. Med. Chem., 2011, 21(4), 1122-1125, J. Org. Chem., 2010, 75(5), 1569-1581, Nucleic Acids Symposium Series, 2008, 52(1), 553-554); and also see published PCT International Applications (WO 2011/115818, WO 2010/077578, WO2010/036698, WO2009/143369, WO 2009/006478, and WO 2007/090071), and U.S. Pat. No. 7,569,686).

Preparation of Compound 7

Compounds 3 (2-acetamido-1,3,4,6-tetra-O-acetyl-2-deoxy-β-Dgalactopyranose or galactosamine pentaacetate) is commercially available. Compound 5 was prepared according to published procedures (Weber et al., J. Med. Chem., 1991, 34, 2692).

Preparation of Compound 11

Compounds 8 and 9 are commercially available.

Preparation of Compound 18

Compound 11 was prepared as per the procedures illustrated in Example 3. Compound 14 is commercially available. Compound 17 was prepared using similar procedures reported by Rensen et al., J. Med. Chem., 2004, 47, 5798-5808.

Preparation of Compound 23

Compounds 19 and 21 are commercially available.

Preparation of Compound 24

Compounds 18 and 23 were prepared as per the procedures illustrated in Examples 4 and 5.

Preparation of Compound 25

Compound 24 was prepared as per the procedures illustrated in Example 6.

Preparation of Compound 26

Compound 24 is prepared as per the procedures illustrated in Example 6.

General Preparation of Conjugated ASOs Comprising GalNAc 3 -1 at the 3′ Terminus, Compound 29

Wherein the protected GalNAc 3 -1 has the structure:

The GalNAc 3 cluster portion of the conjugate group GalNAc 3 -1 (GalNAc 3 -1 a ) can be combined with any cleavable moiety to provide a variety of conjugate groups. Wherein GalNAc 3 -1 a has the formula:

The solid support bound protected GalNAc 3 -1, Compound 25, was prepared as per the procedures illustrated in Example 7. Oligomeric Compound 29 comprising GalNAc 3 -1 at the 3′ terminus was prepared using standard procedures in automated DNA/RNA synthesis (see Dupouy et al., Angew. Chem. Int. Ed., 2006, 45, 3623-3627). Phosphoramidite building blocks, Compounds 1 and 1a were prepared as per the procedures illustrated in Example 1. The phosphoramidites illustrated are meant to be representative and not intended to be limiting as other phosphoramidite building blocks can be used to prepare oligomeric compounds having a predetermined sequence and composition. The order and quantity of phosphoramidites added to the solid support can be adjusted to prepare gapped oligomeric compounds as described herein. Such gapped oligomeric compounds can have predetermined composition and base sequence as dictated by any given target.

General Preparation Conjugated ASOs Comprising GalNAc 3 -1 at the 5′ Terminus, Compound 34

The Unylinker™ 30 is commercially available. Oligomeric Compound 34 comprising a GalNAc 3 -1 cluster at the 5′ terminus is prepared using standard procedures in automated DNA/RNA synthesis (see Dupouy et al., Angew. Chem. Int. Ed., 2006, 45, 3623-3627). Phosphoramidite building blocks, Compounds 1 and 1a were prepared as per the procedures illustrated in Example 1. The phosphoramidites illustrated are meant to be representative and not intended to be limiting as other phosphoramidite building blocks can be used to prepare an oligomeric compound having a predetermined sequence and composition. The order and quantity of phosphoramidites added to the solid support can be adjusted to prepare gapped oligomeric compounds as described herein. Such gapped oligomeric compounds can have predetermined composition and base sequence as dictated by any given target.

Preparation of Compound 39

Compounds 4, 13 and 23 were prepared as per the procedures illustrated in Examples 2, 4, and 5. Compound 35 is prepared using similar procedures published in Rouchaud et al., Eur. J. Org. Chem., 2011, 12, 2346-2353.

Preparation of Compound 40

Compound 38 is prepared as per the procedures illustrated in Example 11.

Preparation of Compound 44

Compounds 23 and 36 are prepared as per the procedures illustrated in Examples 5 and 11. Compound 41 is prepared using similar procedures published in WO 2009082607.

Preparation of Compound 45

Compound 43 is prepared as per the procedures illustrated in Example 13.

Preparation of Compound 47

Compound 46 is commercially available.

Preparation of Compound 53

Compounds 48 and 49 are commercially available. Compounds 17 and 47 are prepared as per the procedures illustrated in Examples 4 and 15.

Preparation of Compound 54

Compound 53 is prepared as per the procedures illustrated in Example 16.

Preparation of Compound 55

Compound 53 is prepared as per the procedures illustrated in Example 16.

General Method for the Preparation of Conjugated ASOs Comprising GalNAc 3 -1 at the 3′ Position Via Solid Phase Techniques (Preparation of ISIS 647535, 647536 and 651900)

Unless otherwise stated, all reagents and solutions used for the synthesis of oligomeric compounds are purchased from commercial sources. Standard phosphoramidite building blocks and solid support are used for incorporation nucleoside residues which include for example T, A, G, and m C residues. A 0.1 M solution of phosphoramidite in anhydrous acetonitrile was used for β-D-2′-deoxyribonucleoside and 2′-MOE.

The ASO syntheses were performed on ABI 394 synthesizer (1-2 μmol scale) or on GE Healthcare Bioscience ÄKTA oligopilot synthesizer (40-200 μmol scale) by the phosphoramidite coupling method on an GalNAc 3 -1 loaded VIMAD solid support (110 μmol/g, Guzaev et al., 2003) packed in the column. For the coupling step, the phosphoramidites were delivered 4 fold excess over the loading on the solid support and phosphoramidite condensation was carried out for 10 min. All other steps followed standard protocols supplied by the manufacturer. A solution of 6% dichloroacetic acid in toluene was used for removing dimethoxytrityl (DMT) group from 5′-hydroxyl group of the nucleotide. 4,5-Dicyanoimidazole (0.7 M) in anhydrous CH 3 CN was used as activator during coupling step. Phosphorothioate linkages were introduced by sulfurization with 0.1 M solution of xanthane hydride in 1:1 pyridine/CH 3 CN for a contact time of 3 minutes. A solution of 20% tert-butylhydroperoxide in CH 3 CN containing 6% water was used as an oxidizing agent to provide phosphodiester internucleoside linkages with a contact time of 12 minutes.

After the desired sequence was assembled, the cyanoethyl phosphate protecting groups were deprotected using a 1:1 (v/v) mixture of triethylamine and acetonitrile with a contact time of 45 minutes. The solid-support bound ASOs were suspended in aqueous ammonia (28-30 wt %) and heated at 55° C. for 6 h.

The unbound ASOs were then filtered and the ammonia was boiled off. The residue was purified by high pressure liquid chromatography on a strong anion exchange column (GE Healthcare Bioscience, Source 30Q, 30 μm, 2.54×8 cm, A=100 mM ammonium acetate in 30% aqueous CH 3 CN, B=1.5 M NaBr in A, 0-40% of B in 60 min, flow 14 mL min-1, λ=260 nm). The residue was desalted by HPLC on a reverse phase column to yield the desired ASOs in an isolated yield of 15-30% based on the initial loading on the solid support. The ASOs were characterized by ion-pair-HPLC coupled MS analysis with Agilent 1100 MSD system.

Antisense oligonucleotides not comprising a conjugate were synthesized using standard oligonucleotide synthesis procedures well known in the art.

Using these methods, three separate antisense compounds targeting ApoC III were prepared. As summarized in Table 17, below, each of the three antisense compounds targeting ApoC III had the same nucleobase sequence; ISIS 304801 is a 5-10-5 MOE gapmer having all phosphorothioate linkages; ISIS 647535 is the same as ISIS 304801, except that it had a GalNAc 3 -1 conjugated at its 3′ end; and ISIS 647536 is the same as ISIS 647535 except that certain internucleoside linkages of that compound are phosphodiester linkages. As further summarized in Table 17, two separate antisense compounds targeting SRB-1 were synthesized. ISIS 440762 was a 2-10-2 cEt gapmer with all phosphorothioate internucleoside linkages; ISIS 651900 is the same as ISIS 440762, except that it included a GalNAc 3 -1 at its 3′-end.

Subscripts: “e” indicates 2′-MOE modified nucleoside; “d” indicates β-D-2′-deoxyribonucleoside; “k” indicates 6′-(S)—CH 3 bicyclic nucleoside (e.g. cEt); “s” indicates phosphorothioate internucleoside linkages (PS); “o” indicates phosphodiester internucleoside linkages (PO); and “o′” indicates —O—P(═O)(OH)—. Superscript “m” indicates 5-methylcytosines. “GalNAc 3 -1” indicates a conjugate group having the structure shown previously in Example 9. Note that GalNAc 3 -1 comprises a cleavable adenosine which links the ASO to remainder of the conjugate, which is designated “GalNAc 3 -1 a .” This nomenclature is used in the above table to show the full nucleobase sequence, including the adenosine, which is part of the conjugate. Thus, in the above table, the sequences could also be listed as ending with “GalNAc 3 -1” with the “A do ” omitted. This convention of using the subscript “a” to indicate the portion of a conjugate group lacking a cleavable nucleoside or cleavable moiety is used throughout these Examples. This portion of a conjugate group lacking the cleavable moiety is referred to herein as a “cluster” or “conjugate cluster” or “GalNAc 3 cluster.” In certain instances it is convenient to describe a conjugate group by separately providing its cluster and its cleavable moiety.

Dose-dependent Antisense Inhibition of Human ApoC III in huApoC III Transgenic Mice

ISIS 304801 and ISIS 647535, each targeting human ApoC III and described above, were separately tested and evaluated in a dose-dependent study for their ability to inhibit human ApoC III in human ApoC III transgenic mice.

Treatment

Human ApoCIII transgenic mice were maintained on a 12-hour light/dark cycle and fed ad libitum Teklad lab chow. Animals were acclimated for at least 7 days in the research facility before initiation of the experiment. ASOs were prepared in PBS and sterilized by filtering through a 0.2 micron filter. ASOs were dissolved in 0.9% PBS for injection.

Human ApoC III transgenic mice were injected intraperitoneally once a week for two weeks with ISIS 304801 or 647535 at 0.08, 0.25. 0.75, 2.25 or 6.75 μmol/kg or with PBS as a control. Each treatment group consisted of 4 animals. Forty-eight hours after the administration of the last dose, blood was drawn from each mouse and the mice were sacrificed and tissues were collected.

ApoC III mRNA Analysis

ApoC III mRNA levels in the mice's livers were determined using real-time PCR and RIBOGREEN® RNA quantification reagent (Molecular Probes, Inc. Eugene, Oreg.) according to standard protocols. ApoC III mRNA levels were determined relative to total RNA (using Ribogreen), prior to normalization to PBS-treated control. The results below are presented as the average percent of ApoC III mRNA levels for each treatment group, normalized to PBS-treated control and are denoted as “% PBS”. The half maximal effective dosage (ED 50 ) of each ASO is also presented in Table 18, below.

As illustrated, both antisense compounds reduced ApoC III RNA relative to the PBS control. Further, the antisense compound conjugated to GalNAc 3 -1 (ISIS 647535) was substantially more potent than the antisense compound lacking the GalNAc 3 -1 conjugate (ISIS 304801).

ApoC III Protein Analysis (Turbidometric Assay)

Plasma ApoC III protein analysis was determined using procedures reported by Graham et al, Circulation Research, published online before print Mar. 29, 2013.

Approximately 100 μl of plasma isolated from mice was analyzed without dilution using an Olympus Clinical Analyzer and a commercially available turbidometric ApoC III assay (Kamiya, Cat#KAI-006, Kamiya Biomedical, Seattle, Wash.). The assay protocol was performed as described by the vendor.

As shown in the Table 19 below, both antisense compounds reduced ApoC III protein relative to the PBS control. Further, the antisense compound conjugated to GalNAc 3 -1 (ISIS 647535) was substantially more potent than the antisense compound lacking the GalNAc 3 -1 conjugate (ISIS 304801).

Plasma triglycerides and cholesterol were extracted by the method of Bligh and Dyer (Bligh, E. G. and Dyer, W. J. Can. J. Biochem. Physiol. 37: 911-917, 1959)(Bligh, E and Dyer, W, Can J Biochem Physiol, 37, 911-917, 1959)(Bligh, E and Dyer, W, Can J Biochem Physiol, 37, 911-917, 1959) and measured by using a Beckmann Coulter clinical analyzer and commercially available reagents.

The triglyceride levels were measured relative to PBS injected mice and are denoted as “% PBS”. Results are presented in Table 20. As illustrated, both antisense compounds lowered triglyceride levels. Further, the antisense compound conjugated to GalNAc 3 -1 (ISIS 647535) was substantially more potent than the antisense compound lacking the GalNAc 3 -1 conjugate (ISIS 304801).

Plasma samples were analyzed by HPLC to determine the amount of total cholesterol and of different fractions of cholesterol (HDL and LDL). Results are presented in Tables 21 and 22. As illustrated, both antisense compounds lowered total cholesterol levels; both lowered LDL; and both raised HDL. Further, the antisense compound conjugated to GalNAc 3 -1 (ISIS 647535) was substantially more potent than the antisense compound lacking the GalNAc 3 -1 conjugate (ISIS 304801). An increase in HDL and a decrease in LDL levels is a cardiovascular beneficial effect of antisense inhibition of ApoC III.

Pharmacokinetics Analysis (PK)

The PK of the ASOs was also evaluated. Liver and kidney samples were minced and extracted using standard protocols. Samples were analyzed on MSD1 utilizing IP-HPLC-MS. The tissue level (μg/g) of full-length ISIS 304801 and 647535 was measured and the results are provided in Table 23. As illustrated, liver concentrations of total full-length antisense compounds were similar for the two antisense compounds. Thus, even though the GalNAc 3 -1-conjugated antisense compound is more active in the liver (as demonstrated by the RNA and protein data above), it is not present at substantially higher concentration in the liver. Indeed, the calculated EC 50 (provided in Table 23) confirms that the observed increase in potency of the conjugated compound cannot be entirely attributed to increased accumulation. This result suggests that the conjugate improved potency by a mechanism other than liver accumulation alone, possibly by improving the productive uptake of the antisense compound into cells.

The results also show that the concentration of GalNAc 3 -1 conjugated antisense compound in the kidney is lower than that of antisense compound lacking the GalNAc conjugate. This has several beneficial therapeutic implications. For therapeutic indications where activity in the kidney is not sought, exposure to kidney risks kidney toxicity without corresponding benefit. Moreover, high concentration in kidney typically results in loss of compound to the urine resulting in faster clearance. Accordingly, for non-kidney targets, kidney accumulation is undesired. These data suggest that GalNAc 3 -1 conjugation reduces kidney accumulation.

Metabolites of ISIS 647535 were also identified and their masses were confirmed by high resolution mass spectrometry analysis. The cleavage sites and structures of the observed metabolites are shown below. The relative % of full length ASO was calculated using standard procedures and the results are presented in Table 23a. The major metabolite of ISIS 647535 was full-length ASO lacking the entire conjugate (i.e. ISIS 304801), which results from cleavage at cleavage site A, shown below. Further, additional metabolites resulting from other cleavage sites were also observed. These results suggest that introducing other cleavable bonds such as esters, peptides, disulfides, phosphoramidates or acyl-hydrazones between the GalNAc 3 -1 sugar and the ASO, which can be cleaved by enzymes inside the cell, or which may cleave in the reductive environment of the cytosol, or which are labile to the acidic pH inside endosomes and lyzosomes, can also be useful.

Antisense Inhibition of Human ApoC III in Human ApoC III Transgenic Mice in Single Administration Study

ISIS 304801, 647535 and 647536 each targeting human ApoC III and described in Table 17, were further evaluated in a single administration study for their ability to inhibit human ApoC III in human ApoC III transgenic mice.

Treatment

Human ApoCIII transgenic mice were maintained on a 12-hour light/dark cycle and fed ad libitum Teklad lab chow. Animals were acclimated for at least 7 days in the research facility before initiation of the experiment. ASOs were prepared in PBS and sterilized by filtering through a 0.2 micron filter. ASOs were dissolved in 0.9% PBS for injection.

Human ApoC III transgenic mice were injected intraperitoneally once at the dosage shown below with ISIS 304801, 647535 or 647536 (described above) or with PBS treated control. The treatment group consisted of 3 animals and the control group consisted of 4 animals. Prior to the treatment as well as after the last dose, blood was drawn from each mouse and plasma samples were analyzed. The mice were sacrificed 72 hours following the last administration.

Samples were collected and analyzed to determine the ApoC III mRNA and protein levels in the liver; plasma triglycerides; and cholesterol, including HDL and LDL fractions were assessed as described above (Example 20). Data from those analyses are presented in Tables 24-28, below. Liver transaminase levels, alanine aminotransferase (ALT) and aspartate aminotransferase (AST), in serum were measured relative to saline injected mice using standard protocols. The ALT and AST levels showed that the antisense compounds were well tolerated at all administered doses.

These results show improvement in potency for antisense compounds comprising a GalNAc 3 -1 conjugate at the 3′ terminus (ISIS 647535 and 647536) compared to the antisense compound lacking a GalNAc 3 -1 conjugate (ISIS 304801). Further, ISIS 647536, which comprises a GalNAc 3 -1 conjugate and some phosphodiester linkages was as potent as ISIS 647535, which comprises the same conjugate and all internucleoside linkages within the ASO are phosphorothioate.

These results confirm that the GalNAc 3 -1 conjugate improves potency of an antisense compound. The results also show equal potency of a GalNAc 3 -1 conjugated antisense compounds where the antisense oligonucleotides have mixed linkages (ISIS 647536 which has six phosphodiester linkages) and a full phosphorothioate version of the same antisense compound (ISIS 647535).

Phosphorothioate linkages provide several properties to antisense compounds. For example, they resist nuclease digestion and they bind proteins resulting in accumulation of compound in the liver, rather than in the kidney/urine. These are desirable properties, particularly when treating an indication in the liver. However, phosphorothioate linkages have also been associated with an inflammatory response. Accordingly, reducing the number of phosphorothioate linkages in a compound is expected to reduce the risk of inflammation, but also lower concentration of the compound in liver, increase concentration in the kidney and urine, decrease stability in the presence of nucleases, and lower overall potency. The present results show that a GalNAc 3 -1 conjugated antisense compound where certain phosphorothioate linkages have been replaced with phosphodiester linkages is as potent against a target in the liver as a counterpart having full phosphorothioate linkages. Such compounds are expected to be less proinflammatory (See Example 24 describing an experiment showing reduction of PS results in reduced inflammatory effect).

Effect of GalNAc 3 -1 Conjugated Modified ASO Targeting SRB-1 In Vivo

ISIS 440762 and 651900, each targeting SRB-1 and described in Table 17, were evaluated in a dose-dependent study for their ability to inhibit SRB-1 in Balb/c mice.

Treatment

Six week old male Balb/c mice (Jackson Laboratory, Bar Harbor, Me.) were injected subcutaneously once at the dosage shown below with ISIS 440762, 651900 or with PBS treated control. Each treatment group consisted of 4 animals. The mice were sacrificed 48 hours following the final administration to determine the SRB-1 mRNA levels in liver using real-time PCR and RIBOGREEN® RNA quantification reagent (Molecular Probes, Inc. Eugene, Oreg.) according to standard protocols. SRB-1 mRNA levels were determined relative to total RNA (using Ribogreen), prior to normalization to PBS-treated control. The results below are presented as the average percent of SRB-1 mRNA levels for each treatment group, normalized to PBS-treated control and is denoted as “% PBS”.

As illustrated in Table 29, both antisense compounds lowered SRB-1 mRNA levels. Further, the antisense compound comprising the GalNAc 3 -1 conjugate (ISIS 651900) was substantially more potent than the antisense compound lacking the GalNAc 3 -1 conjugate (ISIS 440762). These results demonstrate that the potency benefit of GalNAc 3 -1 conjugates are observed using antisense oligonucleotides complementary to a different target and having different chemically modified nucleosides, in this instance modified nucleosides comprise constrained ethyl sugar moieties (a bicyclic sugar moiety).

Human Peripheral Blood Mononuclear Cells (hPBMC) Assay Protocol

The hPBMC assay was performed using BD Vautainer CPT tube method. A sample of whole blood from volunteered donors with informed consent at US HealthWorks clinic (Faraday & El Camino Real, Carlsbad) was obtained and collected in 4-15 BD Vacutainer CPT 8 ml tubes (VWR Cat. #BD362753). The approximate starting total whole blood volume in the CPT tubes for each donor was recorded using the PBMC assay data sheet.

The blood sample was remixed immediately prior to centrifugation by gently inverting tubes 8-10 times. CPT tubes were centrifuged at rt (18-25° C.) in a horizontal (swing-out) rotor for 30 min. at 1500-1800 RCF with brake off (2700 RPM Beckman Allegra 6R). The cells were retrieved from the buffy coat interface (between Ficoll and polymer gel layers); transferred to a sterile 50 ml conical tube and pooled up to 5 CPT tubes/50 ml conical tube/donor. The cells were then washed twice with PBS (Ca ++ , Mg ++ free; GIBCO). The tubes were topped up to 50 ml and mixed by inverting several times. The sample was then centrifuged at 330×g for 15 minutes at rt (1215 RPM in Beckman Allegra 6R) and aspirated as much supernatant as possible without disturbing pellet. The cell pellet was dislodged by gently swirling tube and resuspended cells in RPMI+10% FBS+pen/strep (˜1 ml/10 ml starting whole blood volume). A 60 μl sample was pipette into a sample vial (Beckman Coulter) with 600 μl VersaLyse reagent (Beckman Coulter Cat#A09777) and was gently vortexed for 10-15 sec. The sample was allowed to incubate for 10 min. at rt and being mixed again before counting. The cell suspension was counted on Vicell XR cell viability analyzer (Beckman Coulter) using PBMC cell type (dilution factor of 1:11 was stored with other parameters). The live cell/ml and viability were recorded. The cell suspension was diluted to 1×10 7 live PBMC/ml in RPMI+10% FBS+pen/strep.

The cells were plated at 5×10 5 in 50 μl/well of 96-well tissue culture plate (Falcon Microtest). 50 μl/well of 2x concentration oligos/controls diluted in RPMI+10% FBS+pen/strep. was added according to experiment template (100 μl/well total). Plates were placed on the shaker and allowed to mix for approx. 1 min. After being incubated for 24 hrs at 37° C.; 5% CO 2 , the plates were centrifuged at 400×g for 10 minutes before removing the supernatant for MSD cytokine assay (i.e. human IL-6, IL-10, IL-8 and MCP-1).

Evaluation of Proinflammatory Effects in hPBMC Assay for GalNAc 3 -1 Conjugated ASOs

The antisense oligonucleotides (ASOs) listed in Table 30 were evaluated for proinflammatory effect in hPBMC assay using the protocol described in Example 23. ISIS 353512 is an internal standard known to be a high responder for IL-6 release in the assay. The hPBMCs were isolated from fresh, volunteered donors and were treated with ASOs at 0, 0.0128, 0.064, 0.32, 1.6, 8, 40 and 200 μM concentrations. After a 24 hr treatment, the cytokine levels were measured.

The levels of IL-6 were used as the primary readout. The EC 50 and E max was calculated using standard procedures. Results are expressed as the average ratio of E max /EC 50 from two donors and is denoted as “E max /EC 50 .” The lower ratio indicates a relative decrease in the proinflammatory response and the higher ratio indicates a relative increase in the proinflammatory response.

With regard to the test compounds, the least proinflammatory compound was the PS/PO linked ASO (ISIS 616468). The GalNAc 3 -1 conjugated ASO, ISIS 647535 was slightly less proinflammatory than its non-conjugated counterpart ISIS 304801. These results indicate that incorporation of some PO linkages reduces proinflammatory reaction and addition of a GalNAc 3 -1 conjugate does not make a compound more proinflammatory and may reduce proinflammatory response. Accordingly, one would expect that an antisense compound comprising both mixed PS/PO linkages and a GalNAc 3 -1 conjugate would produce lower proinflammatory responses relative to full PS linked antisense compound with or without a GalNAc 3 -1 conjugate. These results show that GalNAc 3 -1 conjugated antisense compounds, particularly those having reduced PS content are less proinflammatory.

Together, these results suggest that a GalNAc 3 -1 conjugated compound, particularly one with reduced PS content, can be administered at a higher dose than a counterpart full PS antisense compound lacking a GalNAc 3 -1 conjugate. Since half-life is not expected to be substantially different for these compounds, such higher administration would result in less frequent dosing. Indeed such administration could be even less frequent, because the GalNAc 3 -1 conjugated compounds are more potent (See Examples 20-22) and re-dosing is necessary once the concentration of a compound has dropped below a desired level, where such desired level is based on potency.

Subscripts: “e” indicates 2′-MOE modified nucleoside; “d” indicates β-D-2′-deoxyribonucleoside; “k” indicates 6′-(S)—CH 3 bicyclic nucleoside (e.g. cEt); “s” indicates phosphorothioate internucleoside linkages (PS); “o” indicates phosphodiester internucleoside linkages (PO); and “o′” indicates —O—P(═O)(OH)—. Superscript “m” indicates 5-methylcytosines. “A do′ -GalNAc 3 -1 a ” indicates a conjugate having the structure GalNAc 3 -1 shown in Example 9 attached to the 3′-end of the antisense oligonucleotide, as indicated.

Effect of GalNAc 3 -1 Conjugated Modified ASO Targeting Human ApoC III In Vitro

ISIS 304801 and 647535 described above were tested in vitro. Primary hepatocyte cells from transgenic mice at a density of 25,000 cells per well were treated with 0.03, 0.08, 0.24, 0.74, 2.22, 6.67 and 20 μM concentrations of modified oligonucleotides. After a treatment period of approximately 16 hours, RNA was isolated from the cells and mRNA levels were measured by quantitative real-time PCR and the hApoC III mRNA levels were adjusted according to total RNA content, as measured by RIBOGREEN.

The IC 50 was calculated using the standard methods and the results are presented in Table 32. As illustrated, comparable potency was observed in cells treated with ISIS 647535 as compared to the control, ISIS 304801.

In this experiment, the large potency benefits of GalNAc 3 -1 conjugation that are observed in vivo were not observed in vitro. Subsequent free uptake experiments in primary hepatocytes in vitro did show increased potency of oligonucleotides comprising various GalNAc conjugates relative to oligonucleotides that lacking the GalNAc conjugate. (see Examples 60, 82, and 92)

Effect of PO/PS Linkages on ApoC III ASO Activity

Human ApoC III transgenic mice were injected intraperitoneally once at 25 mg/kg of ISIS 304801, or ISIS 616468 (both described above) or with PBS treated control once per week for two weeks. The treatment group consisted of 3 animals and the control group consisted of 4 animals. Prior to the treatment as well as after the last dose, blood was drawn from each mouse and plasma samples were analyzed. The mice were sacrificed 72 hours following the last administration.

Samples were collected and analyzed to determine the ApoC III protein levels in the liver as described above (Example 20). Data from those analyses are presented in Table 33, below.

These results show reduction in potency for antisense compounds with PO/PS (ISIS 616468) in the wings relative to full PS (ISIS 304801).

Compound 56

Compound 56 is commercially available from Glen Research or may be prepared according to published procedures reported by Shchepinov et al., Nucleic Acids Research, 1997, 25(22), 4447-4454.

Preparation of Compound 60

Compound 4 was prepared as per the procedures illustrated in Example 2. Compound 57 is commercially available. Compound 60 was confirmed by structural analysis.

Compound 57 is meant to be representative and not intended to be limiting as other monoprotected substituted or unsubstituted alkyl diols including but not limited to those presented in the specification herein can be used to prepare phosphoramidites having a predetermined composition.

Preparation of Compound 63

Compounds 61 and 62 are prepared using procedures similar to those reported by Tober et al., Eur. J. Org. Chem., 2013, 3, 566-577; and Jiang et al., Tetrahedron, 2007, 63(19), 3982-3988.

Alternatively, Compound 63 is prepared using procedures similar to those reported in scientific and patent literature by Kim et al., Synlett, 2003, 12, 1838-1840; and Kim et al., published PCT International Application, WO 2004063208.

Preparation of Compound 63b

Compound 63a is prepared using procedures similar to those reported by Hanessian et al., Canadian Journal of Chemistry, 1996, 74(9), 1731-1737.

Preparation of Compound 63d

Compound 63c is prepared using procedures similar to those reported by Chen et al., Chinese Chemical Letters, 1998, 9(5), 451-453.

Preparation of Compound 67

Compound 64 was prepared as per the procedures illustrated in Example 2. Compound 65 is prepared using procedures similar to those reported by Or et al., published PCT International Application, WO 2009003009. The protecting groups used for Compound 65 are meant to be representative and not intended to be limiting as other protecting groups including but not limited to those presented in the specification herein can be used.

Preparation of Compound 70

Compound 64 was prepared as per the procedures illustrated in Example 2. Compound 68 is commercially available. The protecting group used for Compound 68 is meant to be representative and not intended to be limiting as other protecting groups including but not limited to those presented in the specification herein can be used.

Preparation of Compound 75a

Compound 75 is prepared according to published procedures reported by Shchepinov et al., Nucleic Acids Research, 1997, 25(22), 4447-4454.

Preparation of Compound 79

Compound 76 was prepared according to published procedures reported by Shchepinov et al., Nucleic Acids Research, 1997, 25(22), 4447-4454.

Preparation of Compound 79a

Compound 77 is prepared as per the procedures illustrated in Example 35.

General Method for the Preparation of Conjugated Oligomeric Compound 82 Comprising a Phosphodiester Linked GalNAc 3 -2 Conjugate at 5′ Terminus Via Solid Support (Method I)

wherein GalNAc 3 -2 has the structure:

The GalNAc 3 cluster portion of the conjugate group GalNAc 3 -2 (GalNAc 3 -2 a ) can be combined with any cleavable moiety to provide a variety of conjugate groups. Wherein GalNAc 3 -2 a has the formula:

The VIMAD-bound oligomeric compound 79b was prepared using standard procedures for automated DNA/RNA synthesis (see Dupouy et al., Angew. Chem. Int. Ed., 2006, 45, 3623-3627). The phosphoramidite Compounds 56 and 60 were prepared as per the procedures illustrated in Examples 27 and 28, respectively. The phosphoramidites illustrated are meant to be representative and not intended to be limiting as other phosphoramidite building blocks including but not limited those presented in the specification herein can be used to prepare an oligomeric compound having a phosphodiester linked conjugate group at the 5′ terminus. The order and quantity of phosphoramidites added to the solid support can be adjusted to prepare the oligomeric compounds as described herein having any predetermined sequence and composition.

Alternative Method for the Preparation of Oligomeric Compound 82 Comprising a Phosphodiester Linked GalNAc 3 -2 Conjugate at 5′ Terminus (Method II)

The VIMAD-bound oligomeric compound 79b was prepared using standard procedures for automated DNA/RNA synthesis (see Dupouy et al., Angew. Chem. Int. Ed., 2006, 45, 3623-3627). The GalNAc 3 -2 cluster phosphoramidite, Compound 79 was prepared as per the procedures illustrated in Example 35. This alternative method allows a one-step installation of the phosphodiester linked GalNAc 3 -2 conjugate to the oligomeric compound at the final step of the synthesis. The phosphoramidites illustrated are meant to be representative and not intended to be limiting, as other phosphoramidite building blocks including but not limited to those presented in the specification herein can be used to prepare oligomeric compounds having a phosphodiester conjugate at the 5′ terminus. The order and quantity of phosphoramidites added to the solid support can be adjusted to prepare the oligomeric compounds as described herein having any predetermined sequence and composition.

General Method for the Preparation of Oligomeric Compound 83h Comprising a GalNAc 3 -3 Conjugate at the 5′ Terminus (GalNAc 3 -1 Modified for 5′ End Attachment) Via Solid Support

Compound 18 was prepared as per the procedures illustrated in Example 4. Compounds 83a and 83b are commercially available. Oligomeric Compound 83e comprising a phosphodiester linked hexylamine was prepared using standard oligonucleotide synthesis procedures. Treatment of the protected oligomeric compound with aqueous ammonia provided the 5′-GalNAc 3 -3 conjugated oligomeric compound (83h).

Wherein GalNAc 3 -3 has the structure:

The GalNAc 3 cluster portion of the conjugate group GalNAc 3 -3 (GalNAc 3 -3 a ) can be combined with any cleavable moiety to provide a variety of conjugate groups. Wherein GalNAc 3 -3 a has the formula:

General Method for the Preparation of Oligomeric Compound 89 Comprising a Phosphodiester Linked GalNAc 3 -4 Conjugate at the 3′ Terminus Via Solid Support

Wherein GalNAc 3 -4 has the structure:

Wherein CM is a cleavable moiety. In certain embodiments, cleavable moiety is:

The GalNAc 3 cluster portion of the conjugate group GalNAc 3 -4 (GalNAc 3 -4 a ) can be combined with any cleavable moiety to provide a variety of conjugate groups. Wherein GalNAc 3 -4 a has the formula:

The protected Unylinker functionalized solid support Compound 30 is commercially available. Compound 84 is prepared using procedures similar to those reported in the literature (see Shchepinov et al., Nucleic Acids Research, 1997, 25(22), 4447-4454; Shchepinov et al., Nucleic Acids Research, 1999, 27, 3035-3041; and Hornet et al., Nucleic Acids Research, 1997, 25, 4842-4849).

The phosphoramidite building blocks, Compounds 60 and 79a are prepared as per the procedures illustrated in Examples 28 and 36. The phosphoramidites illustrated are meant to be representative and not intended to be limiting as other phosphoramidite building blocks can be used to prepare an oligomeric compound having a phosphodiester linked conjugate at the 3′ terminus with a predetermined sequence and composition. The order and quantity of phosphoramidites added to the solid support can be adjusted to prepare the oligomeric compounds as described herein having any predetermined sequence and composition.

General Method for the Preparation of ASOs Comprising a Phosphodiester Linked GalNAc 3 -2 (See Example 37, Bx is Adenine) Conjugate at the 5′ Position Via Solid Phase Techniques (Preparation of ISIS 661134)

Unless otherwise stated, all reagents and solutions used for the synthesis of oligomeric compounds are purchased from commercial sources. Standard phosphoramidite building blocks and solid support are used for incorporation nucleoside residues which include for example T, A, G, and m C residues. Phosphoramidite compounds 56 and 60 were used to synthesize the phosphodiester linked GalNAc 3 -2 conjugate at the 5′ terminus. A 0.1 M solution of phosphoramidite in anhydrous acetonitrile was used for β-D-2′-deoxyribonucleoside and 2′-MOE.

The ASO syntheses were performed on ABI 394 synthesizer (1-2 μmol scale) or on GE Healthcare Bioscience ÄKTA oligopilot synthesizer (40-200 μmol scale) by the phosphoramidite coupling method on VIMAD solid support (110 μmol/g, Guzaev et al., 2003) packed in the column. For the coupling step, the phosphoramidites were delivered at a 4 fold excess over the initial loading of the solid support and phosphoramidite coupling was carried out for 10 min. All other steps followed standard protocols supplied by the manufacturer. A solution of 6% dichloroacetic acid in toluene was used for removing the dimethoxytrityl (DMT) groups from 5′-hydroxyl groups of the nucleotide. 4,5-Dicyanoimidazole (0.7 M) in anhydrous CH 3 CN was used as activator during the coupling step. Phosphorothioate linkages were introduced by sulfurization with 0.1 M solution of xanthane hydride in 1:1 pyridine/CH 3 CN for a contact time of 3 minutes. A solution of 20% tert-butylhydroperoxide in CH 3 CN containing 6% water was used as an oxidizing agent to provide phosphodiester internucleoside linkages with a contact time of 12 minutes.

After the desired sequence was assembled, the cyanoethyl phosphate protecting groups were deprotected using a 20% diethylamine in toluene (v/v) with a contact time of 45 minutes. The solid-support bound ASOs were suspended in aqueous ammonia (28-30 wt %) and heated at 55° C. for 6 h.

The unbound ASOs were then filtered and the ammonia was boiled off. The residue was purified by high pressure liquid chromatography on a strong anion exchange column (GE Healthcare Bioscience, Source 30Q, 30 μm, 2.54×8 cm, A=100 mM ammonium acetate in 30% aqueous CH 3 CN, B=1.5 M NaBr in A, 0-40% of B in 60 min, flow 14 mL min-1, λ=260 nm). The residue was desalted by HPLC on a reverse phase column to yield the desired ASOs in an isolated yield of 15-30% based on the initial loading on the solid support. The ASOs were characterized by ion-pair-HPLC coupled MS analysis with Agilent 1100 MSD system.

Subscripts: “e” indicates 2′-MOE modified nucleoside; “d” indicates β-D-2′-deoxyribonucleoside; “k” indicates 6′-(S)—CH 3 bicyclic nucleoside (e.g. cEt); “s” indicates phosphorothioate internucleoside linkages (PS); “o” indicates phosphodiester internucleoside linkages (PO); and “o′” indicates —O—P(═O)(OH)—. Superscript “m” indicates 5-methylcytosines. The structure of GalNAc 3 -2 a is shown in Example 37.

General Method for the Preparation of ASOs Comprising a GalNAc 3 -3 Conjugate at the 5′ Position Via Solid Phase Techniques (Preparation of ISIS 661166)

The synthesis for ISIS 661166 was performed using similar procedures as illustrated in Examples 39 and 41.

ISIS 661166 is a 5-10-5 MOE gapmer, wherein the 5′ position comprises a GalNAc 3 -3 conjugate. The ASO was characterized by ion-pair-HPLC coupled MS analysis with Agilent 1100 MSD system.

Subscripts: “e” indicates 2′-MOE modified nucleoside; “d” indicates β-D-2′-deoxyribonucleoside; “s” indicates phosphorothioate internucleoside linkages (PS); “o” indicates phosphodiester internucleoside linkages (PO); and “o′” indicates —O—P(═O)(OH)—. Superscript “m” indicates 5-methylcytosines. The structure of “5′-GalNAc 3 -3a” is shown in Example 39.

Dose-dependent Study of Phosphodiester Linked GalNAc 3 -2 (See Examples 37 and 41, Bx is Adenine) at the 5′ Terminus Targeting SRB-1 In Vivo

ISIS 661134 (see Example 41) comprising a phosphodiester linked GalNAc 3 -2 conjugate at the 5′ terminus was tested in a dose-dependent study for antisense inhibition of SRB-1 in mice. Unconjugated ISIS 440762 and 651900 (GalNAc 3 -1 conjugate at 3′ terminus, see Example 9) were included in the study for comparison and are described previously in Table 17.

Treatment

Six week old male Balb/c mice (Jackson Laboratory, Bar Harbor, Me.) were injected subcutaneously once at the dosage shown below with ISIS 440762, 651900, 661134 or with PBS treated control. Each treatment group consisted of 4 animals. The mice were sacrificed 72 hours following the final administration to determine the liver SRB-1 mRNA levels using real-time PCR and RIBOGREEN® RNA quantification reagent (Molecular Probes, Inc. Eugene, Oreg.) according to standard protocols. SRB-1 mRNA levels were determined relative to total RNA (using Ribogreen), prior to normalization to PBS-treated control. The results below are presented as the average percent of SRB-1 mRNA levels for each treatment group, normalized to PBS-treated control and is denoted as “% PBS”. The ED 50 s were measured using similar methods as described previously and are presented below.

As illustrated in Table 35, treatment with antisense oligonucleotides lowered SRB-1 mRNA levels in a dose-dependent manner. Indeed, the antisense oligonucleotides comprising the phosphodiester linked GalNAc 3 -2 conjugate at the 5′ terminus (ISIS 661134) or the GalNAc 3 -1 conjugate linked at the 3′ terminus (ISIS 651900) showed substantial improvement in potency compared to the unconjugated antisense oligonucleotide (ISIS 440762). Further, ISIS 661134, which comprises the phosphodiester linked GalNAc 3 -2 conjugate at the 5′ terminus was equipotent compared to ISIS 651900, which comprises the GalNAc 3 -1 conjugate at the 3′ terminus.

Structures for 3′ GalNAc 3 -1 and 5′ GalNAc 3 -2 were described previously in Examples 9 and 37.

Pharmacokinetics Analysis (PK)

The PK of the ASOs from the high dose group (7 mg/kg) was examined and evaluated in the same manner as illustrated in Example 20. Liver sample was minced and extracted using standard protocols. The full length metabolites of 661134 (5′ GalNAc 3 -2) and ISIS 651900 (3′ GalNAc 3 -1) were identified and their masses were confirmed by high resolution mass spectrometry analysis. The results showed that the major metabolite detected for the ASO comprising a phosphodiester linked GalNAc 3 -2 conjugate at the 5′ terminus (ISIS 661134) was ISIS 440762 (data not shown). No additional metabolites, at a detectable level, were observed. Unlike its counterpart, additional metabolites similar to those reported previously in Table 23a were observed for the ASO having the GalNAc 3 -1 conjugate at the 3′ terminus (ISIS 651900). These results suggest that having the phosphodiester linked GalNAc 3 -1 or GalNAc 3 -2 conjugate may improve the PK profile of ASOs without compromising their potency.

Effect of PO/PS Linkages on Antisense Inhibition of ASOs Comprising GalNAc 3 -1 Conjugate (See Example 9) at the 3′ Terminus Targeting SRB-1

ISIS 655861 and 655862 comprising a GalNAc 3 -1 conjugate at the 3′ terminus each targeting SRB-1 were tested in a single administration study for their ability to inhibit SRB-1 in mice. The parent unconjugated compound, ISIS 353382 was included in the study for comparison.

The ASOs are 5-10-5 MOE gapmers, wherein the gap region comprises ten 2′-deoxyribonucleosides and each wing region comprises five 2′-MOE modified nucleosides. The ASOs were prepared using similar methods as illustrated previously in Example 19 and are described Table 36, below.

Subscripts: “e” indicates 2′-MOE modified nucleoside; “d” indicates β-D-2′-deoxyribonucleoside; “s” indicates phosphorothioate internucleoside linkages (PS); “o” indicates phosphodiester internucleoside linkages (PO); and “o′” indicates —O—P(═O)(OH)—. Superscript “m” indicates 5-methylcytosines. The structure of “GalNAc 3 -1” is shown in Example 9.

Treatment

Six week old male Balb/c mice (Jackson Laboratory, Bar Harbor, Me.) were injected subcutaneously once at the dosage shown below with ISIS 353382, 655861, 655862 or with PBS treated control. Each treatment group consisted of 4 animals. Prior to the treatment as well as after the last dose, blood was drawn from each mouse and plasma samples were analyzed. The mice were sacrificed 72 hours following the final administration to determine the liver SRB-1 mRNA levels using real-time PCR and RIBOGREEN® RNA quantification reagent (Molecular Probes, Inc. Eugene, Oreg.) according to standard protocols. SRB-1 mRNA levels were determined relative to total RNA (using Ribogreen), prior to normalization to PBS-treated control. The results below are presented as the average percent of SRB-1 mRNA levels for each treatment group, normalized to PBS-treated control and is denoted as “% PBS”. The ED 50 s were measured using similar methods as described previously and are reported below.

As illustrated in Table 37, treatment with antisense oligonucleotides lowered SRB-1 mRNA levels in a dose-dependent manner compared to PBS treated control. Indeed, the antisense oligonucleotides comprising the GalNAc 3 -1 conjugate at the 3′ terminus (ISIS 655861 and 655862) showed substantial improvement in potency comparing to the unconjugated antisense oligonucleotide (ISIS 353382). Further, ISIS 655862 with mixed PS/PO linkages showed an improvement in potency relative to full PS (ISIS 655861).

Liver transaminase levels, alanine aminotransferase (ALT) and aspartate aminotransferase (AST), in serum were measured relative to saline injected mice using standard protocols. Organ weights were also evaluated. The results demonstrated that no elevation in transaminase levels (Table 38) or organ weights (data not shown) were observed in mice treated with ASOs compared to PBS control. Further, the ASO with mixed PS/PO linkages (ISIS 655862) showed similar transaminase levels compared to full PS (ISIS 655861).

Preparation of PFP Ester, Compound 110a

Compound 4 (9.5 g, 28.8 mmoles) was treated with compound 103a or 103b (38 mmoles), individually, and TMSOTf (0.5 eq.) and molecular sieves in dichloromethane (200 mL), and stirred for 16 hours at room temperature. At that time, the organic layer was filtered thru celite, then washed with sodium bicarbonate, water and brine. The organic layer was then separated and dried over sodium sulfate, filtered and reduced under reduced pressure. The resultant oil was purified by silica gel chromatography (2%-->10% methanol/dichloromethane) to give compounds 104a and 104b in >80% yield. LCMS and proton NMR was consistent with the structure.

Compounds 104a and 104b were treated to the same conditions as for compounds 100a-d (Example 47), to give compounds 105a and 105b in >90% yield. LCMS and proton NMR was consistent with the structure.

Compounds 105a and 105b were treated, individually, with compound 90 under the same conditions as for compounds 901a-d, to give compounds 106a (80%) and 106b (20%). LCMS and proton NMR was consistent with the structure.

Compounds 106a and 106b were treated to the same conditions as for compounds 96a-d (Example 47), to give 107a (60%) and 107b (20%). LCMS and proton NMR was consistent with the structure.

Compounds 107a and 107b were treated to the same conditions as for compounds 97a-d (Example 47), to give compounds 108a and 108b in 40-60% yield. LCMS and proton NMR was consistent with the structure.

Compounds 108a (60%) and 108b (40%) were treated to the same conditions as for compounds 100a-d (Example 47), to give compounds 109a and 109b in >80% yields. LCMS and proton NMR was consistent with the structure.

Compound 109a was treated to the same conditions as for compounds 101a-d (Example 47), to give Compound 110a in 30-60% yield. LCMS and proton NMR was consistent with the structure. Alternatively, Compound 110b can be prepared in a similar manner starting with Compound 109b.

General Procedure for Conjugation with PFP Esters (Oligonucleotide 111); Preparation of ISIS 666881 (GalNAc 3 -10)

A 5′-hexylamino modified oligonucleotide was synthesized and purified using standard solid-phase oligonucleotide procedures. The 5′-hexylamino modified oligonucleotide was dissolved in 0.1 M sodium tetraborate, pH 8.5 (200 μL) and 3 equivalents of a selected PFP esterified GalNAc 3 cluster dissolved in DMSO (50 μL) was added. If the PFP ester precipitated upon addition to the ASO solution DMSO was added until all PFP ester was in solution. The reaction was complete after about 16 h of mixing at room temperature. The resulting solution was diluted with water to 12 mL and then spun down at 3000 rpm in a spin filter with a mass cut off of 3000 Da. This process was repeated twice to remove small molecule impurities. The solution was then lyophilized to dryness and redissolved in concentrated aqueous ammonia and mixed at room temperature for 2.5 h followed by concentration in vacuo to remove most of the ammonia. The conjugated oligonucleotide was purified and desalted by RP-HPLC and lyophilized to provide the GalNAc 3 conjugated oligonucleotide.

Oligonucleotide 111 is conjugated with GalNAc 3 -10. The GalNAc 3 cluster portion of the conjugate group GalNAc 3 -10 (GalNAc 3 -10 a ) can be combined with any cleavable moiety to provide a variety of conjugate groups. In certain embodiments, the cleavable moiety is —P(═O)(OH)-A d -P(═O)(OH)— as shown in the oligonucleotide (ISIS 666881) synthesized with GalNAc 3 -10 below. The structure of GalNAc 3 -10 (GalNAc 3 -10 a -CM-) is shown below:

Following this general procedure ISIS 666881 was prepared. 5′-hexylamino modified oligonucleotide, ISIS 660254, was synthesized and purified using standard solid-phase oligonucleotide procedures. ISIS 660254 (40 mg, 5.2 μmol) was dissolved in 0.1 M sodium tetraborate, pH 8.5 (200 μL) and 3 equivalents PFP ester (Compound 110a) dissolved in DMSO (50 μL) was added. The PFP ester precipitated upon addition to the ASO solution requiring additional DMSO (600 μL) to fully dissolve the PFP ester. The reaction was complete after 16 h of mixing at room temperature. The solution was diluted with water to 12 mL total volume and spun down at 3000 rpm in a spin filter with a mass cut off of 3000 Da. This process was repeated twice to remove small molecule impurities. The solution was lyophilized to dryness and redissolved in concentrated aqueous ammonia with mixing at room temperature for 2.5 h followed by concentration in vacuo to remove most of the ammonia. The conjugated oligonucleotide was purified and desalted by RP-HPLC and lyophilized to give ISIS 666881 in 90% yield by weight (42 mg, 4.7 μmol).

Capital letters indicate the nucleobase for each nucleoside and m C indicates a 5-methyl cytosine. Subscripts: “e” indicates a 2′-MOE modified nucleoside; “d” indicates a β-D-2′-deoxyribonucleoside; “s” indicates a phosphorothioate internucleoside linkage (PS); “o” indicates a phosphodiester internucleoside linkage (PO); and “o′” indicates —O—P(═O)(OH)—. Conjugate groups are in bold.

Preparation of Oligonucleotide 102 Comprising GalNAc 3 -8

The triacid 90 (4 g, 14.43 mmol) was dissolved in DMF (120 mL) and N,N-Diisopropylethylamine (12.35 mL, 72 mmoles). Pentafluorophenyl trifluoroacetate (8.9 mL, 52 mmoles) was added dropwise, under argon, and the reaction was allowed to stir at room temperature for 30 minutes. Boc-diamine 91a or 91b (68.87 mmol) was added, along with N,N-Diisopropylethylamine (12.35 mL, 72 mmoles), and the reaction was allowed to stir at room temperature for 16 hours. At that time, the DMF was reduced by >75% under reduced pressure, and then the mixture was dissolved in dichloromethane. The organic layer was washed with sodium bicarbonate, water and brine. The organic layer was then separated and dried over sodium sulfate, filtered and reduced to an oil under reduced pressure. The resultant oil was purified by silica gel chromatography (2%-->10% methanol/dichloromethane) to give compounds 92a and 92b in an approximate 80% yield. LCMS and proton NMR were consistent with the structure.

Compound 92a or 92b (6.7 mmoles) was treated with 20 mL of dichloromethane and 20 mL of trifluoroacetic acid at room temperature for 16 hours. The resultant solution was evaporated and then dissolved in methanol and treated with DOWEX-OH resin for 30 minutes. The resultant solution was filtered and reduced to an oil under reduced pressure to give 85-90% yield of compounds 93a and 93b.

Compounds 7 or 64 (9.6 mmoles) were treated with HBTU (3.7 g, 9.6 mmoles) and N,N-Diisopropylethylamine (5 mL) in DMF (20 mL) for 15 minutes. To this was added either compounds 93a or 93b (3 mmoles), and allowed to stir at room temperature for 16 hours. At that time, the DMF was reduced by >75% under reduced pressure, and then the mixture was dissolved in dichloromethane. The organic layer was washed with sodium bicarbonate, water and brine. The organic layer was then separated and dried over sodium sulfate, filtered and reduced to an oil under reduced pressure. The resultant oil was purified by silica gel chromatography (5%-->20% methanol/dichloromethane) to give compounds 96a-d in 20-40% yield. LCMS and proton NMR was consistent with the structure.

Compounds 96a-d (0.75 mmoles), individually, were hydrogenated over Raney Nickel for 3 hours in Ethanol (75 mL). At that time, the catalyst was removed by filtration thru celite, and the ethanol removed under reduced pressure to give compounds 97a-d in 80-90% yield. LCMS and proton NMR were consistent with the structure.

Compound 23 (0.32 g, 0.53 mmoles) was treated with HBTU (0.2 g, 0.53 mmoles) and N,N-Diisopropylethylamine (0.19 mL, 1.14 mmoles) in DMF (30 mL) for 15 minutes. To this was added compounds 97a-d (0.38 mmoles), individually, and allowed to stir at room temperature for 16 hours. At that time, the DMF was reduced by >75% under reduced pressure, and then the mixture was dissolved in dichloromethane. The organic layer was washed with sodium bicarbonate, water and brine. The organic layer was then separated and dried over sodium sulfate, filtered and reduced to an oil under reduced pressure. The resultant oil was purified by silica gel chromatography (2%-->20% methanol/dichloromethane) to give compounds 98a-d in 30-40% yield. LCMS and proton NMR was consistent with the structure.

Compound 99 (0.17 g, 0.76 mmoles) was treated with HBTU (0.29 g, 0.76 mmoles) and N,N-Diisopropylethylamine (0.35 mL, 2.0 mmoles) in DMF (50 mL) for 15 minutes. To this was added compounds 97a-d (0.51 mmoles), individually, and allowed to stir at room temperature for 16 hours. At that time, the DMF was reduced by >75% under reduced pressure, and then the mixture was dissolved in dichloromethane. The organic layer was washed with sodium bicarbonate, water and brine. The organic layer was then separated and dried over sodium sulfate, filtered and reduced to an oil under reduced pressure. The resultant oil was purified by silica gel chromatography (5%-->20% methanol/dichloromethane) to give compounds 100a-d in 40-60% yield. LCMS and proton NMR was consistent with the structure.

Compounds 100a-d (0.16 mmoles), individually, were hydrogenated over 10% Pd(OH) 2 /C for 3 hours in methanol/ethyl acetate (1:1, 50 mL). At that time, the catalyst was removed by filtration thru celite, and the organics removed under reduced pressure to give compounds 101a-d in 80-90% yield. LCMS and proton NMR was consistent with the structure.

Compounds 101a-d (0.15 mmoles), individually, were dissolved in DMF (15 mL) and pyridine (0.016 mL, 0.2 mmoles). Pentafluorophenyl trifluoroacetate (0.034 mL, 0.2 mmoles) was added dropwise, under argon, and the reaction was allowed to stir at room temperature for 30 minutes. At that time, the DMF was reduced by >75% under reduced pressure, and then the mixture was dissolved in dichloromethane. The organic layer was washed with sodium bicarbonate, water and brine. The organic layer was then separated and dried over sodium sulfate, filtered and reduced to an oil under reduced pressure. The resultant oil was purified by silica gel chromatography (2%-->5% methanol/dichloromethane) to give compounds 102a-d in an approximate 80% yield. LCMS and proton NMR were consistent with the structure.

Oligomeric Compound 102, comprising a GalNAc 3 -8 conjugate group, was prepared using the general procedures illustrated in Example 46. The GalNAc 3 cluster portion of the conjugate group GalNAc 3 -8 (GalNAc 3 -8 a ) can be combined with any cleavable moiety to provide a variety of conjugate groups. In a preferred embodiment, the cleavable moiety is —P(═O)(OH)-A d -P(═O)(OH)—.

The structure of GalNAc 3 -8 (GalNAc 3 -8 a -CM-) is shown below:

Preparation of Oligonucleotide 119 Comprising GalNAc 3 -7

Compound 112 was synthesized following the procedure described in the literature ( J. Med. Chem. 2004, 47, 5798-5808).

Compound 112 (5 g, 8.6 mmol) was dissolved in 1:1 methanol/ethyl acetate (22 mL/22 mL). Palladium hydroxide on carbon (0.5 g) was added. The reaction mixture was stirred at room temperature under hydrogen for 12 h. The reaction mixture was filtered through a pad of celite and washed the pad with 1:1 methanol/ethyl acetate. The filtrate and the washings were combined and concentrated to dryness to yield Compound 105a (quantitative). The structure was confirmed by LCMS.

Compound 113 (1.25 g, 2.7 mmol), HBTU (3.2 g, 8.4 mmol) and DIEA (2.8 mL, 16.2 mmol) were dissolved in anhydrous DMF (17 mL) and the reaction mixture was stirred at room temperature for 5 min. To this a solution of Compound 105a (3.77 g, 8.4 mmol) in anhydrous DMF (20 mL) was added. The reaction was stirred at room temperature for 6 h. Solvent was removed under reduced pressure to get an oil. The residue was dissolved in CH 2 Cl 2 (100 mL) and washed with aqueous saturated NaHCO 3 solution (100 mL) and brine (100 mL). The organic phase was separated, dried (Na 2 SO 4 ), filtered and evaporated. The residue was purified by silica gel column chromatography and eluted with 10 to 20% MeOH in dichloromethane to yield Compound 114 (1.45 g, 30%). The structure was confirmed by LCMS and 1 H NMR analysis.

Compound 114 (1.43 g, 0.8 mmol) was dissolved in 1:1 methanol/ethyl acetate (4 mL/4 mL). Palladium on carbon (wet, 0.14 g) was added. The reaction mixture was flushed with hydrogen and stirred at room temperature under hydrogen for 12 h. The reaction mixture was filtered through a pad of celite. The celite pad was washed with methanol/ethyl acetate (1:1). The filtrate and the washings were combined together and evaporated under reduced pressure to yield Compound 115 (quantitative). The structure was confirmed by LCMS and 1 H NMR analysis.

Compound 83a (0.17 g, 0.75 mmol), HBTU (0.31 g, 0.83 mmol) and DIEA (0.26 mL, 1.5 mmol) were dissolved in anhydrous DMF (5 mL) and the reaction mixture was stirred at room temperature for 5 min. To this a solution of Compound 115 (1.22 g, 0.75 mmol) in anhydrous DMF was added and the reaction was stirred at room temperature for 6 h. The solvent was removed under reduced pressure and the residue was dissolved in CH 2 Cl 2 . The organic layer was washed aqueous saturated NaHCO 3 solution and brine and dried over anhydrous Na 2 SO 4 and filtered. The organic layer was concentrated to dryness and the residue obtained was purified by silica gel column chromatography and eluted with 3 to 15% MeOH in dichloromethane to yield Compound 116 (0.84 g, 61%). The structure was confirmed by LC MS and 1 H NMR analysis.

Compound 116 (0.74 g, 0.4 mmol) was dissolved in 1:1 methanol/ethyl acetate (5 mL/5 mL). Palladium on carbon (wet, 0.074 g) was added. The reaction mixture was flushed with hydrogen and stirred at room temperature under hydrogen for 12 h. The reaction mixture was filtered through a pad of celite. The celite pad was washed with methanol/ethyl acetate (1:1). The filtrate and the washings were combined together and evaporated under reduced pressure to yield compound 117 (0.73 g, 98%). The structure was confirmed by LCMS and 1 H NMR analysis.

Compound 117 (0.63 g, 0.36 mmol) was dissolved in anhydrous DMF (3 mL). To this solution N,N-Diisopropylethylamine (70 μL, 0.4 mmol) and pentafluorophenyl trifluoroacetate (72 μL, 0.42 mmol) were added. The reaction mixture was stirred at room temperature for 12 h and poured into a aqueous saturated NaHCO 3 solution. The mixture was extracted with dichloromethane, washed with brine and dried over anhydrous Na 2 SO 4 . The dichloromethane solution was concentrated to dryness and purified with silica gel column chromatography and eluted with 5 to 10% MeOH in dichloromethane to yield compound 118 (0.51 g, 79%). The structure was confirmed by LCMS and 1 H and 1 H and 19 F NMR.

Oligomeric Compound 119, comprising a GalNAc 3 -7 conjugate group, was prepared using the general procedures illustrated in Example 46. The GalNAc 3 cluster portion of the conjugate group GalNAc 3 -7 (GalNAc 3 -7 a ) can be combined with any cleavable moiety to provide a variety of conjugate groups. In certain embodiments, the cleavable moiety is —P(═O)(OH)-A d -P(═O)(OH)—.

The structure of GalNAc 3 -7 (GalNAc 3 -7 a -CM-) is shown below:

Preparation of Oligonucleotide 132 Comprising GalNAc 3 -5

Compound 120 (14.01 g, 40 mmol) and HBTU (14.06 g, 37 mmol) were dissolved in anhydrous DMF (80 mL). Triethylamine (11.2 mL, 80.35 mmol) was added and stirred for 5 min. The reaction mixture was cooled in an ice bath and a solution of compound 121 (10 g, mmol) in anhydrous DMF (20 mL) was added. Additional triethylamine (4.5 mL, 32.28 mmol) was added and the reaction mixture was stirred for 18 h under an argon atmosphere. The reaction was monitored by TLC (ethyl acetate:hexane; 1:1; Rf=0.47). The solvent was removed under reduced pressure. The residue was taken up in EtOAc (300 mL) and washed with 1M NaHSO 4 (3×150 mL), aqueous saturated NaHCO 3 solution (3×150 mL) and brine (2×100 mL). Organic layer was dried with Na 2 SO 4 . Drying agent was removed by filtration and organic layer was concentrated by rotary evaporation. Crude mixture was purified by silica gel column chromatography and eluted by using 35-50% EtOAc in hexane to yield a compound 122 (15.50 g, 78.13%). The structure was confirmed by LCMS and 1 H NMR analysis. Mass m/z 589.3 [M+H] + .

A solution of LiOH (92.15 mmol) in water (20 mL) and THF (10 mL) was added to a cooled solution of Compound 122 (7.75 g, 13.16 mmol) dissolved in methanol (15 mL). The reaction mixture was stirred at room temperature for 45 min. and monitored by TLC (EtOAc:hexane; 1:1). The reaction mixture was concentrated to half the volume under reduced pressure. The remaining solution was cooled an ice bath and neutralized by adding concentrated HCl. The reaction mixture was diluted, extracted with EtOAc (120 mL) and washed with brine (100 mL). An emulsion formed and cleared upon standing overnight. The organic layer was separated dried (Na 2 SO 4 ), filtered and evaporated to yield Compound 123 (8.42 g). Residual salt is the likely cause of excess mass. LCMS is consistent with structure. Product was used without any further purification. M.W.cal: 574.36; M.W.fd: 575.3 [M+H] + .

Compound 126 was synthesized following the procedure described in the literature ( J. Am. Chem. Soc. 2011, 133, 958-963).

Compound 123 (7.419 g, 12.91 mmol), HOBt (3.49 g, 25.82 mmol) and compound 126 (6.33 g, 16.14 mmol) were dissolved in and DMF (40 mL) and the resulting reaction mixture was cooled in an ice bath. To this N,N-Diisopropylethylamine (4.42 mL, 25.82 mmol), PyBop (8.7 g, 16.7 mmol) followed by Bop coupling reagent (1.17 g, 2.66 mmol) were added under an argon atmosphere. The ice bath was removed and the solution was allowed to warm to room temperature. The reaction was completed after 1 h as determined by TLC (DCM:MeOH:AA; 89:10:1). The reaction mixture was concentrated under reduced pressure. The residue was dissolved in EtOAc (200 mL) and washed with 1 M NaHSO 4 (3×100 mL), aqueous saturated NaHCO 3 (3×100 mL) and brine (2×100 mL). The organic phase separated dried (Na 2 SO 4 ), filtered and concentrated. The residue was purified by silica gel column chromatography with a gradient of 50% hexanes/EtOAC to 100% EtOAc to yield Compound 127 (9.4 g) as a white foam. LCMS and 1 H NMR were consistent with structure. Mass m/z 778.4 [M+H] + .

Trifluoroacetic acid (12 mL) was added to a solution of compound 127 (1.57 g, 2.02 mmol) in dichloromethane (12 mL) and stirred at room temperature for 1 h. The reaction mixture was co-evaporated with toluene (30 mL) under reduced pressure to dryness. The residue obtained was co-evaporated twice with acetonitrile (30 mL) and toluene (40 mL) to yield Compound 128 (1.67 g) as trifluoro acetate salt and used for next step without further purification. LCMS and 1 H NMR were consistent with structure. Mass m/z 478.2 [M+H] + .

Compound 7 (0.43 g, 0.963 mmol), HATU (0.35 g, 0.91 mmol), and HOAt (0.035 g, 0.26 mmol) were combined together and dried for 4 h over P 2 O 5 under reduced pressure in a round bottom flask and then dissolved in anhydrous DMF (1 mL) and stirred for 5 min. To this a solution of compound 128 (0.20 g, 0.26 mmol) in anhydrous DMF (0.2 mL) and N,N-Diisopropylethylamine (0.2 mL) was added. The reaction mixture was stirred at room temperature under an argon atmosphere. The reaction was complete after 30 min as determined by LCMS and TLC (7% MeOH/DCM). The reaction mixture was concentrated under reduced pressure. The residue was dissolved in DCM (30 mL) and washed with 1 M NaHSO 4 (3×20 mL), aqueous saturated NaHCO 3 (3×20 mL) and brine (3×20 mL). The organic phase was separated, dried over Na 2 SO 4 , filtered and concentrated. The residue was purified by silica gel column chromatography using 5-15% MeOH in dichloromethane to yield Compound 129 (96.6 mg). LC MS and 1 H NMR are consistent with structure. Mass m/z 883.4 [M+2H] + .

Compound 129 (0.09 g, 0.051 mmol) was dissolved in methanol (5 mL) in 20 mL scintillation vial. To this was added a small amount of 10% Pd/C (0.015 mg) and the reaction vessel was flushed with H 2 gas. The reaction mixture was stirred at room temperature under H 2 atmosphere for 18 h. The reaction mixture was filtered through a pad of Celite and the Celite pad was washed with methanol. The filtrate washings were pooled together and concentrated under reduced pressure to yield Compound 130 (0.08 g). LCMS and 1 H NMR were consistent with structure. The product was used without further purification. Mass m/z 838.3 [M+2H] + .

To a 10 mL pointed round bottom flask were added compound 130 (75.8 mg, 0.046 mmol), 0.37 M pyridine/DMF (200 μL) and a stir bar. To this solution was added 0.7 M pentafluorophenyl trifluoroacetate/DMF (100 μL) drop wise with stirring. The reaction was completed after 1 h as determined by LC MS. The solvent was removed under reduced pressure and the residue was dissolved in CHCl 3 (˜10 mL). The organic layer was partitioned against NaHSO 4 (1 M, 10 mL), aqueous saturated NaHCO 3 (10 mL) and brine (10 mL) three times each. The organic phase separated and dried over Na 2 SO 4 , filtered and concentrated to yield Compound 131 (77.7 mg). LCMS is consistent with structure. Used without further purification. Mass m/z 921.3 [M+2H] + .

Oligomeric Compound 132, comprising a GalNAc 3 -5 conjugate group, was prepared using the general procedures illustrated in Example 46. The GalNAc 3 cluster portion of the conjugate group GalNAc 3 -5 (GalNAc 3 -5 a ) can be combined with any cleavable moiety to provide a variety of conjugate groups. In certain embodiments, the cleavable moiety is —P(═O)(OH)-A d -P(═O)(OH)—.

The structure of GalNAc 3 -5 (GalNAc 3 -5 a -CM-) is shown below:

Preparation of Oligonucleotide 144 Comprising GalNAc 4 -11

Synthesis of Compound 134. To a Merrifield flask was added aminomethyl VIMAD resin (2.5 g, 450 μmol/g) that was washed with acetonitrile, dimethylformamide, dichloromethane and acetonitrile. The resin was swelled in acetonitrile (4 mL). Compound 133 was pre-activated in a 100 mL round bottom flask by adding 20 (1.0 mmol, 0.747 g), TBTU (1.0 mmol, 0.321 g), acetonitrile (5 mL) and DIEA (3.0 mmol, 0.5 mL). This solution was allowed to stir for 5 min and was then added to the Merrifield flask with shaking. The suspension was allowed to shake for 3 h. The reaction mixture was drained and the resin was washed with acetonitrile, DMF and DCM. New resin loading was quantitated by measuring the absorbance of the DMT cation at 500 nm (extinction coefficient=76000) in DCM and determined to be 238 μmol/g. The resin was capped by suspending in an acetic anhydride solution for ten minutes three times.

The solid support bound compound 141 was synthesized using iterative Fmoc-based solid phase peptide synthesis methods. A small amount of solid support was withdrawn and suspended in aqueous ammonia (28-30 wt %) for 6 h. The cleaved compound was analyzed by LC-MS and the observed mass was consistent with structure. Mass m/z 1063.8 [M+2H] + .

The solid support bound compound 142 was synthesized using solid phase peptide synthesis methods.

The solid support bound compound 143 was synthesized using standard solid phase synthesis on a DNA synthesizer.

The solid support bound compound 143 was suspended in aqueous ammonia (28-30 wt %) and heated at 55° C. for 16 h. The solution was cooled and the solid support was filtered. The filtrate was concentrated and the residue dissolved in water and purified by HPLC on a strong anion exchange column. The fractions containing full length compound 144 were pooled together and desalted. The resulting GalNAc 4 -11 conjugated oligomeric compound was analyzed by LC-MS and the observed mass was consistent with structure.

The GalNAc 4 cluster portion of the conjugate group GalNAc 4 -11 (GalNAc 4 -11 a ) can be combined with any cleavable moiety to provide a variety of conjugate groups. In certain embodiments, the cleavable moiety is —P(═O)(OH)-A d -P(═O)(OH)—.

The structure of GalNAc 4 -11 (GalNAc 4 -11 a -CM) is shown below:

Preparation of Oligonucleotide 155 Comprising GalNAc 3 -6

Compound 146 was synthesized as described in the literature ( Analytical Biochemistry 1995, 229, 54-60).

Compound 4 (15 g, 45.55 mmol) and compound 35b (14.3 grams, 57 mmol) were dissolved in CH 2 Cl 2 (200 ml). Activated molecular sieves (4 Å. 2 g, powdered) were added, and the reaction was allowed to stir for 30 minutes under nitrogen atmosphere. TMS-OTf was added (4.1 ml, 22.77 mmol) and the reaction was allowed to stir at room temp overnight. Upon completion, the reaction was quenched by pouring into solution of saturated aqueous NaHCO 3 (500 ml) and crushed ice (˜150 g). The organic layer was separated, washed with brine, dried over MgSO 4 , filtered, and was concentrated to an orange oil under reduced pressure. The crude material was purified by silica gel column chromatography and eluted with 2-10% MeOH in CH 2 Cl 2 to yield Compound 112 (16.53 g, 63%). LCMS and 1 H NMR were consistent with the expected compound.

Compound 112 (4.27 g, 7.35 mmol) was dissolved in 1:1 MeOH/EtOAc (40 ml). The reaction mixture was purged by bubbling a stream of argon through the solution for 15 minutes. Pearlman's catalyst (palladium hydroxide on carbon, 400 mg) was added, and hydrogen gas was bubbled through the solution for 30 minutes. Upon completion (TLC 10% MeOH in CH 2 Cl 2 , and LCMS), the catalyst was removed by filtration through a pad of celite. The filtrate was concentrated by rotary evaporation, and was dried briefly under high vacuum to yield Compound 105a (3.28 g). LCMS and 1H NMR were consistent with desired product.

Compound 147 (2.31 g, 11 mmol) was dissolved in anhydrous DMF (100 mL). N,N-Diisopropylethylamine (DIEA, 3.9 mL, 22 mmol) was added, followed by HBTU (4 g, 10.5 mmol). The reaction mixture was allowed to stir for ˜15 minutes under nitrogen. To this a solution of compound 105a (3.3 g, 7.4 mmol) in dry DMF was added and stirred for 2 h under nitrogen atmosphere. The reaction was diluted with EtOAc and washed with saturated aqueous NaHCO 3 and brine. The organics phase was separated, dried (MgSO 4 ), filtered, and concentrated to an orange syrup. The crude material was purified by column chromatography 2-5% MeOH in CH 2 Cl 2 to yield Compound 148 (3.44 g, 73%). LCMS and 1 H NMR were consistent with the expected product.

Compound 148 (3.3 g, 5.2 mmol) was dissolved in 1:1 MeOH/EtOAc (75 ml). The reaction mixture was purged by bubbling a stream of argon through the solution for 15 minutes. Pearlman's catalyst (palladium hydroxide on carbon) was added (350 mg). Hydrogen gas was bubbled through the solution for 30 minutes. Upon completion (TLC 10% MeOH in DCM, and LCMS), the catalyst was removed by filtration through a pad of celite. The filtrate was concentrated by rotary evaporation, and was dried briefly under high vacuum to yield Compound 149 (2.6 g). LCMS was consistent with desired product. The residue was dissolved in dry DMF (10 ml) was used immediately in the next step.

Compound 146 (0.68 g, 1.73 mmol) was dissolved in dry DMF (20 ml). To this DIEA (450 μL, 2.6 mmol, 1.5 eq.) and HBTU (1.96 g, 0.5.2 mmol) were added. The reaction mixture was allowed to stir for 15 minutes at room temperature under nitrogen. A solution of compound 149 (2.6 g) in anhydrous DMF (10 mL) was added. The pH of the reaction was adjusted to pH=9-10 by addition of DIEA (if necessary). The reaction was allowed to stir at room temperature under nitrogen for 2 h. Upon completion the reaction was diluted with EtOAc (100 mL), and washed with aqueous saturated aqueous NaHCO 3 , followed by brine. The organic phase was separated, dried over MgSO 4 , filtered, and concentrated. The residue was purified by silica gel column chromatography and eluted with 2-10% MeOH in CH 2 Cl 2 to yield Compound 150 (0.62 g, 20%). LCMS and 1 H NMR were consistent with the desired product.

Compound 150 (0.62 g) was dissolved in 1:1 MeOH/EtOAc (5 L). The reaction mixture was purged by bubbling a stream of argon through the solution for 15 minutes. Pearlman's catalyst (palladium hydroxide on carbon) was added (60 mg). Hydrogen gas was bubbled through the solution for 30 minutes. Upon completion (TLC 10% MeOH in DCM, and LCMS), the catalyst was removed by filtration (syringe-tip Teflon filter, 0.45 μm). The filtrate was concentrated by rotary evaporation, and was dried briefly under high vacuum to yield Compound 151 (0.57 g). The LCMS was consistent with the desired product. The product was dissolved in 4 mL dry DMF and was used immediately in the next step.

Compound 83a (0.11 g, 0.33 mmol) was dissolved in anhydrous DMF (5 mL) and N,N-Diisopropylethylamine (75 μL, 1 mmol) and PFP-TFA (90 μL, 0.76 mmol) were added. The reaction mixture turned magenta upon contact, and gradually turned orange over the next 30 minutes. Progress of reaction was monitored by TLC and LCMS. Upon completion (formation of the PFP ester), a solution of compound 151 (0.57 g, 0.33 mmol) in DMF was added. The pH of the reaction was adjusted to pH=9-10 by addition of N,N-Diisopropylethylamine (if necessary). The reaction mixture was stirred under nitrogen for ˜30 min. Upon completion, the majority of the solvent was removed under reduced pressure. The residue was diluted with CH 2 Cl 2 and washed with aqueous saturated NaHCO 3 , followed by brine. The organic phase separated, dried over MgSO 4 , filtered, and concentrated to an orange syrup. The residue was purified by silica gel column chromatography (2-10% MeOH in CH 2 Cl 2 ) to yield Compound 152 (0.35 g, 55%). LCMS and 1 H NMR were consistent with the desired product.

Compound 152 (0.35 g, 0.182 mmol) was dissolved in 1:1 MeOH/EtOAc (10 mL). The reaction mixture was purged by bubbling a stream of argon thru the solution for 15 minutes. Pearlman's catalyst (palladium hydroxide on carbon) was added (35 mg). Hydrogen gas was bubbled thru the solution for 30 minutes. Upon completion (TLC 10% MeOH in DCM, and LCMS), the catalyst was removed by filtration (syringe-tip Teflon filter, 0.45 μm). The filtrate was concentrated by rotary evaporation, and was dried briefly under high vacuum to yield Compound 153 (0.33 g, quantitative). The LCMS was consistent with desired product.

Compound 153 (0.33 g, 0.18 mmol) was dissolved in anhydrous DMF (5 mL) with stirring under nitrogen. To this N,N-Diisopropylethylamine (65 μL, 0.37 mmol) and PFP-TFA (35 μL, 0.28 mmol) were added. The reaction mixture was stirred under nitrogen for ˜30 min. The reaction mixture turned magenta upon contact, and gradually turned orange. The pH of the reaction mixture was maintained at pH=9-10 by adding more N,-Diisopropylethylamine. The progress of the reaction was monitored by TLC and LCMS. Upon completion, the majority of the solvent was removed under reduced pressure. The residue was diluted with CH 2 Cl 2 (50 mL), and washed with saturated aqueous NaHCO 3 , followed by brine. The organic layer was dried over MgSO 4 , filtered, and concentrated to an orange syrup. The residue was purified by column chromatography and eluted with 2-10% MeOH in CH 2 Cl 2 to yield Compound 154 (0.29 g, 79%). LCMS and 1 H NMR were consistent with the desired product.

Oligomeric Compound 155, comprising a GalNAc 3 -6 conjugate group, was prepared using the general procedures illustrated in Example 46. The GalNAc 3 cluster portion of the conjugate group GalNAc 3 -6 (GalNAc 3 -6 a ) can be combined with any cleavable moiety to provide a variety of conjugate groups. In certain embodiments, the cleavable moiety is —P(═O)(OH)-A d -P(═O)(OH)—.

The structure of GalNAc 3 -6 (GalNAc 3 -6 a -CM-) is shown below:

Preparation of Oligonucleotide 160 Comprising GalNAc 3 -9

Compound 156 was synthesized following the procedure described in the literature ( J. Med. Chem. 2004, 47, 5798-5808).

Compound 156, (18.60 g, 29.28 mmol) was dissolved in methanol (200 mL). Palladium on carbon (6.15 g, 10 wt %, loading (dry basis), matrix carbon powder, wet) was added. The reaction mixture was stirred at room temperature under hydrogen for 18 h. The reaction mixture was filtered through a pad of celite and the celite pad was washed thoroughly with methanol. The combined filtrate was washed and concentrated to dryness. The residue was purified by silica gel column chromatography and eluted with 5-10% methanol in dichloromethane to yield Compound 157 (14.26 g, 89%). Mass m/z 544.1 [M−H] − .

Compound 157 (5 g, 9.17 mmol) was dissolved in anhydrous DMF (30 mL). HBTU (3.65 g, 9.61 mmol) and N,N-Diisopropylethylamine (13.73 mL, 78.81 mmol) were added and the reaction mixture was stirred at room temperature for 5 minutes. To this a solution of compound 47 (2.96 g, 7.04 mmol) was added. The reaction was stirred at room temperature for 8 h. The reaction mixture was poured into a saturated NaHCO 3 aqueous solution. The mixture was extracted with ethyl acetate and the organic layer was washed with brine and dried (Na 2 SO 4 ), filtered and evaporated. The residue obtained was purified by silica gel column chromatography and eluted with 50% ethyl acetate in hexane to yield compound 158 (8.25 g, 73.3%). The structure was confirmed by MS and 1 H NMR analysis.

Compound 158 (7.2 g, 7.61 mmol) was dried over P 2 O 5 under reduced pressure. The dried compound was dissolved in anhydrous DMF (50 mL). To this 1H-tetrazole (0.43 g, 6.09 mmol) and N-methylimidazole (0.3 mL, 3.81 mmol) and 2-cyanoethyl-N,N,N′,N′-tetraisopropyl phosphorodiamidite (3.65 mL, 11.50 mmol) were added. The reaction mixture was stirred t under an argon atmosphere for 4 h. The reaction mixture was diluted with ethyl acetate (200 mL). The reaction mixture was washed with saturated NaHCO 3 and brine. The organic phase was separated, dried (Na 2 SO 4 ), filtered and evaporated. The residue was purified by silica gel column chromatography and eluted with 50-90% ethyl acetate in hexane to yield Compound 159 (7.82 g, 80.5%). The structure was confirmed by LCMS and 31 P NMR analysis.

Oligomeric Compound 160, comprising a GalNAc 3 -9 conjugate group, was prepared using standard oligonucleotide synthesis procedures. Three units of compound 159 were coupled to the solid support, followed by nucleotide phosphoramidites. Treatment of the protected oligomeric compound with aqueous ammonia yielded compound 160. The GalNAc 3 cluster portion of the conjugate group GalNAc 3 -9 (GalNAc 3 -9 a ) can be combined with any cleavable moiety to provide a variety of conjugate groups. In certain embodiments, the cleavable moiety is —P(═O)(OH)-A d -P(═O)(OH)—. The structure of GalNAc 3 -9 (GalNAc 3 -9 a -CM) is shown below:

Alternate Procedure for Preparation of Compound 18 (GalNAc 3 -1a and GalNAc 3 -3a)

Lactone 161 was reacted with diamino propane (3-5 eq) or Mono-Boc protected diamino propane (1 eq) to provide alcohol 162a or 162b. When unprotected propanediamine was used for the above reaction, the excess diamine was removed by evaporation under high vacuum and the free amino group in 162a was protected using CbzCl to provide 162b as a white solid after purification by column chromatography. Alcohol 162b was further reacted with compound 4 in the presence of TMSOTf to provide 163a which was converted to 163b by removal of the Cbz group using catalytic hydrogenation. The pentafluorophenyl (PFP) ester 164 was prepared by reacting triacid 113 (see Example 48) with PFPTFA (3.5 eq) and pyridine (3.5 eq) in DMF (0.1 to 0.5 M). The triester 164 was directly reacted with the amine 163b (3-4 eq) and DIPEA (3-4 eq) to provide Compound 18. The above method greatly facilitates purification of intermediates and minimizes the formation of byproducts which are formed using the procedure described in Example 4.

Alternate Procedure for Preparation of Compound 18 (GalNAc 3 -1 a and GalNAc 3 -3a)

The triPFP ester 164 was prepared from acid 113 using the procedure outlined in example 53 above and reacted with mono-Boc protected diamine to provide 165 in essentially quantitative yield. The Boc groups were removed with hydrochloric acid or trifluoroacetic acid to provide the triamine which was reacted with the PFP activated acid 166 in the presence of a suitable base such as DIPEA to provide Compound 18.

The PFP protected Gal-NAc acid 166 was prepared from the corresponding acid by treatment with PFPTFA (1-1.2 eq) and pyridine (1-1.2 eq) in DMF. The precursor acid in turn was prepared from the corresponding alcohol by oxidation using TEMPO (0.2 eq) and BAIB in acetonitrile and water. The precursor alcohol was prepared from sugar intermediate 4 by reaction with 1,6-hexanediol (or 1,5-pentanediol or other diol for other n values) (2-4 eq) and TMSOTf using conditions described previously in example 47.

Dose-dependent Study of Oligonucleotides Comprising Either a 3′ or 5′-Conjugate Group (Comparison of GalNAc 3 -1, 3, 8 and 9) Targeting SRB-1 In Vivo

The oligonucleotides listed below were tested in a dose-dependent study for antisense inhibition of SRB-1 in mice. Unconjugated ISIS 353382 was included as a standard. Each of the various GalNAc 3 conjugate groups was attached at either the 3′ or 5′ terminus of the respective oligonucleotide by a phosphodiester linked 2′-deoxyadenosine nucleoside (cleavable moiety).

Capital letters indicate the nucleobase for each nucleoside and m C indicates a 5-methyl cytosine. Subscripts: “e” indicates a 2′-MOE modified nucleoside; “d” indicates a β-D-2′-deoxyribonucleoside; “s” indicates a phosphorothioate internucleoside linkage (PS); “o” indicates a phosphodiester internucleoside linkage (PO); and “o′” indicates —O—P(═O)(OH)—. Conjugate groups are in bold.

The structure of GalNAc 3 -1 a was shown previously in Example 9. The structure of GalNAc 3 -9 was shown previously in Example 52. The structure of GalNAc 3 -3 was shown previously in Example 39. The structure of GalNAc 3 -8 was shown previously in Example 47.

Treatment

Six week old male Balb/c mice (Jackson Laboratory, Bar Harbor, Me.) were injected subcutaneously once at the dosage shown below with ISIS 353382, 655861, 664078, 661161, 665001 or with saline. Each treatment group consisted of 4 animals. The mice were sacrificed 72 hours following the final administration to determine the liver SRB-1 mRNA levels using real-time PCR and RIBOGREEN® RNA quantification reagent (Molecular Probes, Inc. Eugene, Oreg.) according to standard protocols. The results below are presented as the average percent of SRB-1 mRNA levels for each treatment group, normalized to the saline control.

As illustrated in Table 40, treatment with antisense oligonucleotides lowered SRB-1 mRNA levels in a dose-dependent manner. Indeed, the antisense oligonucleotides comprising the phosphodiester linked GalNAc 3 -1 and GalNAc 3 -9 conjugates at the 3′ terminus (ISIS 655861 and ISIS 664078) and the GalNAc 3 -3 and GalNAc 3 -8 conjugates linked at the 5′ terminus (ISIS 661161 and ISIS 665001) showed substantial improvement in potency compared to the unconjugated antisense oligonucleotide (ISIS 353382). Furthermore, ISIS 664078, comprising a GalNAc 3 -9 conjugate at the 3′ terminus was essentially equipotent compared to ISIS 655861, which comprises a GalNAc 3 -1 conjugate at the 3′ terminus. The 5′ conjugated antisense oligonucleotides, ISIS 661161 and ISIS 665001, comprising a GalNAc 3 -3 or GalNAc 3 -9, respectively, had increased potency compared to the 3′ conjugated antisense oligonucleotides (ISIS 655861 and ISIS 664078).

Liver transaminase levels, alanine aminotransferase (ALT) and aspartate aminotransferase (AST), in serum were measured relative to saline injected mice using standard protocols. Total bilirubin and BUN were also evaluated. The change in body weights was evaluated with no significant change from the saline group. ALTs, ASTs, total bilirubin and BUN values are shown in the table below.

Dose-dependent Study of Oligonucleotides Comprising Either a 3′ or 5′-Conjugate Group (Comparison of GalNAc 3 -1, 2, 3, 5, 6, 7 and 10) Targeting SRB-1 In Vivo

The oligonucleotides listed below were tested in a dose-dependent study for antisense inhibition of SRB-1 in mice. Unconjugated ISIS 353382 was included as a standard. Each of the various GalNAc 3 conjugate groups was attached at the 5′ terminus of the respective oligonucleotide by a phosphodiester linked 2′-deoxyadenosine nucleoside (cleavable moiety) except for ISIS 655861 which had the GalNAc 3 conjugate group attached at the 3′ terminus.

Capital letters indicate the nucleobase for each nucleoside and m C indicates a 5-methyl cytosine. Subscripts: “e” indicates a 2′-MOE modified nucleoside; “d” indicates a β-D-2′-deoxyribonucleoside; “s” indicates a phosphorothioate internucleoside linkage (PS); “o” indicates a phosphodiester internucleoside linkage (PO); and “o′” indicates —O—P(═O)(OH)—. Conjugate groups are in bold.

The structure of GalNAc 3 -1 a was shown previously in Example 9. The structure of GalNAc 3 -2 a was shown previously in Example 37. The structure of GalNAc 3 -3 a was shown previously in Example 39. The structure of GalNAc 3 -5 a was shown previously in Example 49. The structure of GalNAc 3 -6 a was shown previously in Example 51. The structure of GalNAc 3 -7 a was shown previously in Example 48. The structure of GalNAc 3 -10 a was shown previously in Example 46.

Treatment

Six week old male Balb/c mice (Jackson Laboratory, Bar Harbor, Me.) were injected subcutaneously once at the dosage shown below with ISIS 353382, 655861, 664507, 661161, 666224, 666961, 666981, 666881 or with saline. Each treatment group consisted of 4 animals. The mice were sacrificed 72 hours following the final administration to determine the liver SRB-1 mRNA levels using real-time PCR and RIBOGREEN® RNA quantification reagent (Molecular Probes, Inc. Eugene, Oreg.) according to standard protocols. The results below are presented as the average percent of SRB-1 mRNA levels for each treatment group, normalized to the saline control.

As illustrated in Table 43, treatment with antisense oligonucleotides lowered SRB-1 mRNA levels in a dose-dependent manner. Indeed, the conjugated antisense oligonucleotides showed substantial improvement in potency compared to the unconjugated antisense oligonucleotide (ISIS 353382). The 5′ conjugated antisense oligonucleotides showed a slight increase in potency compared to the 3′ conjugated antisense oligonucleotide.

Liver transaminase levels, alanine aminotransferase (ALT) and aspartate aminotransferase (AST), in serum were measured relative to saline injected mice using standard protocols. Total bilirubin and BUN were also evaluated. The change in body weights was evaluated with no significant change from the saline group. ALTs, ASTs, total bilirubin and BUN values are shown in Table 44 below.

Duration of Action Study of Oligonucleotides Comprising a 3′-Conjugate Group Targeting ApoC III In Vivo

Mice were injected once with the doses indicated below and monitored over the course of 42 days for ApoC-III and plasma triglycerides (Plasma TG) levels. The study was performed using 3 transgenic mice that express human APOC-III in each group.

Capital letters indicate the nucleobase for each nucleoside and m C indicates a 5-methyl cytosine. Subscripts: “e” indicates a 2′-MOE modified nucleoside; “d” indicates a β-D-2′-deoxyribonucleoside; “s” indicates a phosphorothioate internucleoside linkage (PS); “o” indicates a phosphodiester internucleoside linkage (PO); and “o′” indicates —O—P(═O)(OH)—. Conjugate groups are in bold.

The structure of GalNAc 3 -1 a was shown previously in Example 9.

As can be seen in the table above the duration of action increased with addition of the 3′-conjugate group compared to the unconjugated oligonucleotide. There was a further increase in the duration of action for the conjugated mixed PO/PS oligonucleotide 647536 as compared to the conjugated full PS oligonucleotide 647535.

Dose-dependent Study of Oligonucleotides Comprising a 3′-Conjugate Group (Comparison of GalNAc 3 -1 and GalNAc 4 -11) Targeting SRB-1 In Vivo

The oligonucleotides listed below were tested in a dose-dependent study for antisense inhibition of SRB-1 in mice. Unconjugated ISIS 440762 was included as an unconjugated standard. Each of the conjugate groups were attached at the 3′ terminus of the respective oligonucleotide by a phosphodiester linked 2′-deoxyadenosine nucleoside cleavable moiety.

The structure of GalNAc 3 -1 a was shown previously in Example 9. The structure of GalNAc 3 -11 a was shown previously in Example 50.

Treatment

Six week old male Balb/c mice (Jackson Laboratory, Bar Harbor, Me.) were injected subcutaneously once at the dosage shown below with ISIS 440762, 651900, 663748 or with saline. Each treatment group consisted of 4 animals. The mice were sacrificed 72 hours following the final administration to determine the liver SRB-1 mRNA levels using real-time PCR and RIBOGREEN® RNA quantification reagent (Molecular Probes, Inc. Eugene, Oreg.) according to standard protocols. The results below are presented as the average percent of SRB-1 mRNA levels for each treatment group, normalized to the saline control.

As illustrated in Table 47, treatment with antisense oligonucleotides lowered SRB-1 mRNA levels in a dose-dependent manner. The antisense oligonucleotides comprising the phosphodiester linked GalNAc 3 -1 and GalNAc 4 -11 conjugates at the 3′ terminus (ISIS 651900 and ISIS 663748) showed substantial improvement in potency compared to the unconjugated antisense oligonucleotide (ISIS 440762). The two conjugated oligonucleotides, GalNAc 3 -1 and GalNAc 4 -11, were equipotent.

Capital letters indicate the nucleobase for each nucleoside and m C indicates a 5-methyl cytosine. Subscripts: “e” indicates a 2′-MOE modified nucleoside; “k” indicates 6′-(S)—CH 3 bicyclic nucleoside; “d” indicates a β-D-2′-deoxyribonucleoside; “s” indicates a phosphorothioate internucleoside linkage (PS); “o” indicates a phosphodiester internucleoside linkage (PO); and “o′” indicates —O—P(═O)(OH)—. Conjugate groups are in bold.

Liver transaminase levels, alanine aminotransferase (ALT) and aspartate aminotransferase (AST), in serum were measured relative to saline injected mice using standard protocols. Total bilirubin and BUN were also evaluated. The change in body weights was evaluated with no significant change from the saline group. ALTs, ASTs, total bilirubin and BUN values are shown in Table 48 below.

Effects of GalNAc 3 -1 Conjugated ASOs Targeting FXI In Vivo

The oligonucleotides listed below were tested in a multiple dose study for antisense inhibition of FXI in mice. ISIS 404071 was included as an unconjugated standard. Each of the conjugate groups was attached at the 3′ terminus of the respective oligonucleotide by a phosphodiester linked 2′-deoxyadenosine nucleoside cleavable moiety.

Capital letters indicate the nucleobase for each nucleoside and m C indicates a 5-methyl cytosine. Subscripts: “e” indicates a 2′-MOE modified nucleoside; “d” indicates a β-D-2′-deoxyribonucleoside; “s” indicates a phosphorothioate internucleoside linkage (PS); “o” indicates a phosphodiester internucleoside linkage (PO); and “o′” indicates —O—P(═O)(OH)—. Conjugate groups are in bold.

The structure of GalNAc 3 -1 a as shown previously in Example 9.

Treatment

Six week old male Balb/c mice (Jackson Laboratory, Bar Harbor, Me.) were injected subcutaneously twice a week for 3 weeks at the dosage shown below with ISIS 404071, 656172, 656173 or with PBS treated control. Each treatment group consisted of 4 animals. The mice were sacrificed 72 hours following the final administration to determine the liver FXI mRNA levels using real-time PCR and RIBOGREEN® RNA quantification reagent (Molecular Probes, Inc. Eugene, Oreg.) according to standard protocols. Plasma FXI protein levels were also measured using ELISA. FXI mRNA levels were determined relative to total RNA (using RIBOGREEN®), prior to normalization to PBS-treated control. The results below are presented as the average percent of FXI mRNA levels for each treatment group. The data was normalized to PBS-treated control and is denoted as “% PBS”. The ED 50 s were measured using similar methods as described previously and are presented below.

As illustrated in Table 50, treatment with antisense oligonucleotides lowered FXI mRNA levels in a dose-dependent manner. The oligonucleotides comprising a 3′-GalNAc 3 -1 conjugate group showed substantial improvement in potency compared to the unconjugated antisense oligonucleotide (ISIS 404071). Between the two conjugated oligonucleotides an improvement in potency was further provided by substituting some of the PS linkages with PO (ISIS 656173).

As illustrated in Table 50a, treatment with antisense oligonucleotides lowered FXI protein levels in a dose-dependent manner. The oligonucleotides comprising a 3′-GalNAc 3 -1 conjugate group showed substantial improvement in potency compared to the unconjugated antisense oligonucleotide (ISIS 404071). Between the two conjugated oligonucleotides an improvement in potency was further provided by substituting some of the PS linkages with PO (ISIS 656173).

Liver transaminase levels, alanine aminotransferase (ALT) and aspartate aminotransferase (AST), in serum were measured relative to saline injected mice using standard protocols. Total bilirubin, total albumin, CRE and BUN were also evaluated. The change in body weights was evaluated with no significant change from the saline group. ALTs, ASTs, total bilirubin and BUN values are shown in the table below.

Effects of Conjugated ASOs Targeting SRB-1 In Vitro

The oligonucleotides listed below were tested in a multiple dose study for antisense inhibition of SRB-1 in primary mouse hepatocytes. ISIS 353382 was included as an unconjugated standard. Each of the conjugate groups were attached at the 3′ or 5′ terminus of the respective oligonucleotide by a phosphodiester linked 2′-deoxyadenosine nucleoside cleavable moiety.

Capital letters indicate the nucleobase for each nucleoside and m C indicates a 5-methyl cytosine. Subscripts: “e” indicates a 2′-MOE modified nucleoside; “d” indicates a β-D-2′-deoxyribonucleoside; “s” indicates a phosphorothioate internucleoside linkage (PS); “o” indicates a phosphodiester internucleoside linkage (PO); and “o′” indicates —O—P(═O)(OH)—. Conjugate groups are in bold.

The structure of GalNAc 3 -1 a was shown previously in Example 9. The structure of GalNAc 3 -3a was shown previously in Example 39. The structure of GalNAc 3 -8a was shown previously in Example 47. The structure of GalNAc 3 -9a was shown previously in Example 52. The structure of GalNAc 3 -6a was shown previously in Example 51. The structure of GalNAc 3 -2a was shown previously in Example 37. The structure of GalNAc 3 -10a was shown previously in Example 46. The structure of GalNAc 3 -5a was shown previously in Example 49. The structure of GalNAc 3 -7a was shown previously in Example 48.

Treatment

The oligonucleotides listed above were tested in vitro in primary mouse hepatocyte cells plated at a density of 25,000 cells per well and treated with 0.03, 0.08, 0.24, 0.74, 2.22, 6.67 or 20 nM modified oligonucleotide. After a treatment period of approximately 16 hours, RNA was isolated from the cells and mRNA levels were measured by quantitative real-time PCR and the SRB-1 mRNA levels were adjusted according to total RNA content, as measured by RIBOGREEN®.

The IC 50 was calculated using standard methods and the results are presented in Table 53. The results show that, under free uptake conditions in which no reagents or electroporation techniques are used to artificially promote entry of the oligonucleotides into cells, the oligonucleotides comprising a GalNAc conjugate were significantly more potent in hepatocytes than the parent oligonucleotide (ISIS 353382) that does not comprise a GalNAc conjugate.

Preparation of Oligomeric Compound 175 Comprising GalNAc 3 -12

Compound 169 is commercially available. Compound 172 was prepared by addition of benzyl (perfluorophenyl) glutarate to compound 171. The benzyl (perfluorophenyl) glutarate was prepared by adding PFP-TFA and DIEA to 5-(benzyloxy)-5-oxopentanoic acid in DMF. Oligomeric compound 175, comprising a GalNAc 3 -12 conjugate group, was prepared from compound 174 using the general procedures illustrated in Example 46. The GalNAc 3 cluster portion of the conjugate group GalNAc 3 -12 (GalNAc 3 -12 a ) can be combined with any cleavable moiety to provide a variety of conjugate groups. In a certain embodiments, the cleavable moiety is —P(═O)(OH)-A d -P(═O)(OH)—. The structure of GalNAc 3 -12 (GalNAc 3 -12 a -CM-) is shown below:

Preparation of Oligomeric Compound 180 Comprising GalNAc 3 -13

Compound 176 was prepared using the general procedure shown in Example 2. Oligomeric compound 180, comprising a GalNAc 3 -13 conjugate group, was prepared from compound 177 using the general procedures illustrated in Example 49. The GalNAc 3 cluster portion of the conjugate group GalNAc 3 -13 (GalNAc 3 -13 a ) can be combined with any cleavable moiety to provide a variety of conjugate groups. In a certain embodiments, the cleavable moiety is —P(═O)(OH)-A d -P(═O)(OH)—. The structure of GalNAc 3 -13 (GalNAc 3 -13 a -CM-) is shown below:

Preparation of Oligomeric Compound 188 Comprising GalNAc 3 -14

Compounds 181 and 185 are commercially available. Oligomeric compound 188, comprising a GalNAc 3 -14 conjugate group, was prepared from compound 187 using the general procedures illustrated in Example 46. The GalNAc 3 cluster portion of the conjugate group GalNAc 3 -14 (GalNAc 3 -14 a ) can be combined with any cleavable moiety to provide a variety of conjugate groups. In certain embodiments, the cleavable moiety is —P(═O)(OH)-A d -P(═O)(OH)—. The structure of GalNAc 3 -14 (GalNAc 3 -14 a -CM-) is shown below:

Preparation of Oligomeric Compound 197 Comprising GalNAc 3 -15

Compound 189 is commercially available. Compound 195 was prepared using the general procedure shown in Example 31. Oligomeric compound 197, comprising a GalNAc 3 -15 conjugate group, was prepared from compounds 194 and 195 using standard oligonucleotide synthesis procedures. The GalNAc 3 cluster portion of the conjugate group GalNAc 3 -15 (GalNAc 3 -15 a ) can be combined with any cleavable moiety to provide a variety of conjugate groups. In certain embodiments, the cleavable moiety is —P(═O)(OH)-A d -P(═O)(OH)—. The structure of GalNAc 3 -15 (GalNAc 3 -15 a -CM-) is shown below:

Dose-dependent Study of Oligonucleotides Comprising a 5′-Conjugate Group (Comparison of GalNAc 3 -3, 12, 13, 14, and 15) Targeting SRB-1 In Vivo

The oligonucleotides listed below were tested in a dose-dependent study for antisense inhibition of SRB-1 in mice. Unconjugated ISIS 353382 was included as a standard. Each of the GalNAc 3 conjugate groups was attached at the 5′ terminus of the respective oligonucleotide by a phosphodiester linked 2′-deoxyadenosine nucleoside (cleavable moiety).

Capital letters indicate the nucleobase for each nucleoside and m C indicates a 5-methyl cytosine. Subscripts: “e” indicates a 2′-MOE modified nucleoside; “d” indicates a β-D-2′-deoxyribonucleoside; “s” indicates a phosphorothioate internucleoside linkage (PS); “o” indicates a phosphodiester internucleoside linkage (PO); and “o′” indicates —O—P(═O)(OH)—. Conjugate groups are in bold.

The structure of GalNAc 3 -3 a was shown previously in Example 39. The structure of GalNAc 3 -12a was shown previously in Example 61. The structure of GalNAc 3 -13a was shown previously in Example 62. The structure of GalNAc 3 -14a was shown previously in Example 63. The structure of GalNAc 3 -15a was shown previously in Example 64.

Treatment

Six to eight week old C57bl6 mice (Jackson Laboratory, Bar Harbor, Me.) were injected subcutaneously once or twice at the dosage shown below with ISIS 353382, 661161, 671144, 670061, 671261, 671262, or with saline. Mice that were dosed twice received the second dose three days after the first dose. Each treatment group consisted of 4 animals. The mice were sacrificed 72 hours following the final administration to determine the liver SRB-1 mRNA levels using real-time PCR and RIBOGREEN® RNA quantification reagent (Molecular Probes, Inc. Eugene, Oreg.) according to standard protocols. The results below are presented as the average percent of SRB-1 mRNA levels for each treatment group, normalized to the saline control.

As illustrated in Table 55, treatment with antisense oligonucleotides lowered SRB-1 mRNA levels in a dose-dependent manner. No significant differences in target knockdown were observed between animals that received a single dose and animals that received two doses (see ISIS 353382 dosages 30 and 2×15 mg/kg; and ISIS 661161 dosages 5 and 2×2.5 mg/kg). The antisense oligonucleotides comprising the phosphodiester linked GalNAc 3 -3, 12, 13, 14, and 15 conjugates showed substantial improvement in potency compared to the unconjugated antisense oligonucleotide (ISIS 335382).

Liver transaminase levels, alanine aminotransferase (ALT) and aspartate aminotransferase (AST), in serum were measured relative to saline injected mice using standard protocols. Total bilirubin and BUN were also evaluated. The changes in body weights were evaluated with no significant differences from the saline group (data not shown). ALTs, ASTs, total bilirubin and BUN values are shown in Table 56 below.

Effect of Various Cleavable Moieties on Antisense Inhibition In Vivo by Oligonucleotides Targeting SRB-1 Comprising a 5′-GalNAc 3 Cluster

The oligonucleotides listed below were tested in a dose-dependent study for antisense inhibition of SRB-1 in mice. Each of the GalNAc 3 conjugate groups was attached at the 5′ terminus oligonucleotide by a phosphodiester linked nucleoside (cleavable moiety (CM)).

Capital letters indicate the nucleobase for each nucleoside and m C indicates a 5-methyl cytosine. Subscripts: “e” indicates a 2′-MOE modified nucleoside; “d” indicates a β-D-2′-deoxyribonucleoside; “s” indicates a phosphorothioate internucleoside linkage (PS); “o” indicates a phosphodiester internucleoside linkage (PO); and “o′” indicates —O—P(═O)(OH)—. Conjugate groups are in bold.

The structure of GalNAc 3 -3 a was shown previously in Example 39. The structure of GalNAc 3 -13a was shown previously in Example 62.

Treatment

Six to eight week old C57bl6 mice (Jackson Laboratory, Bar Harbor, Me.) were injected subcutaneously once at the dosage shown below with ISIS 661161, 670699, 670700, 670701, 671165, or with saline. Each treatment group consisted of 4 animals. The mice were sacrificed 72 hours following the final administration to determine the liver SRB-1 mRNA levels using real-time PCR and RIBOGREEN® RNA quantification reagent (Molecular Probes, Inc. Eugene, Oreg.) according to standard protocols. The results below are presented as the average percent of SRB-1 mRNA levels for each treatment group, normalized to the saline control.

As illustrated in Table 58, treatment with antisense oligonucleotides lowered SRB-1 mRNA levels in a dose-dependent manner. The antisense oligonucleotides comprising various cleavable moieties all showed similar potencies.

Liver transaminase levels, alanine aminotransferase (ALT) and aspartate aminotransferase (AST), in serum were measured relative to saline injected mice using standard protocols. Total bilirubin and BUN were also evaluated. The changes in body weights were evaluated with no significant differences from the saline group (data not shown). ALTs, ASTs, total bilirubin and BUN values are shown in Table 56 below.

Preparation of Oligomeric Compound 199 Comprising GalNAc 3 -16

Oligomeric compound 199, comprising a GalNAc 3 -16 conjugate group, is prepared using the general procedures illustrated in Examples 7 and 9. The GalNAc 3 cluster portion of the conjugate group GalNAc 3 -16 (GalNAc 3 -16 a ) can be combined with any cleavable moiety to provide a variety of conjugate groups. In certain embodiments, the cleavable moiety is —P(═O)(OH)-A d -P(═O)(OH)—. The structure of GalNAc 3 -16 (GalNAc 3 -16 a -CM-) is shown below:

Preparation of Oligomeric Compound 200 Comprising GalNAc 3 -17

Oligomeric compound 200, comprising a GalNAc 3 -17 conjugate group, was prepared using the general procedures illustrated in Example 46. The GalNAc 3 cluster portion of the conjugate group GalNAc 3 -17 (GalNAc 3 -17 a ) can be combined with any cleavable moiety to provide a variety of conjugate groups. In certain embodiments, the cleavable moiety is —P(═O)(OH)-A d -P(═O)(OH)—. The structure of GalNAc 3 -17 (GalNAc 3 -17 a -CM-) is shown below:

Preparation of Oligomeric Compound 201 Comprising GalNAc 3 -18

Oligomeric compound 201, comprising a GalNAc 3 -18 conjugate group, was prepared using the general procedures illustrated in Example 46. The GalNAc 3 cluster portion of the conjugate group GalNAc 3 -18 (GalNAc 3 -18 a ) can be combined with any cleavable moiety to provide a variety of conjugate groups. In certain embodiments, the cleavable moiety is —P(═O)(OH)-A d -P(═O)(OH)—. The structure of GalNAc 3 -18 (GalNAc 3 -18 a -CM-) is shown below:

Preparation of Oligomeric Compound 204 Comprising GalNAc 3 -19

Oligomeric compound 204, comprising a GalNAc 3 -19 conjugate group, was prepared from compound 64 using the general procedures illustrated in Example 52. The GalNAc 3 cluster portion of the conjugate group GalNAc 3 -19 (GalNAc 3 -19 a ) can be combined with any cleavable moiety to provide a variety of conjugate groups. In certain embodiments, the cleavable moiety is —P(═O)(OH)-A d -P(═O)(OH)—. The structure of GalNAc 3 -19 (GalNAc 3 -19 a -CM-) is shown below:

Preparation of Oligomeric Compound 210 Comprising GalNAc 3 -20

Compound 205 was prepared by adding PFP-TFA and DIEA to 6-(2,2,2-trifluoroacetamido)hexanoic acid in acetonitrile, which was prepared by adding triflic anhydride to 6-aminohexanoic acid. The reaction mixture was heated to 80° C., then lowered to rt. Oligomeric compound 210, comprising a GalNAc 3 -20 conjugate group, was prepared from compound 208 using the general procedures illustrated in Example 52. The GalNAc 3 cluster portion of the conjugate group GalNAc 3 -20 (GalNAc 3 -20 a ) can be combined with any cleavable moiety to provide a variety of conjugate groups. In certain embodiments, the cleavable moiety is —P(═O)(OH)-A d -P(═O)(OH)—. The structure of GalNAc 3 -20 (GalNAc 3 -20 a -CM-) is shown below:

Preparation of Oligomeric Compound 215 Comprising GalNAc 3 -21

Compound 211 is commercially available. Oligomeric compound 215, comprising a GalNAc 3 -21 conjugate group, was prepared from compound 213 using the general procedures illustrated in Example 52. The GalNAc 3 cluster portion of the conjugate group GalNAc 3 -21 (GalNAc 3 -21 a ) can be combined with any cleavable moiety to provide a variety of conjugate groups. In certain embodiments, the cleavable moiety is —P(═O)(OH)-A d -P(═O)(OH)—. The structure of GalNAc 3 -21 (GalNAc 3 -21 a -CM-) is shown below:

Preparation of Oligomeric Compound 221 Comprising GalNAc 3 -22

Compound 220 was prepared from compound 219 using diisopropylammonium tetrazolide. Oligomeric compound 221, comprising a GalNAc 3 -21 conjugate group, is prepared from compound 220 using the general procedure illustrated in Example 52. The GalNAc 3 cluster portion of the conjugate group GalNAc 3 -22 (GalNAc 3 -22 a ) can be combined with any cleavable moiety to provide a variety of conjugate groups. In certain embodiments, the cleavable moiety is —P(═O)(OH)-A d -P(═O)(OH)—. The structure of GalNAc 3 -22 (GalNAc 3 -22 a -CM-) is shown below:

Effect of Various Cleavable Moieties on Antisense Inhibition In Vivo by Oligonucleotides Targeting SRB-1 Comprising a 5′-GalNAc 3 Conjugate

The oligonucleotides listed below were tested in a dose-dependent study for antisense inhibition of SRB-1 in mice. Each of the GalNAc 3 conjugate groups was attached at the 5′ terminus of the respective oligonucleotide.

In all tables, capital letters indicate the nucleobase for each nucleoside and m C indicates a 5-methyl cytosine. Subscripts: “e” indicates a 2′-MOE modified nucleoside; “d” indicates a β-D-2′-deoxyribonucleoside; “s” indicates a phosphorothioate internucleoside linkage (PS); “o” indicates a phosphodiester internucleoside linkage (PO); and “o′” indicates —O—P(═O)(OH)—. Conjugate groups are in bold.

The structure of GalNAc 3 -3 a was shown previously in Example 39. The structure of GalNAc 3 -17a was shown previously in Example 68, and the structure of GalNAc 3 -18a was shown in Example 69.

Treatment

Six to eight week old C57BL/6 mice (Jackson Laboratory, Bar Harbor, Me.) were injected subcutaneously once at the dosage shown below with an oligonucleotide listed in Table 60 or with saline. Each treatment group consisted of 4 animals. The mice were sacrificed 72 hours following the final administration to determine the SRB-1 mRNA levels using real-time PCR and RIBOGREEN® RNA quantification reagent (Molecular Probes, Inc. Eugene, Oreg.) according to standard protocols. The results below are presented as the average percent of SRB-1 mRNA levels for each treatment group, normalized to the saline control.

As illustrated in Table 61, treatment with antisense oligonucleotides lowered SRB-1 mRNA levels in a dose-dependent manner. The antisense oligonucleotides comprising a GalNAc conjugate showed similar potencies and were significantly more potent than the parent oligonucleotide lacking a GalNAc conjugate.

Liver transaminase levels, alanine aminotransferase (ALT) and aspartate aminotransferase (AST), in serum were measured relative to saline injected mice using standard protocols. Total bilirubin and BUN were also evaluated. The change in body weights was evaluated with no significant change from the saline group (data not shown). ALTs, ASTs, total bilirubin and BUN values are shown in Table 62 below.

Pharmacokinetic Analysis of Oligonucleotides Comprising a 5′-Conjugate Group

The PK of the ASOs in Tables 54, 57 and 60 above was evaluated using liver samples that were obtained following the treatment procedures described in Examples 65, 66, and 74. The liver samples were minced and extracted using standard protocols and analyzed by IP-HPLC-MS alongside an internal standard. The combined tissue level (μg/g) of all metabolites was measured by integrating the appropriate UV peaks, and the tissue level of the full-length ASO missing the conjugate (“parent,” which is Isis No. 353382 in this case) was measured using the appropriate extracted ion chromatograms (EIC).

The results in Table 63 above show that there were greater liver tissue levels of the oligonucleotides comprising a GalNAc 3 conjugate group than of the parent oligonucleotide that does not comprise a GalNAc 3 conjugate group (ISIS 353382) 72 hours following oligonucleotide administration, particularly when taking into consideration the differences in dosing between the oligonucleotides with and without a GalNAc 3 conjugate group. Furthermore, by 72 hours, 40-98% of each oligonucleotide comprising a GalNAc 3 conjugate group was metabolized to the parent compound, indicating that the GalNAc 3 conjugate groups were cleaved from the oligonucleotides.

Preparation of Oligomeric Compound 230 Comprising GalNAc 3 -23

Compound 222 is commercially available. 44.48 ml (0.33 mol) of compound 222 was treated with tosyl chloride (25.39 g, 0.13 mol) in pyridine (500 mL) for 16 hours. The reaction was then evaporated to an oil, dissolved in EtOAc and washed with water, sat. NaHCO 3 , brine, and dried over Na 2 SO 4 . The ethyl acetate was concentrated to dryness and purified by column chromatography, eluted with EtOAc/hexanes (1:1) followed by 10% methanol in CH 2 Cl 2 to give compound 223 as a colorless oil. LCMS and NMR were consistent with the structure. 10 g (32.86 mmol) of 1-Tosyltriethylene glycol (compound 223) was treated with sodium azide (10.68 g, 164.28 mmol) in DMSO (100 mL) at room temperature for 17 hours. The reaction mixture was then poured onto water, and extracted with EtOAc. The organic layer was washed with water three times and dried over Na 2 SO 4 . The organic layer was concentrated to dryness to give 5.3 g of compound 224 (92%). LCMS and NMR were consistent with the structure. 1-Azidotriethylene glycol (compound 224, 5.53 g, 23.69 mmol) and compound 4 (6 g, 18.22 mmol) were treated with 4A molecular sieves (5 g), and TMSOTf (1.65 ml, 9.11 mmol) in dichloromethane (100 mL) under an inert atmosphere. After 14 hours, the reaction was filtered to remove the sieves, and the organic layer was washed with sat. NaHCO 3 , water, brine, and dried over Na 2 SO 4 . The organic layer was concentrated to dryness and purified by column chromatography, eluted with a gradient of 2 to 4% methanol in dichloromethane to give compound 225. LCMS and NMR were consistent with the structure. Compound 225 (11.9 g, 23.59 mmol) was hydrogenated in EtOAc/Methanol (4:1, 250 mL) over Pearlman's catalyst. After 8 hours, the catalyst was removed by filtration and the solvents removed to dryness to give compound 226. LCMS and NMR were consistent with the structure.

In order to generate compound 227, a solution of nitromethanetrispropionic acid (4.17 g, 15.04 mmol) and Hunig's base (10.3 ml, 60.17 mmol) in DMF (100 mL) were treated dropwise with pentaflourotrifluoro acetate (9.05 ml, 52.65 mmol). After 30 minutes, the reaction was poured onto ice water and extracted with EtOAc. The organic layer was washed with water, brine, and dried over Na 2 SO 4 . The organic layer was concentrated to dryness and then recrystallized from heptane to give compound 227 as a white solid. LCMS and NMR were consistent with the structure. Compound 227 (1.5 g, 1.93 mmol) and compound 226 (3.7 g, 7.74 mmol) were stirred at room temperature in acetonitrile (15 mL) for 2 hours. The reaction was then evaporated to dryness and purified by column chromatography, eluting with a gradient of 2 to 10% methanol in dichloromethane to give compound 228. LCMS and NMR were consistent with the structure. Compound 228 (1.7 g, 1.02 mmol) was treated with Raney Nickel (about 2 g wet) in ethanol (100 mL) in an atmosphere of hydrogen. After 12 hours, the catalyst was removed by filtration and the organic layer was evaporated to a solid that was used directly in the next step. LCMS and NMR were consistent with the structure. This solid (0.87 g, 0.53 mmol) was treated with benzylglutaric acid (0.18 g, 0.8 mmol), HBTU (0.3 g, 0.8 mmol) and DIEA (273.7 μl, 1.6 mmol) in DMF (5 mL). After 16 hours, the DMF was removed under reduced pressure at 65° C. to an oil, and the oil was dissolved in dichloromethane. The organic layer was washed with sat. NaHCO 3 , brine, and dried over Na 2 SO 4 . After evaporation of the organic layer, the compound was purified by column chromatography and eluted with a gradient of 2 to 20% methanol in dichloromethane to give the coupled product. LCMS and NMR were consistent with the structure. The benzyl ester was deprotected with Pearlman's catalyst under a hydrogen atmosphere for 1 hour. The catalyst was them removed by filtration and the solvents removed to dryness to give the acid. LCMS and NMR were consistent with the structure. The acid (486 mg, 0.27 mmol) was dissolved in dry DMF (3 mL). Pyridine (53.61 μl, 0.66 mmol) was added and the reaction was purged with argon. Pentaflourotriflouro acetate (46.39 μl, 0.4 mmol) was slowly added to the reaction mixture. The color of the reaction changed from pale yellow to burgundy, and gave off a light smoke which was blown away with a stream of argon. The reaction was allowed to stir at room temperature for one hour (completion of reaction was confirmed by LCMS). The solvent was removed under reduced pressure (rotovap) at 70° C. The residue was diluted with DCM and washed with 1N NaHSO 4 , brine, saturated sodium bicarbonate and brine again. The organics were dried over Na 2 SO 4 , filtered, and were concentrated to dryness to give 225 mg of compound 229 as a brittle yellow foam. LCMS and NMR were consistent with the structure.

Oligomeric compound 230, comprising a GalNAc 3 -23 conjugate group, was prepared from compound 229 using the general procedure illustrated in Example 46. The GalNAc 3 cluster portion of the GalNAc 3 -23 conjugate group (GalNAc 3 -23 a ) can be combined with any cleavable moiety to provide a variety of conjugate groups. The structure of GalNAc 3 -23 (GalNAc 3 -23 a -CM) is shown below:

Antisense Inhibition In Vivo by Oligonucleotides Targeting SRB-1 Comprising a GalNAc 3 Conjugate

The oligonucleotides listed below were tested in a dose-dependent study for antisense inhibition of SRB-1 in mice.

The structure of GalNAc 3 -1 a was shown previously in Example 9, GalNAc 3 -3 a was shown in Example 39, GalNAc 3 -9a was shown in Example 52, GalNAc 3 -10a was shown in Example 46, GalNAc 3 -19 a was shown in Example 70, GalNAc 3 -20 a was shown in Example 71, and GalNAc 3 -23 a was shown in Example 76.

Treatment

Six to eight week old C57BL/6 mice (Jackson Laboratory, Bar Harbor, Me.) were each injected subcutaneously once at a dosage shown below with an oligonucleotide listed in Table 64 or with saline. Each treatment group consisted of 4 animals. The mice were sacrificed 72 hours following the final administration to determine the SRB-1 mRNA levels using real-time PCR and RIBOGREEN® RNA quantification reagent (Molecular Probes, Inc. Eugene, Oreg.) according to standard protocols. The results below are presented as the average percent of SRB-1 mRNA levels for each treatment group, normalized to the saline control.

As illustrated in Table 65, treatment with antisense oligonucleotides lowered SRB-1 mRNA levels in a dose-dependent manner.

Liver transaminase levels, alanine aminotransferase (ALT) and aspartate aminotransferase (AST), in serum were also measured using standard protocols. Total bilirubin and BUN were also evaluated. Changes in body weights were evaluated, with no significant change from the saline group (data not shown). ALTs, ASTs, total bilirubin and BUN values are shown in Table 66 below.

Antisense Inhibition In Vivo by Oligonucleotides Targeting Angiotensinogen Comprising a GalNAc 3 Conjugate

The oligonucleotides listed below were tested in a dose-dependent study for antisense inhibition of Angiotensinogen (AGT) in normotensive Sprague Dawley rats.

The structure of GalNAc 3 -1 a was shown previously in Example 9.

Treatment

Six week old, male Sprague Dawley rats were each injected subcutaneously once per week at a dosage shown below, for a total of three doses, with an oligonucleotide listed in Table 67 or with PBS. Each treatment group consisted of 4 animals. The rats were sacrificed 72 hours following the final dose. AGT liver mRNA levels were measured using real-time PCR and RIBOGREEN® RNA quantification reagent (Molecular Probes, Inc. Eugene, Oreg.) according to standard protocols. AGT plasma protein levels were measured using the Total Angiotensinogen ELISA (Catalog #JP27412, IBL International, Toronto, ON) with plasma diluted 1:20,000. The results below are presented as the average percent of AGT mRNA levels in liver or AGT protein levels in plasma for each treatment group, normalized to the PBS control.

As illustrated in Table 68, treatment with antisense oligonucleotides lowered AGT liver mRNA and plasma protein levels in a dose-dependent manner, and the oligonucleotide comprising a GalNAc conjugate was significantly more potent than the parent oligonucleotide lacking a GalNAc conjugate.

Liver transaminase levels, alanine aminotransferase (ALT) and aspartate aminotransferase (AST), in plasma and body weights were also measured at time of sacrifice using standard protocols. The results are shown in Table 69 below.

Duration of Action In Vivo of Oligonucleotides Targeting APOC-III Comprising a GalNAc 3 Conjugate

The oligonucleotides listed in Table 70 below were tested in a single dose study for duration of action in mice.

The structure of GalNAc 3 -1 a was shown previously in Example 9, GalNAc 3 -3 a was shown in Example 39, GalNAc 3 -7 a was shown in Example 48, GalNAc 3 -10 a was shown in Example 46, and GalNAc 3 -13 a was shown in Example 62.

Treatment

Six to eight week old transgenic mice that express human APOC-III were each injected subcutaneously once with an oligonucleotide listed in Table 70 or with PBS. Each treatment group consisted of 3 animals. Blood was drawn before dosing to determine baseline and at 72 hours, 1 week, 2 weeks, 3 weeks, 4 weeks, 5 weeks, and 6 weeks following the dose. Plasma triglyceride and APOC-III protein levels were measured as described in Example 20. The results below are presented as the average percent of plasma triglyceride and APOC-III levels for each treatment group, normalized to baseline levels, showing that the oligonucleotides comprising a GalNAc conjugate group exhibited a longer duration of action than the parent oligonucleotide without a conjugate group (ISIS 304801) even though the dosage of the parent was three times the dosage of the oligonucleotides comprising a GalNAc conjugate group.

Antisense Inhibition In Vivo by Oligonucleotides Targeting Alpha-1 Antitrypsin (A1AT) Comprising a GalNAc 3 Conjugate

The oligonucleotides listed in Table 72 below were tested in a study for dose-dependent inhibition of A1AT in mice.

The structure of GalNAc 3 -1 a was shown previously in Example 9, GalNAc 3 -3 a was shown in Example 39, GalNAc 3 -7 a was shown in Example 48, GalNAc 3 -10 a was shown in Example 46, and GalNAc 3 -13 a was shown in Example 62.

Treatment

Six week old, male C57BL/6 mice (Jackson Laboratory, Bar Harbor, Me.) were each injected subcutaneously once per week at a dosage shown below, for a total of three doses, with an oligonucleotide listed in Table 72 or with PBS. Each treatment group consisted of 4 animals. The mice were sacrificed 72 hours following the final administration. A1AT liver mRNA levels were determined using real-time PCR and RIBOGREEN® RNA quantification reagent (Molecular Probes, Inc. Eugene, Oreg.) according to standard protocols. A1AT plasma protein levels were determined using the Mouse Alpha 1-Antitrypsin ELISA (catalog #41-A1AMS-E01, Alpco, Salem, N.H.). The results below are presented as the average percent of A1AT liver mRNA and plasma protein levels for each treatment group, normalized to the PBS control.

As illustrated in Table 73, treatment with antisense oligonucleotides lowered A1AT liver mRNA and A1AT plasma protein levels in a dose-dependent manner. The oligonucleotides comprising a GalNAc conjugate were significantly more potent than the parent (ISIS 476366).

Liver transaminase and BUN levels in plasma were measured at time of sacrifice using standard protocols. Body weights and organ weights were also measured. The results are shown in Table 74 below. Body weight is shown as % relative to baseline. Organ weights are shown as % of body weight relative to the PBS control group.

Duration of Action In Vivo of Oligonucleotides Targeting A1AT Comprising a GalNAc 3 Cluster

The oligonucleotides listed in Table 72 were tested in a single dose study for duration of action in mice.

Treatment

Six week old, male C57BL/6 mice were each injected subcutaneously once with an oligonucleotide listed in Table 72 or with PBS. Each treatment group consisted of 4 animals. Blood was drawn the day before dosing to determine baseline and at 5, 12, 19, and 25 days following the dose. Plasma A1AT protein levels were measured via ELISA (see Example 80). The results below are presented as the average percent of plasma A1AT protein levels for each treatment group, normalized to baseline levels. The results show that the oligonucleotides comprising a GalNAc conjugate were more potent and had longer duration of action than the parent lacking a GalNAc conjugate (ISIS 476366). Furthermore, the oligonucleotides comprising a 5′-GalNAc conjugate (ISIS 678381, 678382, 678383, and 678384) were generally even more potent with even longer duration of action than the oligonucleotide comprising a 3′-GalNAc conjugate (ISIS 656326).

Antisense Inhibition In Vitro by Oligonucleotides Targeting SRB-1 Comprising a GalNAc 3 Conjugate

Primary mouse liver hepatocytes were seeded in 96 well plates at 15,000 cells/well 2 hours prior to treatment. The oligonucleotides listed in Table 76 were added at 2, 10, 50, or 250 nM in Williams E medium and cells were incubated overnight at 37° C. in 5% CO 2 . Cells were lysed 16 hours following oligonucleotide addition, and total RNA was purified using RNease 3000 BioRobot (Qiagen). SRB-1 mRNA levels were determined using real-time PCR and RIBOGREEN® RNA quantification reagent (Molecular Probes, Inc. Eugene, Oreg.) according to standard protocols. IC 50 values were determined using Prism 4 software (GraphPad). The results show that oligonucleotides comprising a variety of different GalNAc conjugate groups and a variety of different cleavable moieties are significantly more potent in an in vitro free uptake experiment than the parent oligonucleotides lacking a GalNAc conjugate group (ISIS 353382 and 666841).

The structure of GalNAc 3 -1 a was shown previously in Example 9, GalNAc 3 -3 a was shown in Example 39, GalNAc 3 -5 a was shown in Example 49, GalNAc 3 -6 a was shown in Example 51, GalNAc 3 -7 a was shown in Example 48, GalNAc 3 -8 a was shown in Example 47, GalNAc 3 -9 a was shown in Example 52, GalNAc 3 -10 a was shown in Example 46, GalNAc 3 -12 a was shown in Example 61, GalNAc 3 -13 a was shown in Example 62, GalNAc 3 -14 a was shown in Example 63, GalNAc 3 -15 a was shown in Example 64, GalNAc 3 -17 a was shown in Example 68, GalNAc 3 -18 a was shown in Example 69, GalNAc 3 -19 a was shown in Example 70, GalNAc 3 -20 a was shown in Example 71, and GalNAc 3 -23 a was shown in Example 76.

Antisense Inhibition In Vivo by Oligonucleotides Targeting Factor XI Comprising a GalNAc 3 Cluster

The oligonucleotides listed in Table 77 below were tested in a study for dose-dependent inhibition of Factor XI in mice.

The structure of GalNAc 3 -1 a was shown previously in Example 9, GalNAc 3 -3 a was shown in Example 39, GalNAc 3 -7 a was shown in Example 48, GalNAc 3 -10 a was shown in Example 46, and GalNAc 3 -13 a was shown in Example 62.

Treatment

Six to eight week old mice were each injected subcutaneously once per week at a dosage shown below, for a total of three doses, with an oligonucleotide listed below or with PBS. Each treatment group consisted of 4 animals. The mice were sacrificed 72 hours following the final dose. Factor XI liver mRNA levels were measured using real-time PCR and normalized to cyclophilin according to standard protocols. Liver transaminases, BUN, and bilirubin were also measured. The results below are presented as the average percent for each treatment group, normalized to the PBS control.

As illustrated in Table 78, treatment with antisense oligonucleotides lowered Factor XI liver mRNA in a dose-dependent manner. The results show that the oligonucleotides comprising a GalNAc conjugate were more potent than the parent lacking a GalNAc conjugate (ISIS 404071). Furthermore, the oligonucleotides comprising a 5′-GalNAc conjugate (ISIS 663086, 678347, 678348, and 678349) were even more potent than the oligonucleotide comprising a 3′-GalNAc conjugate (ISIS 656173).

Duration of Action In Vivo of Oligonucleotides Targeting Factor XI Comprising a GalNAc 3 Conjugate

The oligonucleotides listed in Table 77 were tested in a single dose study for duration of action in mice.

Treatment

Six to eight week old mice were each injected subcutaneously once with an oligonucleotide listed in Table 77 or with PBS. Each treatment group consisted of 4 animals. Blood was drawn by tail bleeds the day before dosing to determine baseline and at 3, 10, and 17 days following the dose. Plasma Factor XI protein levels were measured by ELISA using Factor XI capture and biotinylated detection antibodies from R & D Systems, Minneapolis, Minn. (catalog #AF2460 and #BAF2460, respectively) and the OptEIA Reagent Set B (Catalog #550534, BD Biosciences, San Jose, Calif.). The results below are presented as the average percent of plasma Factor XI protein levels for each treatment group, normalized to baseline levels. The results show that the oligonucleotides comprising a GalNAc conjugate were more potent with longer duration of action than the parent lacking a GalNAc conjugate (ISIS 404071). Furthermore, the oligonucleotides comprising a 5′-GalNAc conjugate (ISIS 663086, 678347, 678348, and 678349) were even more potent with an even longer duration of action than the oligonucleotide comprising a 3′-GalNAc conjugate (ISIS 656173).

Antisense Inhibition In Vivo by Oligonucleotides Targeting SRB-1 Comprising a GalNAc 3 Conjugate

Oligonucleotides listed in Table 76 were tested in a dose-dependent study for antisense inhibition of SRB-1 in mice.

Treatment

Six to eight week old C57BL/6 mice were each injected subcutaneously once per week at a dosage shown below, for a total of three doses, with an oligonucleotide listed in Table 76 or with saline. Each treatment group consisted of 4 animals. The mice were sacrificed 48 hours following the final administration to determine the SRB-1 mRNA levels using real-time PCR and RIBOGREEN® RNA quantification reagent (Molecular Probes, Inc. Eugene, Oreg.) according to standard protocols. The results below are presented as the average percent of liver SRB-1 mRNA levels for each treatment group, normalized to the saline control.

As illustrated in Tables 80 and 81, treatment with antisense oligonucleotides lowered SRB-1 mRNA levels in a dose-dependent manner.

Liver transaminase levels, total bilirubin, BUN, and body weights were also measured using standard protocols. Average values for each treatment group are shown in Table 82 below.

Antisense Inhibition In Vivo by Oligonucleotides Targeting TTR Comprising a GalNAc 3 Cluster

Oligonucleotides listed in Table 83 below were tested in a dose-dependent study for antisense inhibition of human transthyretin (TTR) in transgenic mice that express the human TTR gene.

Treatment

Eight week old TTR transgenic mice were each injected subcutaneously once per week for three weeks, for a total of three doses, with an oligonucleotide and dosage listed in the tables below or with PBS. Each treatment group consisted of 4 animals. The mice were sacrificed 72 hours following the final administration. Tail bleeds were performed at various time points throughout the experiment, and plasma TTR protein, ALT, and AST levels were measured and reported in Tables 85-87. After the animals were sacrificed, plasma ALT, AST, and human TTR levels were measured, as were body weights, organ weights, and liver human TTR mRNA levels. TTR protein levels were measured using a clinical analyzer (AU480, Beckman Coulter, CA). Real-time PCR and RIBOGREEN® RNA quantification reagent (Molecular Probes, Inc. Eugene, Oreg.) were used according to standard protocols to determine liver human TTR mRNA levels. The results presented in Tables 84-87 are the average values for each treatment group. The mRNA levels are the average values relative to the average for the PBS group. Plasma protein levels are the average values relative to the average value for the PBS group at baseline. Body weights are the average percent weight change from baseline until sacrifice for each individual treatment group. Organ weights shown are normalized to the animal's body weight, and the average normalized organ weight for each treatment group is then presented relative to the average normalized organ weight for the PBS group.

In Tables 84-87, “BL” indicates baseline, measurements that were taken just prior to the first dose. As illustrated in Tables 84 and 85, treatment with antisense oligonucleotides lowered TTR expression levels in a dose-dependent manner. The oligonucleotides comprising a GalNAc conjugate were more potent than the parent lacking a GalNAc conjugate (ISIS 420915). Furthermore, the oligonucleotides comprising a GalNAc conjugate and mixed PS/PO internucleoside linkages were even more potent than the oligonucleotide comprising a GalNAc conjugate and full PS linkages.

The legend for Table 85 can be found in Example 74. The structure of GalNAc 3 -1 was shown in Example 9. The structure of GalNAc 3 -3 a was shown in Example 39. The structure of GalNAc 3 -7 a was shown in Example 48. The structure of GalNAc 3 -10 a was shown in Example 46. The structure of GalNAc 3 -13 a was shown in Example 62. The structure of GalNAc 3 -19 a was shown in Example 70.

Duration of Action In Vivo by Single Doses of Oligonucleotides Targeting TTR Comprising a GalNAc 3 Cluster

ISIS numbers 420915 and 660261 (see Table 83) were tested in a single dose study for duration of action in mice. ISIS numbers 420915, 682883, and 682885 (see Table 83) were also tested in a single dose study for duration of action in mice.

Treatment

Eight week old, male transgenic mice that express human TTR were each injected subcutaneously once with 100 mg/kg ISIS No. 420915 or 13.5 mg/kg ISIS No. 660261. Each treatment group consisted of 4 animals. Tail bleeds were performed before dosing to determine baseline and at days 3, 7, 10, 17, 24, and 39 following the dose. Plasma TTR protein levels were measured as described in Example 86. The results below are presented as the average percent of plasma TTR levels for each treatment group, normalized to baseline levels.

Treatment

Female transgenic mice that express human TTR were each injected subcutaneously once with 100 mg/kg ISIS No. 420915, 10.0 mg/kg ISIS No. 682883, or 10.0 mg/kg 682885. Each treatment group consisted of 4 animals. Tail bleeds were performed before dosing to determine baseline and at days 3, 7, 10, 17, 24, and 39 following the dose. Plasma TTR protein levels were measured as described in Example 86. The results below are presented as the average percent of plasma TTR levels for each treatment group, normalized to baseline levels.

The results in Tables 88 and 89 show that the oligonucleotides comprising a GalNAc conjugate are more potent with a longer duration of action than the parent oligonucleotide lacking a conjugate (ISIS 420915).

Splicing Modulation In Vivo by Oligonucleotides Targeting SMN Comprising a GalNAc 3 Conjugate

The oligonucleotides listed in Table 90 were tested for splicing modulation of human survival of motor neuron (SMN) in mice.

The structure of GalNAc 3 -7 a was shown previously in Example 48. “X” indicates a 5′ primary amine generated by Gene Tools (Philomath, Oreg.), and GalNAc 3 -7 b indicates the structure of GalNAc 3 -7 a lacking the —NH—C 6 —O portion of the linker as shown below:

ISIS numbers 703421 and 703422 are morpholino oligonucleotides, wherein each nucleotide of the two oligonucleotides is a morpholino nucleotide.

Treatment

Six week old transgenic mice that express human SMN were injected subcutaneously once with an oligonucleotide listed in Table 91 or with saline. Each treatment group consisted of 2 males and 2 females. The mice were sacrificed 3 days following the dose to determine the liver human SMN mRNA levels both with and without exon 7 using real-time PCR according to standard protocols. Total RNA was measured using Ribogreen reagent. The SMN mRNA levels were normalized to total mRNA, and further normalized to the averages for the saline treatment group. The resulting average ratios of SMN mRNA including exon 7 to SMN mRNA missing exon 7 are shown in Table 91. The results show that fully modified oligonucleotides that modulate splicing and comprise a GalNAc conjugate are significantly more potent in altering splicing in the liver than the parent oligonucleotides lacking a GlaNAc conjugate. Furthermore, this trend is maintained for multiple modification chemistries, including 2′-MOE and morpholino modified oligonucleotides.

Antisense Inhibition In Vivo by Oligonucleotides Targeting Apolipoprotein A (Apo(a)) Comprising a GalNAc 3 Conjugate

The oligonucleotides listed in Table 92 below were tested in a study for dose-dependent inhibition of Apo(a) in transgenic mice.

Treatment

Eight week old, female C57BL/6 mice (Jackson Laboratory, Bar Harbor, Me.) were each injected subcutaneously once per week at a dosage shown below, for a total of six doses, with an oligonucleotide listed in Table 92 or with PBS. Each treatment group consisted of 3-4 animals. Tail bleeds were performed the day before the first dose and weekly following each dose to determine plasma Apo(a) protein levels. The mice were sacrificed two days following the final administration. Apo(a) liver mRNA levels were determined using real-time PCR and RIBOGREEN® RNA quantification reagent (Molecular Probes, Inc. Eugene, Oreg.) according to standard protocols. Apo(a) plasma protein levels were determined using ELISA, and liver transaminase levels were determined. The mRNA and plasma protein results in Table 93 are presented as the treatment group average percent relative to the PBS treated group. Plasma protein levels were further normalized to the baseline (BL) value for the PBS group. Average absolute transaminase levels and body weights (% relative to baseline averages) are reported in Table 94.

As illustrated in Table 93, treatment with the oligonucleotides lowered Apo(a) liver mRNA and plasma protein levels in a dose-dependent manner. Furthermore, the oligonucleotide comprising the GalNAc conjugate was significantly more potent with a longer duration of action than the parent oligonucleotide lacking a GalNAc conjugate. As illustrated in Table 94, transaminase levels and body weights were unaffected by the oligonucleotides, indicating that the oligonucleotides were well tolerated.

Antisense Inhibition In Vivo by Oligonucleotides Targeting TTR Comprising a GalNAc 3 Cluster

Oligonucleotides listed in Table 95 below were tested in a dose-dependent study for antisense inhibition of human transthyretin (TTR) in transgenic mice that express the human TTR gene.

Treatment

TTR transgenic mice were each injected subcutaneously once per week for three weeks, for a total of three doses, with an oligonucleotide and dosage listed in Table 96 or with PBS. Each treatment group consisted of 4 animals. Prior to the first dose, a tail bleed was performed to determine plasma TTR protein levels at baseline (BL). The mice were sacrificed 72 hours following the final administration. TTR protein levels were measured using a clinical analyzer (AU480, Beckman Coulter, CA). Real-time PCR and RIBOGREEN® RNA quantification reagent (Molecular Probes, Inc. Eugene, Oreg.) were used according to standard protocols to determine liver human TTR mRNA levels. The results presented in Table 96 are the average values for each treatment group. The mRNA levels are the average values relative to the average for the PBS group. Plasma protein levels are the average values relative to the average value for the PBS group at baseline. “BL” indicates baseline, measurements that were taken just prior to the first dose. As illustrated in Table 96, treatment with antisense oligonucleotides lowered TTR expression levels in a dose-dependent manner. The oligonucleotides comprising a GalNAc conjugate were more potent than the parent lacking a GalNAc conjugate (ISIS 420915), and oligonucleotides comprising a phosphodiester or deoxyadenosine cleavable moiety showed significant improvements in potency compared to the parent lacking a conjugate (see ISIS numbers 682883 and 666943 vs 420915 and see Examples 86 and 87).

The legend for Table 95 can be found in Example 74. The structure of GalNAc 3 -3 a was shown in Example 39. The structure of GalNAc 3 -7 a was shown in Example 48. The structure of GalNAc 3 -10 a was shown in Example 46. The structure of GalNAc 3 -13 a was shown in Example 62.

Antisense Inhibition In Vivo by Oligonucleotides Targeting Factor VII Comprising a GalNAc 3 Conjugate in Non-human Primates

Oligonucleotides listed in Table 97 below were tested in a non-terminal, dose escalation study for antisense inhibition of Factor VII in monkeys.

Treatment

Non-naïve monkeys were each injected subcutaneously on days 0, 15, and 29 with escalating doses of an oligonucleotide listed in Table 97 or with PBS. Each treatment group consisted of 4 males and 1 female. Prior to the first dose and at various time points thereafter, blood draws were performed to determine plasma Factor VII protein levels. Factor VII protein levels were measured by ELISA. The results presented in Table 98 are the average values for each treatment group relative to the average value for the PBS group at baseline (BL), the measurements taken just prior to the first dose. As illustrated in Table 98, treatment with antisense oligonucleotides lowered Factor VII expression levels in a dose-dependent manner, and the oligonucleotide comprising the GalNAc conjugate was significantly more potent in monkeys compared to the oligonucleotide lacking a GalNAc conjugate.

The legend for Table 97 can be found in Example 74. The structure of GalNAc 3 -10 a was shown in Example 46.

Antisense Inhibition in Primary Hepatocytes by Antisense Oligonucleotides Targeting ApoCIII Comprising a GalNAc 3 Conjugate

Primary mouse hepatocytes were seeded in 96-well plates at 15,000 cells per well, and the oligonucleotides listed in Table 99, targeting mouse ApoC-III, were added at 0.46, 1.37, 4.12, or 12.35, 37.04, 111.11, or 333.33 nM or 1.00 μM. After incubation with the oligonucleotides for 24 hours, the cells were lysed and total RNA was purified using RNeasy (Qiagen). ApoC-III mRNA levels were determined using real-time PCR and RIBOGREEN® RNA quantification reagent (Molecular Probes, Inc.) according to standard protocols. IC 50 values were determined using Prism 4 software (GraphPad). The results show that regardless of whether the cleavable moiety was a phosphodiester or a phosphodiester-linked deoxyadensoine, the oligonucleotides comprising a GalNAc conjugate were significantly more potent than the parent oligonucleotide lacking a conjugate.

The structure of GalNAc 3 -1 a was shown previously in Example 9, GalNAc 3 -3 a was shown in Example 39, GalNAc 3 -7 a was shown in Example 48, GalNAc 3 -10 a was shown in Example 46, GalNAc 3 -13 a was shown in Example 62, and GalNAc 3 -19 a was shown in Example 70.

Antisense Inhibition In Vivo by Oligonucleotides Targeting SRB-1 Comprising Mixed Wings and a 5′-GalNAc 3 Conjugate

The oligonucleotides listed in Table 100 were tested in a dose-dependent study for antisense inhibition of SRB-1 in mice.

The structure of GalNAc 3 -3 a was shown previously in Example 39, and the structure of GalNAc 3 -7a was shown previously in Example 48. Subscripts: “e” indicates 2′-MOE modified nucleoside; “d” indicates β-D-2′-deoxyribonucleoside; “k” indicates 6′-(S)—CH 3 bicyclic nucleoside (cEt); “s” indicates phosphorothioate internucleoside linkages (PS); “o” indicates phosphodiester internucleoside linkages (PO). Superscript “m” indicates 5-methylcytosines.

Treatment

Six to eight week old C57BL/6 mice (Jackson Laboratory, Bar Harbor, Me.) were injected subcutaneously once at the dosage shown below with an oligonucleotide listed in Table 100 or with saline. Each treatment group consisted of 4 animals. The mice were sacrificed 72 hours following the final administration. Liver SRB-1 mRNA levels were measured using real-time PCR. SRB-1 mRNA levels were normalized to cyclophilin mRNA levels according to standard protocols. The results are presented as the average percent of SRB-1 mRNA levels for each treatment group relative to the saline control group. As illustrated in Table 101, treatment with antisense oligonucleotides lowered SRB-1 mRNA levels in a dose-dependent manner, and the gapmer oligonucleotides comprising a GalNAc conjugate and having wings that were either full cEt or mixed sugar modifications were significantly more potent than the parent oligonucleotide lacking a conjugate and comprising full cEt modified wings.

Body weights, liver transaminases, total bilirubin, and BUN were also measured, and the average values for each treatment group are shown in Table 101. Body weight is shown as the average percent body weight relative to the baseline body weight (% BL) measured just prior to the oligonucleotide dose.

Antisense Inhibition In Vivo by Oligonucleotides Targeting SRB-1 Comprising 2′-Sugar Modifications and a 5′-GalNAc 3 Conjugate

The oligonucleotides listed in Table 102 were tested in a dose-dependent study for antisense inhibition of SRB-1 in mice.

Subscript “m” indicates a 2′-O-methyl modified nucleoside. See Example 74 for complete table legend. The structure of GalNAc 3 -3 a was shown previously in Example 39, and the structure of GalNAc 3 -7a was shown previously in Example 48.

Treatment

The study was completed using the protocol described in Example 93. Results are shown in Table 103 below and show that both the 2′-MOE and 2′-OMe modified oligonucleotides comprising a GalNAc conjugate were significantly more potent than the respective parent oligonucleotides lacking a conjugate. The results of the body weights, liver transaminases, total bilirubin, and BUN measurements indicated that the compounds were all well tolerated.

Antisense Inhibition In Vivo by Oligonucleotides Targeting SRB-1 Comprising Bicyclic Nucleosides and a 5′-GalNAc 3 Conjugate

The oligonucleotides listed in Table 104 were tested in a dose-dependent study for antisense inhibition of SRB-1 in mice.

Subscript “g” indicates a fluoro-HNA nucleoside, subscript “1” indicates a locked nucleoside comprising a 2′-O—CH 2 -4′ bridge. See the Example 74 table legend for other abbreviations. The structure of GalNAc 3 -1 a was shown previously in Example 9, the structure of GalNAc 3 -3 a was shown previously in Example 39, and the structure of GalNAc 3 -7a was shown previously in Example 48.

Treatment

The study was completed using the protocol described in Example 93. Results are shown in Table 105 below and show that oligonucleotides comprising a GalNAc conjugate and various bicyclic nucleoside modifications were significantly more potent than the parent oligonucleotide lacking a conjugate and comprising bicyclic nucleoside modifications. Furthermore, the oligonucleotide comprising a GalNAc conjugate and fluoro-HNA modifications was significantly more potent than the parent lacking a conjugate and comprising fluoro-HNA modifications. The results of the body weights, liver transaminases, total bilirubin, and BUN measurements indicated that the compounds were all well tolerated.

Plasma Protein Binding of Antisense Oligonucleotides Comprising a GalNAc 3 Conjugate Group

Oligonucleotides listed in Table 70 targeting ApoC-III and oligonucleotides in Table 106 targeting Apo(a) were tested in an ultra-filtration assay in order to assess plasma protein binding.

See the Example 74 for table legend. The structure of GalNAc 3 -7a was shown previously in Example 48.

Ultrafree-MC ultrafiltration units (30,000 NMWL, low-binding regenerated cellulose membrane, Millipore, Bedford, Mass.) were pre-conditioned with 300 μL of 0.5% Tween 80 and centrifuged at 2000 g for 10 minutes, then with 300 μL of a 300 μg/mL solution of a control oligonucleotide in H 2 O and centrifuged at 2000 g for 16 minutes. In order to assess non-specific binding to the filters of each test oligonucleotide from Tables 70 and 106 to be used in the studies, 300 μL of a 250 ng/mL solution of oligonucleotide in H 2 O at pH 7.4 was placed in the pre-conditioned filters and centrifuged at 2000 g for 16 minutes. The unfiltered and filtered samples were analyzed by an ELISA assay to determine the oligonucleotide concentrations. Three replicates were used to obtain an average concentration for each sample. The average concentration of the filtered sample relative to the unfiltered sample is used to determine the percent of oligonucleotide that is recovered through the filter in the absence of plasma (% recovery).

Frozen whole plasma samples collected in K3-EDTA from normal, drug-free human volunteers, cynomolgus monkeys, and CD-1 mice, were purchased from Bioreclamation LLC (Westbury, N.Y.). The test oligonucleotides were added to 1.2 mL aliquots of plasma at two concentrations (5 and 150 μg/mL). An aliquot (300 μL) of each spiked plasma sample was placed in a pre-conditioned filter unit and incubated at 37° C. for 30 minutes, immediately followed by centrifugation at 2000 g for 16 minutes. Aliquots of filtered and unfiltered spiked plasma samples were analyzed by an ELISA to determine the oligonucleotide concentration in each sample. Three replicates per concentration were used to determine the average percentage of bound and unbound oligonucleotide in each sample. The average concentration of the filtered sample relative to the concentration of the unfiltered sample is used to determine the percent of oligonucleotide in the plasma that is not bound to plasma proteins (% unbound). The final unbound oligonucleotide values are corrected for non-specific binding by dividing the % unbound by the % recovery for each oligonucleotide. The final % bound oligonucleotide values are determined by subtracting the final % unbound values from 100. The results are shown in Table 107 for the two concentrations of oligonucleotide tested (5 and 150 μg/mL) in each species of plasma. The results show that GalNAc conjugate groups do not have a significant impact on plasma protein binding. Furthermore, oligonucleotides with full PS internucleoside linkages and mixed PO/PS linkages both bind plasma proteins, and those with full PS linkages bind plasma proteins to a somewhat greater extent than those with mixed PO/PS linkages.

Modified Oligonucleotides Targeting TTR Comprising a GalNAc 3 Conjugate Group

The oligonucleotides shown in Table 108 comprising a GalNAc conjugate were designed to target TTR.

The legend for Table 108 can be found in Example 74. The structure of GalNAc 3 -1 was shown in Example 9. The structure of GalNAc 3 -3 a was shown in Example 39. The structure of GalNAc 3 -7 a was shown in Example 48. The structure of GalNAc 3 -10 a was shown in Example 46. The structure of GalNAc 3 -13 a was shown in Example 62. The structure of GalNAc 3 -19 a was shown in Example 70.

Evaluation of Pro-inflammatory Effects of Oligonucleotides Comprising a GalNAc Conjugate in hPMBC Assay

The oligonucleotides listed in Table 109 and were tested for pro-inflammatory effects in an hPMBC assay as described in Examples 23 and 24. (See Tables 30, 83, 95, and 108 for descriptions of the oligonucleotides.) ISIS 353512 is a high responder used as a positive control, and the other oligonucleotides are described in Tables 83, 95, and 108. The results shown in Table 109 were obtained using blood from one volunteer donor. The results show that the oligonucleotides comprising mixed PO/PS internucleoside linkages produced significantly lower pro-inflammatory responses compared to the same oligonucleotides having full PS linkages. Furthermore, the GalNAc conjugate group did not have a significant effect in this assay.

Binding Affinities of Oligonucleotides Comprising a GalNAc Conjugate for the Asialoglycoprotein Receptor

The binding affinities of the oligonucleotides listed in Table 110 (see Table 76 for descriptions of the oligonucleotides) for the asialoglycoprotein receptor were tested in a competitive receptor binding assay. The competitor ligand, α1-acid glycoprotein (AGP), was incubated in 50 mM sodium acetate buffer (pH 5) with 1 U neuraminidase-agarose for 16 hours at 37° C., and >90% desialylation was confirmed by either sialic acid assay or size exclusion chromatography (SEC). Iodine monochloride was used to iodinate the AGP according to the procedure by Atsma et al. (see J Lipid Res. 1991 January; 32(1):173-81.) In this method, desialylated α1-acid glycoprotein (de-AGP) was added to 10 mM iodine chloride, Na 125 I, and 1 M glycine in 0.25 M NaOH. After incubation for 10 minutes at room temperature, 125 I -labeled de-AGP was separated from free 125 I by concentrating the mixture twice utilizing a 3 KDMWCO spin column. The protein was tested for labeling efficiency and purity on a HPLC system equipped with an Agilent SEC-3 column (7.8×300 mm) and a β-RAM counter. Competition experiments utilizing 125 I -labeled de-AGP and various GalNAc-cluster containing ASOs were performed as follows. Human HepG2 cells (10 6 cells/ml) were plated on 6-well plates in 2 ml of appropriate growth media. MEM media supplemented with 10% fetal bovine serum (FBS), 2 mM L-Glutamine and 10 mM HEPES was used. Cells were incubated 16-20 hours @ 37° C. with 5% and 10% CO 2 respectively. Cells were washed with media without FBS prior to the experiment. Cells were incubated for 30 min @37° C. with 1 ml competition mix containing appropriate growth media with 2% FBS, 10 −8 M 125 I-labeled de-AGP and GalNAc-cluster containing ASOs at concentrations ranging from 10 −11 to 10 −5 M. Non-specific binding was determined in the presence of 10 −2 M GalNAc sugar. Cells were washed twice with media without FBS to remove unbound 125 I -labeled de-AGP and competitor GalNAc ASO. Cells were lysed using Qiagen's RLT buffer containing 1% β-mercaptoethanol. Lysates were transferred to round bottom assay tubes after a brief 10 min freeze/thaw cycle and assayed on a γ-counter. Non-specific binding was subtracted before dividing 125 I protein counts by the value of the lowest GalNAc-ASO concentration counts. The inhibition curves were fitted according to a single site competition binding equation using a nonlinear regression algorithm to calculate the binding affinities (K D 's).

The results in Table 110 were obtained from experiments performed on five different days. Results for oligonucleotides marked with superscript “a” are the average of experiments run on two different days. The results show that the oligonucleotides comprising a GalNAc conjugate group on the 5′-end bound the asialoglycoprotein receptor on human HepG2 cells with 1.5 to 16-fold greater affinity than the oligonucleotides comprising a GalNAc conjugate group on the 3′-end.

Antisense Inhibition In Vivo by Oligonucleotides Comprising a GalNAc Conjugate Group Targeting Apo(a) In Vivo

The oligonucleotides listed in Table 111a below were tested in a single dose study for duration of action in mice.

The structure of GalNAc 3 -7 a was shown in Example 48.

Treatment

Female transgenic mice that express human Apo(a) were each injected subcutaneously once per week, for a total of 6 doses, with an oligonucleotide and dosage listed in Table 111b or with PBS. Each treatment group consisted of 3 animals. Blood was drawn the day before dosing to determine baseline levels of Apo(a) protein in plasma and at 72 hours, 1 week, and 2 weeks following the first dose. Additional blood draws will occur at 3 weeks, 4 weeks, 5 weeks, and 6 weeks following the first dose. Plasma Apo(a) protein levels were measured using an ELISA. The results in Table 111b are presented as the average percent of plasma Apo(a) protein levels for each treatment group, normalized to baseline levels (% BL), The results show that the oligonucleotides comprising a GalNAc conjugate group exhibited potent reduction in Apo(a) expression. This potent effect was observed for the oligonucleotide that comprises full PS internucleoside linkages and the oligonucleotide that comprises mixed PO and PS linkages.

Antisense Inhibition by Oligonucleotides Comprising a GalNAc Cluster Linked Via a Stable Moiety

The oligonucleotides listed in Table 112 were tested for inhibition of mouse APOC-III expression in vivo. C57Bl/6 mice were each injected subcutaneously once with an oligonucleotide listed in Table 112 or with PBS. Each treatment group consisted of 4 animals. Each mouse treated with ISIS 440670 received a dose of 2, 6, 20, or 60 mg/kg. Each mouse treated with ISIS 680772 or 696847 received 0.6, 2, 6, or 20 mg/kg. The GalNAc conjugate group of ISIS 696847 is linked via a stable moiety, a phosphorothioate linkage instead of a readily cleavable phosphodiester containing linkage. The animals were sacrificed 72 hours after the dose. Liver APOC-III mRNA levels were measured using real-time PCR. APOC-III mRNA levels were normalized to cyclophilin mRNA levels according to standard protocols. The results are presented in Table 112 as the average percent of APOC-III mRNA levels for each treatment group relative to the saline control group. The results show that the oligonucleotides comprising a GalNAc conjugate group were significantly more potent than the oligonucleotide lacking a conjugate group. Furthermore, the oligonucleotide comprising a GalNAc conjugate group linked to the oligonucleotide via a cleavable moiety (ISIS 680772) was even more potent than the oligonucleotide comprising a GalNAc conjugate group linked to the oligonucleotide via a stable moiety (ISIS 696847).

The structure of GalNAc 3 -7 a was shown in Example 48.

Distribution in Liver of Antisense Oligonucleotides Comprising a GalNAc Conjugate

The liver distribution of ISIS 353382 (see Table 36) that does not comprise a GalNAc conjugate and ISIS 655861 (see Table 36) that does comprise a GalNAc conjugate was evaluated. Male balb/c mice were subcutaneously injected once with ISIS 353382 or 655861 at a dosage listed in Table 113. Each treatment group consisted of 3 animals except for the 18 mg/kg group for ISIS 655861, which consisted of 2 animals. The animals were sacrificed 48 hours following the dose to determine the liver distribution of the oligonucleotides. In order to measure the number of antisense oligonucleotide molecules per cell, a Ruthenium (II) tris-bipyridine tag (MSD TAG, Meso Scale Discovery) was conjugated to an oligonucleotide probe used to detect the antisense oligonucleotides. The results presented in Table 113 are the average concentrations of oligonucleotide for each treatment group in units of millions of oligonucleotide molecules per cell. The results show that at equivalent doses, the oligonucleotide comprising a GalNAc conjugate was present at higher concentrations in the total liver and in hepatocytes than the oligonucleotide that does not comprise a GalNAc conjugate. Furthermore, the oligonucleotide comprising a GalNAc conjugate was present at lower concentrations in non-parenchymal liver cells than the oligonucleotide that does not comprise a GalNAc conjugate. And while the concentrations of ISIS 655861 in hepatocytes and non-parenchymal liver cells were similar per cell, the liver is approximately 80% hepatocytes by volume. Thus, the majority of the ISIS 655861 oligonucleotide that was present in the liver was found in hepatocytes, whereas the majority of the ISIS 353382 oligonucleotide that was present in the liver was found in non-parenchymal liver cells.

Duration of Action In Vivo of Oligonucleotides Targeting APOC-III Comprising a GalNAc 3 Conjugate

The oligonucleotides listed in Table 114 below were tested in a single dose study for duration of action in mice.

The structure of GalNAc 3 -3 a was shown in Example 39, and GalNAc 3 -19 a was shown in Example 70.

Treatment

Female transgenic mice that express human APOC-III were each injected subcutaneously once with an oligonucleotide listed in Table 114 or with PBS. Each treatment group consisted of 3 animals. Blood was drawn before dosing to determine baseline and at 3, 7, 14, 21, 28, 35, and 42 days following the dose. Plasma triglyceride and APOC-III protein levels were measured as described in Example 20. The results in Table 115 are presented as the average percent of plasma triglyceride and APOC-III levels for each treatment group, normalized to baseline levels. A comparison of the results in Table 71 of example 79 with the results in Table 115 below show that oligonucleotides comprising a mixture of phosphodiester and phosphorothioate internucleoside linkages exhibited increased duration of action than equivalent oligonucleotides comprising only phosphorothioate internucleoside linkages.

Synthesis of Oligonucleotides Comprising a 5′-GalNAc 2 Conjugate

Compound 120 is commercially available, and the synthesis of compound 126 is described in Example 49. Compound 120 (1 g, 2.89 mmol), HBTU (0.39 g, 2.89 mmol), and HOBt (1.64 g, 4.33 mmol) were dissolved in DMF (10 mL. and N,N-diisopropylethylamine (1.75 mL, 10.1 mmol) were added. After about 5 min, aminohexanoic acid benzyl ester (1.36 g, 3.46 mmol) was added to the reaction. After 3h, the reaction mixture was poured into 100 mL of 1 M NaHSO4 and extracted with 2×50 mL ethyl acetate. Organic layers were combined and washed with 3×40 mL sat NaHCO 3 and 2× brine, dried with Na 2 SO 4 , filtered and concentrated. The product was purified by silica gel column chromatography (DCM:EA:Hex, 1:1:1) to yield compound 231. LCMS and NMR were consistent with the structure. Compounds 231 (1.34 g, 2.438 mmol) was dissolved in dichloromethane (10 mL) and trifluoracetic acid (10 mL) was added. After stirring at room temperature for 2h, the reaction mixture was concentrated under reduced pressure and co-evaporated with toluene (3×10 mL). The residue was dried under reduced pressure to yield compound 232 as the trifluoracetate salt. The synthesis of compound 166 is described in Example 54. Compound 166 (3.39 g, 5.40 mmol) was dissolved in DMF (3 mL). A solution of compound 232 (1.3 g, 2.25 mmol) was dissolved in DMF (3 mL) and N,N-diisopropylethylamine (1.55 mL) was added. The reaction was stirred at room temperature for 30 minutes, then poured into water (80 mL) and the aqueous layer was extracted with EtOAc (2×100 mL). The organic phase was separated and washed with sat. aqueous NaHCO 3 (3×80 mL), 1 M NaHSO 4 (3×80 mL) and brine (2×80 mL), then dried (Na 2 SO 4 ), filtered, and concentrated. The residue was purified by silica gel column chromatography to yield compound 233. LCMS and NMR were consistent with the structure. Compound 233 (0.59 g, 0.48 mmol) was dissolved in methanol (2.2 mL) and ethyl acetate (2.2 mL). Palladium on carbon (10 wt % Pd/C, wet, 0.07 g) was added, and the reaction mixture was stirred under hydrogen atmosphere for 3 h. The reaction mixture was filtered through a pad of Celite and concentrated to yield the carboxylic acid. The carboxylic acid (1.32 g, 1.15 mmol, cluster free acid) was dissolved in DMF (3.2 mL). To this N,N-diisopropylethylamine (0.3 mL, 1.73 mmol) and PFPTFA (0.30 mL, 1.73 mmol) were added. After 30 min stirring at room temperature the reaction mixture was poured into water (40 mL) and extracted with EtOAc (2×50 mL). A standard work-up was completed as described above to yield compound 234. LCMS and NMR were consistent with the structure. Oligonucleotide 235 was prepared using the general procedure described in Example 46. The GalNAc 2 cluster portion (GalNAc 2 -24 a ) of the conjugate group GalNAc 2 -24 can be combined with any cleavable moiety present on the oligonucleotide to provide a variety of conjugate groups. The structure of GalNAc 2 -24 (GalNAc 2 -24 a -CM) is shown below:

Synthesis of Oligonucleotides Comprising a GalNAc 1 -25 Conjugate

The synthesis of compound 166 is described in Example 54. Oligonucleotide 236 was prepared using the general procedure described in Example 46.

Alternatively, oligonucleotide 236 was synthesized using the scheme shown below, and compound 238 was used to form the oligonucleotide 236 using procedures described in Example 10.

The GalNAc 1 cluster portion (GalNAc 1 -25 a ) of the conjugate group GalNAc 1 -25 can be combined with any cleavable moiety present on the oligonucleotide to provide a variety of conjugate groups. The structure of GalNAc 1 -25 (GalNAc 1 -25 a -CM) is shown below:

Antisense Inhibition In Vivo by Oligonucleotides Targeting SRB-1 Comprising a 5′-GalNAc 2 or a 5′-GalNAc 3 Conjugate

Oligonucleotides listed in Tables 116 and 117 were tested in dose-dependent studies for antisense inhibition of SRB-1 in mice.

Treatment

Six to week old, male C57BL/6 mice (Jackson Laboratory, Bar Harbor, Me.) were injected subcutaneously once with 2, 7, or 20 mg/kg of ISIS No. 440762; or with 0.2, 0.6, 2, 6, or 20 mg/kg of ISIS No. 686221, 686222, or 708561; or with saline. Each treatment group consisted of 4 animals. The mice were sacrificed 72 hours following the final administration. Liver SRB-1 mRNA levels were measured using real-time PCR. SRB-1 mRNA levels were normalized to cyclophilin mRNA levels according to standard protocols. The antisense oligonucleotides lowered SRB-1 mRNA levels in a dose-dependent manner, and the ED 50 results are presented in Tables 116 and 117. Although previous studies showed that trivalent GalNAc-conjugated oligonucleotides were significantly more potent than divalent GalNAc-conjugated oligonucleotides, which were in turn significantly more potent than monovalent GalNAc conjugated oligonucleotides (see, e.g., Khorev et al., Bioorg . & Med. Chem ., Vol. 16, 5216-5231 (2008)), treatment with antisense oligonucleotides comprising monovalent, divalent, and trivalent GalNAc clusters lowered SRB-1 mRNA levels with similar potencies as shown in Tables 116 and 117.

See Example 93 for table legend. The structure of GalNAc 3 -13a was shown in Example 62, and the structure of GalNAc 2 -24a was shown in Example 104.

See Example 93

for table legend. The structure of GalNAc 1 -25a was shown in Example 105.

The concentrations of the oligonucleotides in Tables 116 and 117 in liver were also assessed, using procedures described in Example 75. The results shown in Tables 117a and 117b below are the average total antisense oligonucleotide tissues levels for each treatment group, as measured by UV in units of μg oligonucleotide per gram of liver tissue. The results show that the oligonucleotides comprising a GalNAc conjugate group accumulated in the liver at significantly higher levels than the same dose of the oligonucleotide lacking a GalNAc conjugate group. Furthermore, the antisense oligonucleotides comprising one, two, or three GalNAc ligands in their respective conjugate groups all accumulated in the liver at similar levels. This result is surprising in view of the Khorev et al. literature reference cited above and is consistent with the activity data shown in Tables 116 and 117 above.

Synthesis of Oligonucleotides Comprising a GalNAc 1 -26 or GalNAc 1 -27 Conjugate

Oligonucleotide 239 is synthesized via coupling of compound 47 (see Example 15) to acid 64 (see Example 32) using HBTU and DIEA in DMF. The resulting amide containing compound is phosphitylated, then added to the 5′-end of an oligonucleotide using procedures described in Example 10. The GalNAc 1 cluster portion (GalNAc 1 -26 a ) of the conjugate group GalNAc 1 -26 can be combined with any cleavable moiety present on the oligonucleotide to provide a variety of conjugate groups. The structure of GalNAc 1 -26 (GalNAc 1 -26 a -CM) is shown below:

In order to add the GalNAc 1 conjugate group to the 3′-end of an oligonucleotide, the amide formed from the reaction of compounds 47 and 64 is added to a solid support using procedures described in Example 7. The oligonucleotide synthesis is then completed using procedures described in Example 9 in order to form oligonucleotide 240.

The GalNAc 1 cluster portion (GalNAc 1 -27 a ) of the conjugate group GalNAc 1 -27 can be combined with any cleavable moiety present on the oligonucleotide to provide a variety of conjugate groups. The structure of GalNAc 1 -27 (GalNAc 1 -27 a -CM) is shown below:

Antisense Inhibition In Vivo by Oligonucleotides Comprising a GalNAc Conjugate Group Targeting Apo(a) In Vivo

The oligonucleotides listed in Table 118 below were tested in a single dose study in mice.

The structure of GalNAc 3 -7 a was shown in Example 48.

Treatment

Male transgenic mice that express human Apo(a) were each injected subcutaneously once with an oligonucleotide and dosage listed in Table 119 or with PBS. Each treatment group consisted of 4 animals. Blood was drawn the day before dosing to determine baseline levels of Apo(a) protein in plasma and at 1 week following the first dose. Additional blood draws will occur weekly for approximately 8 weeks. Plasma Apo(a) protein levels were measured using an ELISA. The results in Table 119 are presented as the average percent of plasma Apo(a) protein levels for each treatment group, normalized to baseline levels (% BL), The results show that the antisense oligonucleotides reduced Apo(a) protein expression. Furthermore, the oligonucleotides comprising a GalNAc conjugate group exhibited even more potent reduction in Apo(a) expression than the oligonucleotide that does not comprise a conjugate group.

Synthesis of Oligonucleotides Comprising a GalNAc 1 -28 or GalNAc 1 -29 Conjugate

Oligonucleotide 241 is synthesized using procedures similar to those described in Example 71 to form the phosphoramidite intermediate, followed by procedures described in Example 10 to synthesize the oligonucleotide. The GalNAc 1 cluster portion (GalNAc 1 -28 a ) of the conjugate group GalNAc 1 -28 can be combined with any cleavable moiety present on the oligonucleotide to provide a variety of conjugate groups. The structure of GalNAc 1 -28 (GalNAc 1 -28 a -CM) is shown below:

In order to add the GalNAc 1 conjugate group to the 3′-end of an oligonucleotide, procedures similar to those described in Example 71 are used to form the hydroxyl intermediate, which is then added to the solid support using procedures described in Example 7. The oligonucleotide synthesis is then completed using procedures described in Example 9 in order to form oligonucleotide 242.

The GalNAc 1 cluster portion (GalNAc 1 -29 a ) of the conjugate group GalNAc 1 -29 can be combined with any cleavable moiety present on the oligonucleotide to provide a variety of conjugate groups. The structure of GalNAc 1 -29 (GalNAc 1 -29 a -CM) is shown below:

Synthesis of Oligonucleotides Comprising a GalNAc 1 -30 Conjugate

Oligonucleotide 246 comprising a GalNAc 1 -30 conjugate group, wherein Y is selected from O, S, a substituted or unsubstituted C 1 -C 10 alkyl, amino, substituted amino, azido, alkenyl or alkynyl, is synthesized as shown above. The GalNAc 1 cluster portion (GalNAc 1 -30 a ) of the conjugate group GalNAc 1 -30 can be combined with any cleavable moiety to provide a variety of conjugate groups. In certain embodiments, Y is part of the cleavable moiety. In certain embodiments, Y is part of a stable moiety, and the cleavable moiety is present on the oligonucleotide. The structure of GalNAc 1 -30 a is shown below:

Synthesis of Oligonucleotides Comprising a GalNAc 2 -31 or GalNAc 2 -32 Conjugate

Oligonucleotide 250 comprising a GalNAc 2 -31 conjugate group, wherein Y is selected from O, S, a substituted or unsubstituted C 1 -C 10 alkyl, amino, substituted amino, azido, alkenyl or alkynyl, is synthesized as shown above. The GalNAc 2 cluster portion (GalNAc 2 -31 a ) of the conjugate group GalNAc 2 -31 can be combined with any cleavable moiety to provide a variety of conjugate groups. In certain embodiments, the Y-containing group directly adjacent to the 5′-end of the oligonucleotide is part of the cleavable moiety. In certain embodiments, the Y-containing group directly adjacent to the 5′-end of the oligonucleotide is part of a stable moiety, and the cleavable moiety is present on the oligonucleotide. The structure of GalNAc 2 -31 a is shown below:

The synthesis of an oligonucleotide comprising a GalNAc 2 -32 conjugate is shown below.

Oligonucleotide 252 comprising a GalNAc 2 -32 conjugate group, wherein Y is selected from O, S, a substituted or unsubstituted C 1 -C 10 alkyl, amino, substituted amino, azido, alkenyl or alkynyl, is synthesized as shown above. The GalNAc 2 cluster portion (GalNAc 2 -32 a ) of the conjugate group GalNAc 2 -32 can be combined with any cleavable moiety to provide a variety of conjugate groups. In certain embodiments, the Y-containing group directly adjacent to the 5′-end of the oligonucleotide is part of the cleavable moiety. In certain embodiments, the Y-containing group directly adjacent to the 5′-end of the oligonucleotide is part of a stable moiety, and the cleavable moiety is present on the oligonucleotide. The structure of GalNAc 2 -32 a is shown below:

Modified Oligonucleotides Comprising a GalNAc 1 Conjugate

The oligonucleotides in Table 120 targeting SRB-1 were synthesized with a GalNAc 1 conjugate group in order to further test the potency of oligonucleotides comprising conjugate groups that contain one GalNAc ligand.

Antisense Oligonucleotides Targeting Kallikrein B, Plasma (Fletcher Factor) 1 Comprising a GalNAc Cluster

The oligonucleotides in Table 121 were designed to target human kallikrein B, plasma (Fletcher factor) 1, or prekallikrein (PKK).

Antisense Inhibition of Human PKK in HepaRG™ T Cells by Antisense Oligonucleotides with 2′-MOE Sugar Modifications

Antisense oligonucleotides were designed targeting a PKK nucleic acid and were tested for their effects on PKK mRNA in vitro. HepaRG™ cells, which are terminally differentiated hepatic cells derived from a human hepatic progenitor cell line and retain many characteristics of primary human hepatocytes (Lubberstedt M. et al., J. Pharmacol. Toxicol. Methods 2011 63: 59-68), were used in the screen.

The chimeric antisense oligonucleotides in the tables below were designed as 5-10-5 MOE gapmers. The gapmers are 20 nucleosides in length, wherein the central gap segment comprises of ten 2′-deoxynucleosides and is flanked by wing segments on the 5′ direction and the 3′ direction comprising five nucleosides each. Each nucleoside in the 5′ wing segment and each nucleoside in the 3′ wing segment has a 2′-O-methoxyethyl modification. The internucleoside linkages throughout each gapmer are phosphorothioate linkages. All cytosine residues throughout each gapmer are 5-methylcytosines. “Start site” indicates the 5′-most nucleoside to which the gapmer is targeted in the human gene sequence. “Stop site” indicates the 3′-most nucleoside to which the gapmer is targeted in the human gene sequence. Each gapmer listed in the tables below is targeted to either the human PKK mRNA, designated herein as SEQ ID NO: 1 (GENBANK Accession No. NM_000892.3) or the human PKK genomic sequence, designated herein as SEQ ID NO: 10 (GENBANK Accession No. NT_016354.19 truncated from nucleotides 111693001 to 11730000). ‘n/a’ indicates that the antisense oligonucleotide does not target that particular gene sequence.

Cultured HepaRG™ cells at a density of 20,000 cells per well were transfected using electroporation with 3,000 nM antisense oligonucleotide. After a treatment period of approximately 24 hours, RNA was isolated from the cells and PKK mRNA levels were measured by quantitative real-time PCR. Human primer probe set RTS3454 (forward sequence CCAAAAAAGGTGCACCAGTAACA, designated herein as SEQ ID NO: 20; reverse sequence CCTCCGGGACTGTACTTTAATAGG, designated herein as SEQ ID NO: 21; probe sequence CACGCAAACATTTCACAAGGCAGAGTACC, designated herein as SEQ ID NO: 22) was used to measure mRNA levels. PKK mRNA levels were adjusted according to total RNA content, as measured by RIBOGREEN®. The antisense oligonucleotides were tested in a series of experiments that had similar culture conditions. The results for each experiment are presented in separate tables shown below. Results are presented as percent inhibition of PKK, relative to untreated control cells.

Antisense Inhibition of Human PKK in HepaRG™ Cells by Antisense Oligonucleotides with 2′-MOE Sugar Modifications

Additional antisense oligonucleotides were designed targeting a PKK nucleic acid and were tested for their effects on PKK mRNA in vitro.

The chimeric antisense oligonucleotides in the tables below were designed as 5-10-5 MOE gapmers, 4-9-4 MOE gapmers, 4-10-4 MOE gapmers, 4-10-3 MOE gapmers, 3-10-4 MOE gapmers, or 3-10-3 MOE gapmers. The 5-10-5 MOE gapmers are 20 nucleosides in length, wherein the central gap segment comprises of ten 2′-deoxynucleosides and is flanked by wing segments on the 5′ direction and the 3′ direction comprising five nucleosides each. The 4-9-4 MOE gapmers are 17 nucleosides in length, wherein the central gap segment comprises of nine 2′-deoxynucleosides and is flanked by wing segments on the 5′ direction and the 3′ direction comprising four nucleosides each. The 4-10-4 MOE gapmers are 18 nucleosides in length, wherein the central gap segment comprises of ten 2′-deoxynucleosides and is flanked by wing segments on the 5′ direction and the 3′ direction comprising four nucleosides each. The 4-10-3 MOE gapmers are 17 nucleosides in length, wherein the central gap segment comprises of ten 2′-deoxynucleosides and is flanked by wing segments on the 5′ direction and the 3′ direction comprising four and three nucleosides respectively. The 3-10-4 MOE gapmers are 17 nucleosides in length, wherein the central gap segment comprises of ten 2′-deoxynucleosides and is flanked by wing segments on the 5′ direction and the 3′ direction comprising three and four nucleosides respectively. The 3-10-3 MOE gapmers are 16 nucleosides in length, wherein the central gap segment comprises of ten 2′-deoxynucleosides and is flanked by wing segments on the 5′ direction and the 3′ direction comprising three nucleosides each. Each nucleoside in the 5′ wing segment and each nucleoside in the 3′ wing segment has a 2′-O-methoxyethyl modification. The internucleoside linkages throughout each gapmer are phosphorothioate linkages. All cytosine residues throughout each gapmer are 5-methylcytosines. “Start site” indicates the 5′-most nucleoside to which the gapmer is targeted in the human gene sequence. “Stop site” indicates the 3′-most nucleoside to which the gapmer is targeted in the human gene sequence. Each gapmer listed in the tables below is targeted to either SEQ ID NO: 1 or SEQ ID NO: 10. ‘n/a’ indicates that the antisense oligonucleotide does not target that particular gene sequence.

Cultured HepaRG™ cells at a density of 20,000 cells per well were transfected using electroporation with 5,000 nM antisense oligonucleotide. After a treatment period of approximately 24 hours, RNA was isolated from the cells and PKK mRNA levels were measured by quantitative real-time PCR. Human primer probe set RTS3454 was used to measure mRNA levels. PKK mRNA levels were adjusted according to total RNA content, as measured by RIBOGREEN®. The antisense oligonucleotides were tested in a series of experiments that had similar culture conditions. The results for each experiment are presented in separate tables shown below. Results are presented as percent inhibition of PKK, relative to untreated control cells.

Antisense Inhibition of Human PKK in HepaRG™ Cells by Antisense Oligonucleotides with MOE, Deoxy and cEt Sugar Modifications

Additional antisense oligonucleotides were designed targeting a PKK nucleic acid and were tested for their effects on PKK mRNA in vitro.

The chimeric antisense oligonucleotides in the tables below were designed as deoxy, MOE and cEt gapmers. The gapmers are 16 nucleosides in length wherein the nucleoside have either a MOE sugar modification, a cEt sugar modification, or a deoxy modification. The ‘Chemistry’ column describes the sugar modifications of each oligonucleotide. ‘k’ indicates an cEt sugar modification; the number indicates the number of deoxynucleosides; otherwise, ‘d’ indicates a deoxynucleoside; and ‘e’ indicates a 2′-O-methoxyethyl modification. The internucleoside linkages throughout each gapmer are phosphorothioate linkages. All cytosine residues throughout each oligonucleotide are 5-methylcytosines. “Start site” indicates the 5′-most nucleoside to which the gapmer is targeted in the human gene sequence. “Stop site” indicates the 3′-most nucleoside to which the gapmer is targeted in the human gene sequence. Each gapmer listed in the tables below is targeted to either the human PKK mRNA, designated herein as SEQ ID NO: 1 or the human PKK genomic sequence, designated herein as SEQ ID NO: 10. ‘n/a’ indicates that the antisense oligonucleotide does not target that particular gene sequence.

Cultured HepaRG™ cells at a density of 20,000 cells per well were transfected using electroporation with 1,000 nM antisense oligonucleotide. After a treatment period of approximately 24 hours, RNA was isolated from the cells and PKK mRNA levels were measured by quantitative real-time PCR. Human primer probe set RTS3454 was used to measure mRNA levels. ISIS 531231 was also included in this assay. PKK mRNA levels were adjusted according to total RNA content, as measured by RIBOGREEN®. The antisense oligonucleotides were tested in a series of experiments that had similar culture conditions. The results for each experiment are presented in separate tables shown below. Results are presented as percent inhibition of PKK, relative to untreated control cells.

Dose-dependent Antisense Inhibition of Human PKK in HepaRG™ Cells

Gapmers from the studies described above exhibiting significant in vitro inhibition of PKK mRNA were selected and tested at various doses in HepaRG™ cells. Cells were plated at a density of 20,000 cells per well and transfected using electroporation with 0.12 μM, 0.37 μM, 1.11 μM, 3.33 μM, and 10.00 μM concentrations of antisense oligonucleotide. After a treatment period of approximately 16 hours, RNA was isolated from the cells and PKK mRNA levels were measured by quantitative real-time PCR. Human PKK primer probe set RTS3454 was used to measure mRNA levels. PKK mRNA levels were adjusted according to total RNA content, as measured by RIBOGREEN®. The antisense oligonucleotides were tested in a series of experiments that had similar culture conditions. The results for each experiment are presented in separate tables shown below. Results are presented as percent inhibition of PKK, relative to untreated control cells.

The half maximal inhibitory concentration (IC 50 ) of each oligonucleotide is also presented. PKK mRNA levels were significantly reduced in a dose-dependent manner in antisense oligonucleotide treated cells.

Dose-dependent Antisense Inhibition of Human PKK in HepaRG™ Cells

Gapmers from the studies described above exhibiting significant in vitro inhibition of PKK mRNA were selected and tested at various doses in HepaRG™ cells. Cells were plated at a density of 20,000 cells per well and transfected using electroporation with 0.19 μM, 0.56 μM, 1.67 μM, and 5.00 μM concentrations of antisense oligonucleotide. After a treatment period of approximately 16 hours, RNA was isolated from the cells and PKK mRNA levels were measured by quantitative real-time PCR. Human PKK primer probe set RTS3454 was used to measure mRNA levels. PKK mRNA levels were adjusted according to total RNA content, as measured by RIBOGREEN®. The antisense oligonucleotides were tested in a series of experiments that had similar culture conditions. The results for each experiment are presented in separate tables shown below. Results are presented as percent inhibition of PKK, relative to untreated control cells. ‘n/a’ indicates that there was no measurement done for that particular antisense oligonucleotide for that particular dose.

The half maximal inhibitory concentration (IC 50 ) of each oligonucleotide is also presented. PKK mRNA levels were significantly reduced in a dose-dependent manner in antisense oligonucleotide treated cells.

Dose-dependent Antisense Inhibition of Human PKK in HepaRG™ Cells

Gapmers from the studies described above exhibiting significant in vitro inhibition of PKK mRNA were selected and tested at various doses in HepaRG™ cells. Cells were plated at a density of 20,000 cells per well and transfected using electroporation with 0.11 μM, 0.33 μM, 1.00 μM, and 3.00 μM concentrations of antisense oligonucleotide. After a treatment period of approximately 16 hours, RNA was isolated from the cells and PKK mRNA levels were measured by quantitative real-time PCR. Human PKK primer probe set RTS3454 was used to measure mRNA levels. PKK mRNA levels were adjusted according to total RNA content, as measured by RIBOGREEN®. The antisense oligonucleotides were tested in a series of experiments that had similar culture conditions. The results for each experiment are presented in separate tables shown below. Results are presented as percent inhibition of PKK, relative to untreated control cells. ‘n/a’ indicates that there was no measurement done for that particular antisense oligonucleotide for that particular dose.

The half maximal inhibitory concentration (IC 50 ) of each oligonucleotide is also presented. PKK mRNA levels were significantly reduced in a dose-dependent manner in antisense oligonucleotide treated cells.

Efficacy of Antisense Oligonucleotides Targeting Human PKK in Transgenic Mice

Transgenic mice containing a 37,390 base pair fragment of the human KLKB1 gene sequence (chromosome 4: position 187148672-187179625, accession no: NC_000004.11) were treated with ISIS antisense oligonucleotides selected from studies described above, which were evaluated for efficacy in this model.

Treatment

Groups of transgenic mice were injected subcutaneously twice a week for 3 weeks with 2.5 mg/kg/week, 5.0 mg/kg/week, 10 mg/kg/week or 20 mg/kg/week of ISIS 546232, ISIS 546251, ISIS 546254, ISIS 546343, ISIS 546828, ISIS 547455, ISIS 547457, ISIS 547927, and ISIS 548048. One group of transgenic mice was injected subcutaneously twice a week for 3 weeks with PBS. Mice were euthanized 48 hours after the last dose, and organs and plasma were harvested for further analysis.

RNA Analysis

To evaluate the effect of ISIS oligonucleotides on target reduction, RNA was extracted from liver tissue for real-time PCR analysis of human PKK. Results are presented as percent inhibition of PKK mRNA, relative to PBS control. As shown in Table 169, treatment with ISIS antisense oligonucleotides resulted in significant reduction of PKK mRNA in comparison to the PBS control.

Protein Analysis

Plasma PKK protein levels were evaluated in all groups. Results are presented as percent inhibition of PKK protein, relative to PBS control. As shown in Table 170, treatment with ISIS antisense oligonucleotides resulted in significant reduction of PKK protein levels in comparison to the PBS control.

Effect of ISIS Antisense Oligonucleotides Targeting Human PKK in Cynomolgus Monkeys

Cynomolgus monkeys were treated with ISIS antisense oligonucleotides selected from studies described above. Antisense oligonucleotide efficacy and tolerability were evaluated. The human antisense oligonucleotides tested are cross-reactive with the rhesus genomic sequence (GENBANK Accession No. NW_001118167.1 truncated from nucleotides 2358000 to 2391000 and designated herein as SEQ ID NO: 18). The target start site of each oligonucleotide to SEQ ID NO: 18 is presented in Table 171. ‘Mismatches’ indicates that the number of nucleotides by which the oligonucleotide is mismatched to the rhesus sequence. The greater the complementarity between the human oligonucleotide and the rhesus monkey sequence, the more likely the human oligonucleotide can cross-react with the rhesus monkey sequence. ‘n/a’ indicates that the oligonucleotide is has more than 3 mismatches with the rhesus gene sequence.

Treatment

Prior to the study, the monkeys were kept in quarantine for a 30-day period, during which the animals were observed daily for general health. The monkeys were 2-4 years old and weighed between 2 and 4 kg. Ten groups of four randomly assigned male cynomolgus monkeys each were injected subcutaneously with ISIS oligonucleotide or PBS. PBS solution or ISIS oligonucleotides at a dose of 40 mg/kg were administered initially with a loading regimen consisting of four doses on the first week of the study (days 1, 3, 5, and 7), followed by a maintenance regimen consisting of once weekly administration starting on Day 14 (weeks 2 to 13). Subcutaneous injections were performed in clock-wise rotations at 4 sites on the back; one site per dose. The injection sites were delineated by tattoo, while sedated using ketamine, and were separated by a minimum of 3 cm.

During the study period, the monkeys were observed a minimum of once daily for signs of illness or distress. Any animal experiencing more than momentary or slight pain or distress due to the treatment, injury or illness was promptly reported to the responsible veterinarian and the Study Director. Any animal in poor health or in a possible moribund condition was identified for further monitoring and possible euthanasia. For example, two monkeys treated with ISIS 547445 were euthanized due to subdued behavior, lateral position, lack of response to stimuli and decreased respiration. The protocols described in the Example were approved by the Institutional Animal Care and Use Committee (IACUC).

Target Reduction

RNA Analysis

On day 87 or 88, 48 hours after the final dose, RNA was extracted from liver tissue for real-time PCR analysis of PKK using primer probe set RTS3455 (forward sequence CCTGTGTGGAGGGTCACTCA, designated herein as SEQ ID NO: 23; reverse sequence CCACTATAGATGCGCCAAACATC, designated herein as SEQ ID NO: 24; probe sequence CCCACTGCTTTGATGGGCTTCCC, designated herein as SEQ ID NO: 25). The results were normalized to the housekeeping gene, Cyclophilin. Results are presented as percent inhibition of PKK mRNA, relative to PBS control. As shown in Table 172, treatment with ISIS antisense oligonucleotides resulted in significant reduction of PKK mRNA in comparison to the PBS control.

Protein Analysis

Approximately 0.9 mL of blood was collected each time from all available animals at pre-dose, day 17, day 31, day 45, day 59, and day 73, and placed in tubes containing 3.2% sodium citrate. The tubes were centrifuged (3000 rpm for 10 min at room temperature) to obtain plasma. PKK protein levels were measured in the plasma by ELISA. The results are presented in Table 173, expressed as percentage inhibition compared to the PBS control levels. The results indicate that ISIS oligonucleotides significantly reduced PKK protein levels.

Tolerability Studies

Liver Function

To evaluate the effect of ISIS oligonucleotides on hepatic function, the monkeys were fasted overnight. Approximately, 1.5 mL of blood samples were collected from all the study groups. Blood was collected in tubes without anticoagulant for serum separation. The tubes were kept at room temperature for a minimum of 90 min and then centrifuged at 3,000 rpm for 10 min. Levels of various liver function markers were measured using a Toshiba 120FR NEO chemistry analyzer (Toshiba Co., Japan). The results are presented in Table 174 and indicate that antisense oligonucleotides had no effect on liver function outside the expected range for antisense oligonucleotides.

Hematology

To evaluate any effect of ISIS oligonucleotides in cynomolgus monkeys on hematologic parameters, blood samples of approximately 1.2 mL of blood was collected pre-dose and on day 87 or day 88 from each of the available study animals in tubes containing K 2 -EDTA. Samples were analyzed for red blood cell (RBC) count, white blood cells (WBC) count, platelet count, hemoglobin content and hematocrit, using an ADVIA2120i hematology analyzer (SIEMENS, USA). The data is presented in Table 175.

The data indicate treatment with most of the oligonucleotides did not cause any changes in hematologic parameters outside the expected range for antisense oligonucleotides at this dose.

Kidney Function

To evaluate the effect of ISIS oligonucleotides on kidney function, the monkeys were fasted overnight. Approximately, 1.5 mL of blood samples were collected from all the study groups. Blood was collected in tubes without anticoagulant for serum separation. The tubes were kept at room temperature for a minimum of 90 min and then centrifuged at 3,000 rpm for 10 min. Levels of BUN and creatinine were measured using a Toshiba 120FR NEO chemistry analyzer (Toshiba Co., Japan). Results are presented in Table 176, expressed in mg/dL. The plasma chemistry data indicate that most of the ISIS oligonucleotides did not have any effect on the kidney function outside the expected range for antisense oligonucleotides. Specifically, treatment with ISIS 546254 was well tolerated in terms of the kidney function of the monkeys.

Kidney function was also assessed by urinalysis. Fresh urine from all animals was collected using a clean cage pan on wet ice. Food was removed overnight the day before fresh urine collection was done but water was supplied. The total protein and creatinine levels were measured using a Toshiba 120FR NEO automated chemistry analyzer (Toshiba Co., Japan) and the protein to creatinine ratio was calculated. The results are presented in Table 177.

C-reactive Protein Level Analysis

To evaluate any inflammatory effect of ISIS oligonucleotides in cynomolgus monkeys, the monkeys were fasted overnight. Approximately, 1.5 mL of blood samples were collected from all the study groups. Blood was collected in tubes without anticoagulant for serum separation. The tubes were kept at room temperature for a minimum of 90 min and then centrifuged at 3,000 rpm for 10 min. C-reactive protein (CRP), which is synthesized in the liver and which serves as a marker of inflammation, was measured using a Toshiba 120FR NEO chemistry analyzer (Toshiba Co., Japan). Complement C3 was also measured similarly, and the data is presented as a percentage of baseline values. The results are presented in Table 178 and indicate that treatment with ISIS oligonucleotides did not cause any inflammation in monkeys.

Antisense Inhibition of Murine PKK mRNA in Mouse Primary Hepatocytes

Antisense oligonucleotides targeting a murine PKK nucleic acid were designed and tested for their effects on PKK mRNA in vitro. Cultured mouse primary hepatocytes at a density of 10,000 cells per well were transfected using Cytofectin reagent with 12.5 nM, 25.0 nM, 50.0 nM, 100.0 nM, and 200.0 nM of antisense oligonucleotide. After a treatment period of approximately 24 hours, RNA was isolated from the cells and mouse PKK mRNA levels were measured by quantitative real-time PCR using the murine primer probe set RTS3313 (forward sequence TGCCTGCTGTTCAGCTTTCTC, designated herein as SEQ ID NO: 2228; reverse sequence TGGCAAAGTCCCTGTAATGCT, designated herein as SEQ ID NO: 2229; probe sequence CGTGACTCCACCCAAAGAGACAAATAAACG, designated herein as SEQ ID NO: 2230). PKK mRNA levels were adjusted according to total RNA content, as measured by RIBOGREEN.

The chimeric antisense oligonucleotides were designed as 5-10-5 MOE gapmers. The gapmers are 20 nucleotides in length, wherein the central gap segment is comprised of ten 2′-deoxynucleosides and is flanked on both sides (in the 5′ and 3′ directions) by wings comprising 5 nucleosides each. Each nucleoside in the 5′ wing segment and each nucleoside in the 3′ wing segment has a 2′-O-methoxyethyl modification. The internucleoside linkages throughout each gapmer are phosphorothioate linkages. All cytosine residues throughout each gapmer are 5-methylcytosines. Results demonstrate that PKK mRNA levels were significantly reduced in a dose dependent manner.

In one specific example, ISIS 482584 (GGCATATTGGTTTTTGGAAT; SEQ ID NO: 2244) reduced PKK mRNA in a dose dependent manner yielding a half maximal inhibitory concentration (IC 50 ) of 84 nM (see Table 179). ISIS 482584 is targeted to SEQ ID NO: 11 (GENBANK Accession No. NM_008455.2) and has a target start site of 1586 and a target stop site of 1605. “Target start site” indicates the 5′-most nucleotide to which the gapmer is targeted. “Target stop site” indicates the 3′-most nucleotide to which the gapmer is targeted.

Antisense Inhibition of PKK mRNA in BALB/c Mice

ISIS 482584 was tested for its effect on murine PKK mRNA in vivo.

Treatment

Six groups of male BALB/c mice each were treated with 2.5 mg/kg, 5.0 mg/kg, 10.0 mg/kg, 20.0 mg/kg, 40.0 mg/kg, or 80.0 mg/kg of ISIS 482584, administered subcutaneously twice a week for 3 weeks (weekly doses of 5.0 mg/kg, 10.0 mg/kg, 20.0 mg/kg, 40.0 mg/kg, 80.0 mg/kg, or 160.0 mg/kg). A control group of BALB/c mice was treated with PBS, administered subcutaneously twice a week for 3 weeks. Two days after the last dose of antisense oligonucleotide or PBS, mice from all groups were anesthetized with 150 mg/kg ketamine mixed with 10 mg/kg xylazine, administered by intraperitoneal injection. Liver was collected for RNA analysis.

RNA Analysis

RNA was extracted from liver tissue for real-time PCR analysis of PKK. PKK mRNA levels were measured using the murine primer probe set (forward sequence ACAAGTGCATTTTACAGACCAGAGTAC, designated herein as SEQ ID NO: 2231; reverse sequence GGTTGTCCGCTGACTTTATGCT, designated herein as SEQ ID NO: 2232; probe sequence AAGCACAGTGCAAGCGGAACACCC, designated herein as SEQ ID NO: 2233). Results are presented as percent inhibition of PKK, relative to PBS control. As shown in Table 180, treatment with ISIS 482584 resulted in significant dose-dependent reduction of PKK mRNA in comparison to the PBS control.

Protein Analysis

Plasma was collected in tubes containing sodium citrate as an anticoagulant. The samples were run on a 4-12% gradient SDS-polyacrylamide gel (Invitrogen), followed by immunoblotting with murine PKK antibody (R&D Systems). Blots were incubated with secondary fluorophore-labeled antibodies (LI-COR) and imaged in an Odyssey Imager (LI-COR). Results are presented as percent inhibition of PKK, relative to PBS control. As shown in Table 181, treatment with ISIS 482584 resulted in significant dose-dependent reduction of PKK plasma protein in comparison to the PBS control.

In Vivo Effect of Antisense Inhibition of Murine PKK in an Angioedema Mouse Model

Hereditary angioedema (HAE) is characterized by local swelling and increase in vascular permeability in subcutaneous tissues (Morgan, B. P. N. Engl. J. Med. 363: 581-83, 2010). It is caused by a deficiency of the C1 inhibitor, a protein of the complement system. Two mouse models were used in this study including an established mouse model of C1-INH deficiency and a captopril-induced edema model, both of which cause vascular permeability, a hallmark of HAE. Reversal of vascular permeability is accompanied by increased plasma levels of high molecular weight kininogen (HMWK).

In the first model, angioedema was induced by treatment with Captopril, a known antihypertensive agent, which increases vascular permeability in mice and replicates the pathology of hereditary angioedema.

In the second model, angioedema was induced by treatment with ISIS 461756, an antisense oligonucleotide which targets murine C1 inhibitor mRNA, which increases vascular permeability in mice and replicates the pathology of hereditary angioedema. ISIS 461756 (SEQ ID NO: 2245; AAAGTGGTTGATACCCTGGG) is a 5-10-5 MOE gapmer targeting nucleosides 1730-1749 of NM_009776.3 (SEQ ID NO: 2243).

The effect of HOE-140 and ISIS 482584, an antisense oligonucleotide inhibitor of PKK, were evaluated in the Captopril and ISIS 461756-induced mouse models of vascular permeability. Some of the murine groups were treated with HOE-140, a selective antagonist of the bradykinin B2 receptor, which blocks vasodilation and vascular permeability (Cruden and Newby, Expert Opin. Pharmacol. 9: 2383-90, 2008). Other mice were treated with ISIS 482584, which inhibits PKK mRNA expression. The effect of treatment with HOE-140 was compared with the effect of treatment with ISIS 482584.

Treatment

The various treatment groups for this assay are presented in Table 182.

Group 1 consisted of 4 C57BL/6J-Tyrc-2J mice treated with PBS administered subcutaneously twice a week for 4 weeks. No other treatment was administered to Group 1 which served as a control group to measure the basal level of vascular permeability.

Group 2 consisted of 8 C57BL/6J-Tyrc-2J mice treated with PBS administered subcutaneously twice a week for 4 weeks. At the end of the treatment, the mice were intraperitoneally administered 20 μg of captopril. Group 2 served as a PBS control group for captopril-induced vascular permeability.

Group 3 consisted of 8 C57BL/6J-Tyrc-2J mice treated with PBS administered subcutaneously twice a week for 4 weeks. On day 14, the mice were treated with 50 mg/kg of the antisense oligonucleotide targeting C1 inhibitor, ISIS 461756, administered subcutaneously twice a week for 2 weeks. At the end of the treatment period, the mice were intraperitoneally administered 20 μg of captopril. Group 3 served as a PBS control group for captopril and ISIS 461756-induced vascular permeability.

Group 4 consisted of 8 C57BL/6J-Tyrc-2J mice treated with PBS administered subcutaneously twice a week for 4 weeks. On day 14, the mice were treated with 50 mg/kg of the antisense oligonucleotide targeting C1 inhibitor, ISIS 461756, administered subcutaneously twice a week for 2 weeks. At the end of the treatment period, the mice were intraperitoneally administered 20 μg of captopril. The mice were then also intraperitoneally administered 30 μg of HOE-140. Group 4 served as a positive control for inhibition of vascular permeability with HOE-140.

Group 5 consisted of 8 C57BL/6J-Tyrc-2J mice treated with 40 mg/kg of control oligonucleotide ISIS 141923, a 5-10-5 MOE gapmer with no known murine target, (CCTTCCCTGAAGGTTCCTCC; SEQ ID NO: 2246) administered subcutaneously twice a week for 4 weeks. On day 14, the mice were treated with 50 mg/kg of the antisense oligonucleotide targeting C1 inhibitor, ISIS 461756, administered subcutaneously twice a week for 2 weeks. At the end of the treatment period, the mice were intraperitoneally administered 20 μg of captopril. Group 5 served as a control group for captopril and ISIS 461756-induced vascular permeability.

Group 6 consisted of 8 C57BL/6J-Tyrc-2J mice and was treated with 40 mg/kg of ISIS 482584 administered subcutaneously twice a week for 4 weeks. At the end of the treatment period, the mice were intraperitoneally administered 20 μg of captopril. Group 6 served as the experimental treatment group for examining the effect of PKK ASO on captopril-induced vascular permeability.

Group 7 consisted of 8 C57BL/6J-Tyrc-2J mice treated with 40 mg/kg of ISIS 482584 administered subcutaneously twice a week for 4 weeks. On day 14, the mice were treated with 50 mg/kg of the antisense oligonucleotide targeting C1 inhibitor, ISIS 461756, administered subcutaneously twice a week for 2 weeks. At the end of the treatment period, the mice were intraperitoneally administered 20 μg of captopril. Group 7 served as the experimental treatment group for examining the effect of PKK ASO on captopril and ISIS 461756-induced vascular permeability.

All the groups were then injected with 30 mg/kg of Evans Blue solution into the tail vein. The mice were sacrificed 30 min after the Evans Blue solution administration and colons, feet, ears, and intestines were harvested. Blood samples were taken through cardiac puncture.

Quantification of Vascular Permeability

The harvested tissues from the feet, colon, ears, and intestines were placed separately in formamide solution overnight to leach out the Evans Blue dye. The formamide solution containing ear and feet tissue was heated to 55° C. and left overnight. The color intensity of the dye-infused formamide solution was then measured at OD 600nm , and is presented in Table 183. Mice displaying any manifestation of angioedema take up more dye and, therefore, demonstrate high OD values.

As presented in Table 183, treatment with ISIS 482584 prevents vascular permeability in mice treated with captopril (Group 6) and in mice treated with captopril and ISIS 461756 (Group 7) compared to the respective PBS control groups (Groups 2 and 3). Measures of vascular permeability in mice of Groups 6 and 7 were also reduced in most of the tissues in comparison to the mice treated with the control oligonucleotide, ISIS 141923 (Group 5), where vascular permeability was induced with captopril and ISIS 461756. Measures of vascular permeability in the colon and feet tissues of both the treatment groups (Groups 6 and 7) were comparable to basal levels, as observed in mice treated with only PBS (Group 1). Reduction in vascular permeability in mice treated with ISIS 482584 was comparable to that seen in mice treated with the bradykinin 2 receptor antagonist, HOE140, which served as a positive control in this assay.

Therefore, antisense inhibition of PKK mRNA may be beneficial for the treatment and prevention of vascular permeability, which is symptomatic of HAE.

Quantification of High Molecular Weight Kininogen (HMWK)

Western blot quantification of HMWK from blood samples shows that Groups 1 and 2 have low levels of HMWK as compared to Groups 6 and 7 indicating that vascular permeability is reversed in Groups 6 and 7. Samples from Groups 1 and 2 have increased HMWK cleavage product as compared to Groups 6 and 7. Thus, lack of HMWK is caused by PKK cleavage of HMWK into cleavage products (including bradykinin and HKa).

In Vivo Effect of Antisense Inhibition of Murine PKK on Basal Permeability and Captopril-induced Permeability in Mice

Basal permeability is the level of vascular permeability occurring in the tissues of naïve, untreated mice. The effect of ISIS 482584 in the prevention of vascular permeability, either basal or captopril-induced, was evaluated.

Treatment

The various treatment groups for this assay are presented in Table 184.

Group 1 consisted of 8 mice and was treated with PBS administered subcutaneously twice a week for 4 weeks. No other treatment was administered to Group 1 which served as a control group to measure the basal levels of vascular permeability.

Group 2 consisted of 8 mice and was treated with PBS administered subcutaneously twice a week for 4 weeks. At the end of the treatment period, the mice were intraperitoneally administered 20 μg of captopril. Group 2 served as the negative control group for captopril-induced vascular permeability.

Group 3 consisted of 8 mice and was treated with PBS administered subcutaneously twice a week for 4 weeks. At the end of the treatment period, the mice were intraperitoneally administered 30 μg of HOE-140. Group 3 served as a positive control for inhibition of basal vascular permeability.

Group 4 consisted of 8 mice and was treated with PBS administered subcutaneously twice a week for 4 weeks. At the end of the treatment period, the mice were intraperitoneally administered 20 μg of captopril. The mice were also intraperitoneally administered 30 μg of HOE-140. Group 4 served as a positive control for inhibition of captopril-induced vascular permeability.

Group 5 consisted of 8 mice and was treated with 40 mg/kg of ISIS 482584 administered subcutaneously twice a week for 4 weeks. Group 5 served as an experimental treatment group for examining the effect of ISIS 482584 on basal vascular permeability.

Group 6 consisted of 8 mice and was treated with 40 mg/kg of ISIS 482584 administered subcutaneously twice a week for 4 weeks. At the end of the treatment period, the mice were intraperitoneally administered 20 μg of captopril. Group 6 served as an experimental treatment group for examining the effect of ISIS 482584 on captopril-induced vascular permeability.

All the groups were then injected with 30 mg/kg of Evans Blue solution. The mice were sacrificed 30 min after the Evans Blue solution administration and colons, feet, ears, and intestines were harvested.

Quantification of Vascular Permeability

The harvested tissues from the feet, colon, intestine, and ears were placed separately in formamide solution overnight to leach out the Evans Blue dye. The formamide solution containing feet and ear tissue was heated to 55° C. and left overnight. The color intensity of the dye-infused formamide solution was then measured at OD 600nm , and is presented in Table 185. Mice displaying any manifestation of angioedema take up more dye and, therefore, demonstrate high OD values.

As presented in Table 185, mice treated with ISIS 482584 demonstrated reduced basal vascular permeability compared to the PBS control (Group 5 vs. Group 1). The reduction in basal vascular permeability by treatment with ISIS 482584 was comparable to that caused by treatment with HOE-140 (Group 3, which served as the positive control). Mice treated with ISIS 482584 also demonstrated reduced captopril-induced vascular permeability in most tissues compared to the PBS control (Group 6 vs. Group 2). The reduction in captopril-induced vascular permeability by treatment with ISIS 482584 was comparable to that caused by treatment with HOE-140 (Group 4, which served as the positive control).

Dose-dependent Effect of Antisense Inhibition of Murine PKK on Captopril-induced Vascular Permeability

The effect of varying doses on ISIS 482584 on captopril-induced vascular permeability was evaluated.

Treatment

The various treatment groups for this assay are presented in Table 186.

Group 1 consisted of 4 mice and was treated with PBS administered subcutaneously twice a week for 3 weeks. No other treatment was administered to Group 1 which served as a control group to measure the basal levels of vascular permeability.

Group 2 consisted of 8 mice and was treated with PBS administered subcutaneously twice a week for 3 weeks. At the end of the treatment period, the mice were intraperitoneally administered 20 μg of captopril. Group 2 served as the control group for captopril-induced vascular permeability.

Group 3 consisted of 4 mice and was treated with PBS administered subcutaneously twice a week for 3 weeks. At the end of the treatment period, the mice were intraperitoneally administered 20 μg of captopril. The mice were also intraperitoneally administered 30 μg of Icatibant (HOE-140). Group 4 served as a positive control for inhibition of captopril-induced vascular permeability.

Groups 4, 5, 6, 7, 8, and 9 consisted of 8 mice each and were treated with 2.5 mg/kg, 5 mg/kg, 10 mg/kg, 20 mg/kg, 40 mg/kg, or 80 mg/kg (corresponding to 5 mg/kg, 10 mg/kg, 20 mg/kg, 40 mg/kg, 80 mg/kg, or 160 mg/kg per week), respectively of ISIS 482584 administered subcutaneously twice a week for 3 weeks. At the end of the treatment period, the mice of all the groups were intraperitoneally administered 20 g of captopril. Groups 4-9 served as the experimental treatment groups for examining the effect of varying doses of ISIS 482584 on captopril-induced vascular permeability.

All the groups were then injected with 30 mg/kg of Evans Blue solution in the tail vein. The mice were sacrificed 30 min after the Evans Blue solution administration and colons, feet, ears, and intestines were harvested. Blood samples were taken through cardiac puncture.

Quantification of Vascular Permeability

The harvested tissues were placed in formamide solution overnight to leach out the Evans Blue dye. The formamide solution containing feet and ear tissue was heated to 55° C. and left overnight. The color intensity of the dye-infused formamide solution was then measured at OD 600nm , and is presented in Table 187. Mice displaying any manifestation of angioedema take up more dye and, therefore, demonstrate high OD values.

As presented in Table 187, mice treated with higher doses of ISIS 482584 (Groups 4, 5, and 6) had reduced levels of captopril-induced vascular permeability compared to the corresponding PBS control group (Group 2). The reduction in vascular permeability in mice of these treatment groups (Groups 4 and 5) was comparable to the levels of basal vascular permeability (as shown in Group 1) as well as in mice treated with HOE-140 (Group 3).

Quantification of Vascular Leakage

The blood drawn through cardiac puncture was immediately mixed with 3 times the volume of ice-cold ethanol. The solution was centrifuged at 15,000 g for 20 minutes at 4° C. to remove cell debris and precipitated plasma proteins. The ethanol extracts were further purified by ultra-filtration through a 10 kDa MWCO filter. The color intensity of the ethanol extracted plasma solution was then measured at OD 620nm . The results are presented in Table 188 as percentage increase or decrease of the OD values of the Group 1 PBS control. It was expected that tissues from mice displaying manifestation of angioedema would leak more dye from the plasma and, therefore, demonstrate low OD values, whereas treatment groups may display higher OD values due to reduced vascular leakage. Mice treated with 160 mg/kg/week and 80 mg/kg/week of ISIS 482584 (Groups 4 and 5) demonstrated less vascular leakage compared to the PBS negative control treated with captopril (Group 2). The results from Groups 4 and 5 were comparable to the positive control treated with HOE-140 (Group 3).

Dose-dependent Effect of Antisense Inhibition of Murine PKK on Basal Permeability in Mice

The effect of varying doses on ISIS 482584 on basal vascular permeability was evaluated.

Treatment

The various treatment groups for this assay are presented in Table 189.

Group 1 consisted of 8 mice and was treated with PBS administered subcutaneously twice a week for 3 weeks. No other treatment was administered to Group 1 which served as a control group to measure the basal levels of vascular permeability.

Group 2 consisted of 4 mice and was treated with PBS administered subcutaneously twice a week for 3 weeks. At the end of the treatment period, the mice were intraperitoneally administered 30 μg of HOE-140. Group 2 served as a positive control for inhibition of basal vascular permeability.

Groups 3, 4, 5, 6, 7, and 8 consisted of 8 mice each and were treated with 2.5 mg/kg, 5 mg/kg, 10 mg/kg, 20 mg/kg, 40 mg/kg, or 80 mg/kg (corresponding to 5 mg/kg, 10 mg/kg, 20 mg/kg, 40 mg/kg, 80 mg/kg, or 160 mg/kg per week), respectively of ISIS 482584 administered subcutaneously twice a week for 3 weeks. Groups 4-9 served as the experimental treatment groups for examining the effect of varying doses of ISIS 482584 on basal vascular permeability.

All the groups were then injected with 30 mg/kg of Evans Blue solution in the tail vein. The mice were sacrificed 30 min after the Evans Blue solution administration and colons, feet, and ears were harvested and examined for permeability defects. Blood samples were taken through cardiac puncture.

Quantification of Vascular Permeability

The harvested tissues from the feet, colon, and ears were placed in formamide solution overnight to leach out the Evans Blue dye. The formamide solution containing feet and ear tissue was heated to 55° C. and left overnight. The color intensity of the dye-infused formamide solution was then measured at OD 600nm , and is presented in Table 190. Higher OD values are associated with higher levels of permeability.

As presented in Table 190, most of the tissues of mice treated with ISIS 482584 at all doses (Groups 3-8) demonstrated reduced basal vascular permeability compared to the PBS control (Group 1). The reduction in basal vascular permeability of the ISIS oligonucleotide-treated groups was comparable to the same demonstrated in the positive control group treated with HOE-140 (Group 2).

Quantification of Vascular Leakage

The blood drawn through cardiac puncture was immediately mixed with 3 times the volume of ice-cold ethanol. The solution was centrifuged at 15,000 g for 20 minutes at 4° C. to remove cell debris and precipitated plasma proteins. The ethanol extracts were further purified by ultra-filtration through a 10 kDa MWCO filter. The color intensity of the ethanol extracted plasma solution was then measured at OD 620nm . The results are presented in Table 191 as percentage increase or decrease of the OD values of the Group 1 PBS control. It was expected that treatment groups may display higher OD values due to reduced vascular leakage. All the mice in the ISIS oligonucleotide-treated groups demonstrated significantly reduced vascular leakage compared to the PBS negative control.

Quantification of High Molecular Weight Kininogen (HMWK)

Western blot quantification of HMWK from blood samples are presented in Tables 192 and 193.

As shown in Table 192, Groups treated with 482584 have higher levels of HMWK as compared to PBS control, increasing in a dose-dependent manner. Treatment with PKK antisense oligonucleotide results in stabilization of HMWK. Thus, vascular permeability is reduced in ISIS 482584-treated groups in a dose-dependent manner. As shown in Table 193, Groups treated with ISIS 482584 have lower HMWK cleavage product as compared to PBS control, decreasing in a dose-dependent manner. Thus, reduced HMWK is caused by PKK cleavage of HMWK into cleavage products (including bradykinin and HKa). Data are presented in Intensity Units as measured by densitometer.

Combination Therapy of Antisense Oligonucleotides Targeting PKK and Factor 12 on Captopril-induced Vascular Permeability in Mice

Mice were treated varying doses of ISIS 410944, a 5-10-5 MOE gapmer targeting Factor 12 (GCATGGGACAGAGATGGTGC; SEQ ID NO: 2247), and ISIS 482584 in a captopril-induced vascular permeability model.

Treatment

The various treatment groups for this assay are presented in Table 194.

Group 1 consisted of 4 mice and was treated with PBS administered subcutaneously twice a week for 3 weeks. No other treatment was administered to Group 1 which served as a control group to measure the basal levels of vascular permeability.

Group 2 consisted of 8 mice and was treated with PBS administered subcutaneously twice a week for 3 weeks. At the end of the treatment period, the mice were intraperitoneally administered 20 μg of captopril. Group 2 served as the control group for captopril-induced vascular permeability.

Group 3 consisted of 4 mice and was treated with PBS administered subcutaneously twice a week for 3 weeks. At the end of the treatment period, the mice were intraperitoneally administered 20 μg of captopril. The mice were also intraperitoneally administered 30 μg of HOE-140. Group 3 served as a positive control for inhibition of captopril-induced vascular permeability.

Groups 4, 5, 6, 7, and 8 consisted of 8 mice each and were treated with 2.5 mg/kg, 5 mg/kg, 10 mg/kg, 20 mg/kg, or 40 mg/kg (corresponding to 5 mg/kg, 10 mg/kg, 20 mg/kg, 40 mg/kg, or 80 mg/kg per week), respectively of ISIS 482584 and ISIS 410944 each administered subcutaneously twice a week for 3 weeks. At the end of the treatment period, the mice of all the groups were intraperitoneally administered 20 μg of captopril. Groups 4-8 served as the experimental treatment groups for examining the effect of ISIS 410944 and ISIS 482584 on captopril-induced vascular permeability.

All the groups were then injected with 30 mg/kg of Evans Blue solution in the tail vein. The mice were sacrificed 30 min after the Evans Blue solution administration and colons, feet, ears, and intestines were harvested.

Quantification of Vascular Permeability

The harvested tissues from the feet, colon, and ears were placed in formamide solution overnight to leach out the Evans Blue dye. The formamide solution containing feet and ear tissue was heated to 55° C. and left overnight. The color intensity of the dye-infused formamide solution was then measured at OD 600nm , and is presented in Table 195. Higher OD values are associated with higher levels of permeability.

As presented in Table 195, most of the tissues of mice treated with a combination of ISIS 482584 and ISIS 410944 at all doses (Groups 3-8) demonstrated reduced vascular permeability compared to the PBS control (Group 1). The reduction in vascular permeability of the ISIS oligonucleotide-treated groups was comparable to the same demonstrated in the basal PBS control (Group 1), as well as the positive control group treated with HOE140 (Group 2). Combination of PKK and Factor 12 antisense oligonucleotides results in synergistic decrease in permeability. As expected, a corresponding synergistic decrease in vascular leakage was also observed.

Combination Therapy of Antisense Oligonucleotides Targeting PKK and Factor 12 on Basal Vascular Permeability in Mice

Mice were treated with varying doses of ISIS 410944, an antisense oligonucleotide targeting Factor 12, and ISIS 482584 in a basal vascular permeability model.

Treatment

The various treatment groups for this assay are presented in Table 196.

Group 1 consisted of 8 mice and was treated with PBS administered subcutaneously twice a week for 3 weeks. No other treatment was administered to Group 1 which served as a control group to measure the basal levels of vascular permeability.

Group 2 consisted of 4 mice and was treated with PBS administered subcutaneously twice a week for 3 weeks. At the end of the treatment period, the mice were intraperitoneally administered 30 μg of HOE-140. Group 2 served as a positive control for inhibition of basal vascular permeability.

Groups 3, 4, 5, 6, and 7 consisted of 8 mice each and were treated with 2.5 mg/kg, 5 mg/kg, 10 mg/kg, 20 mg/kg, or 40 mg/kg (corresponding to 5 mg/kg, 10 mg/kg, 20 mg/kg, 40 mg/kg, or 80 mg/kg per week), respectively of ISIS 482584 and ISIS 410944 each administered subcutaneously twice a week for 3 weeks. Groups 3-7 served as the experimental treatment groups for examining the effect of ISIS 410944 and ISIS 482584 on basal vascular permeability.

All the groups were then injected with 30 mg/kg of Evans Blue solution in the tail vein. The mice were sacrificed 30 min after the Evans Blue solution administration and colons, feet, ears, and intestines were harvested.

Quantification of Vascular Permeability

The harvested tissues from the feet, colon, intestines, and ears were placed in formamide solution overnight to leach out the Evans Blue dye. The formamide solution containing feet and ear tissue was heated to 55° C. and left overnight. The color intensity of the dye-infused formamide solution was then measured at OD 600nm , and is presented in Table 197. Higher OD values are associated with higher levels of permeability.

As presented in Table 197, most of the tissues of mice treated with a combination of ISIS 482584 and ISIS 410944 at all doses (Groups 2-7) demonstrated reduced vascular permeability compared to the PBS control (Group 1). The reduction in vascular permeability of the ISIS oligonucleotide-treated groups was comparable to the same demonstrated in positive control group treated with HOE140 (Group 2). Combination of PKK and Factor 12 antisense oligonucleotides results in synergistic decrease in permeability. As expected, a corresponding synergistic decrease in vascular leakage was also observed.

Inhibition of Factor 12 Protein Activation by ISIS 482584

The effect of antisense inhibition of PKK mRNA on Factor 12 protein activation was evaluated.

Treatment

The various treatment groups for this assay are presented in Table 198.

Group 1 consisted of 8 mice and was treated with PBS administered subcutaneously twice a week for 3 weeks. No other treatment was administered to Group 1 which served as a control group to measure Factor 12 activation.

Groups 2, 3, 4, 5, and 6 consisted of 8 mice each and were treated with 2.5 mg/kg, 5 mg/kg, 10 mg/kg, 20 mg/kg, or 40 mg/kg (corresponding to 5 mg/kg, 10 mg/kg, 20 mg/kg, 40 mg/kg, or 80 mg/kg per week), respectively of ISIS 482584 administered subcutaneously twice a week for 3 weeks. Groups 2-6 served as the treatment groups for measuring the effect of ISIS 482584 on Factor 12 activation.

At the end of the treatment period, plasma was harvested from the mice for the Spectrozyme® Factor 12a based amidolytic assay for Factor 12 in plasma.

Assay for Factor 12 Activation in Plasma

Plasma (5 μL) was added to 85 μL of PBS with 1 ug/ml dextran sulfate (500 kDa) in a 96 well polypropelene microplate and the solution was incubated for 5 minutes at room temperature. Spectrozyme® FXIIa (10 μL of a 2 mM solution) and 0.2 mM KALLISTOP™ solution was added and the absorbance kinetic was measured at 405 nm. Factor 12 activation was measured in the linear phase of absorbance accumulation. The results are presented in Table 199 as a percentage of Factor 12 activation measured in the PBS control sample. As observed in Table 199, inhibition of PKK by ISIS 482584 results in decreased activation of Factor 12 by its substrate, implying the that PKK is required for proper factor 12 activation.

In Vivo Effect of Antisense Inhibition of Murine PKK on C1-INH Antisense Oligonucleotide-induced Vascular Permeability

Vascular permeability induced by ISIS 461756, an antisense oligonucleotide which targets murine C1 inhibitor mRNA, increases vascular permeability in mice and replicates the pathology of hereditary angioedema. The effect of ISIS 482584 on this model was evaluated.

Treatment

One group of 8 mice was treated with 40 mg/kg ISIS 482584 administered subcutaneously twice a week for 3 weeks (weekly dose of 80 mg/kg). A second group of 8 mice was treated with 40 mg/kg of the control oligonucleotide, ISIS 141923, administered subcutaneously twice a week for 3 weeks (weekly dose of 80 mg/kg). A third group of 8 mice was treated with PBS administered subcutaneously twice a week for 3 weeks. On day 14, all the groups were treated with 12.5 mg/kg ISIS 461756 administered subcutaneously twice a week for 3 weeks (weekly dose of 25 mg/kg). A control group of mice was treated with PBS administered subcutaneously twice a week for 3 weeks but was not administered ISIS 461756.

At the end of the treatment period, all the groups were injected with 30 mg/kg of Evans Blue solution into the tail vein. The mice were sacrificed 30 min after the Evans Blue solution administration and colons, feet, ears, and intestines were harvested. The liver was also harvested for RNA analysis.

RNA Analysis

RNA was isolated from the liver for RT-PCR analysis of C1-INH and PKK mRNAs. The primer probe set for C1-INH is RTS3218 (forward sequence GAGTCCCCCAGAGCCTACAGT, designated herein as SEQ ID NO: 2234; reverse sequence TGTCATTTGTTATTGTGATGGCTACA, designated herein as SEQ ID NO: 2235; probe sequence CTGCCCTCTACCTGGCCAACAACCA, designated herein as SEQ ID NO: 2236). The primer probe set for PKK is RTS3287 (forward sequence ACAAGTGCATTTTACAGACCAGAGTAC, designated herein as SEQ ID NO: 2237; reverse sequence GGTTGTCCGCTGACTTTATGCT, designated herein as SEQ ID NO: 2238; probe sequence AAGCACAGTGCAAGCGGAACACCC, designated herein as SEQ ID NO: 2239). The results are presented in Table 200 as percent inhibition compared to the PBS control not treated with ISIS 461756. The data indicates that ISIS 461756 significantly reduced C1-INH mRNA expression and that treatment with ISIS 482584 significantly reduced PKK expression.

Quantification of Vascular Permeability

The harvested tissues from the feet, colon, and intestines were placed in formamide solution overnight to leach out the Evans Blue dye. The formamide solution containing feet tissue was heated to 55° C. and left overnight. The color intensity of the dye-infused formamide solution was then measured at OD 600nm . The data is presented in Table 201 as percent increase or reduction compared to the PBS control not treated with ISIS 461756. The data indicates that treatment with ISIS 482584 prevented vascular permeability induced by ISIS 461756.

In Vivo Effect of Antisense Inhibition of Murine PKK in the FeCl 3 -induced Inferior Vena Cava Thrombosis Model

ISIS 482584, which demonstrated significant in vitro and in vivo inhibition of PKK, was evaluated in the FeCl 3 -induced inferior vena cava thrombosis mouse model.

Treatment

Three groups of 8 male BALB/c mice were treated with 10 mg/kg, 20 mg/kg, or 40 mg/kg of ISIS 482584, administered subcutaneously twice a week for 3 weeks (weekly doses of 20 mg/kg, 40 mg/kg, or 80 mg/kg). Two control groups of 12 BALB/c mice each were treated with PBS, administered subcutaneously twice a week for 3 weeks. Two days after the last dose of antisense oligonucleotide or PBS, mice from all groups were anesthetized with 150 mg/kg ketamine mixed with 10 mg/kg xylazine, administered by intraperitoneal injection. Thrombus formation was induced with FeCl 3 in all groups of anesthetized mice except the first control group.

In mice undergoing FeCl 3 treatment, thrombus formation was induced by applying a piece of filter paper (2×4 mm) pre-saturated with 10% FeCl 3 solution directly on the vena cava. After 3 minutes of exposure, the filter paper was removed. Thirty minutes after the filter paper application, a fixed length of the vein containing the thrombus was dissected out for platelet analysis. Liver was collected for RNA analysis.

Quantification of Platelet Composition

Real-time PCR quantification of platelet factor-4 (PF-4) was used to quantify platelets in the vena cava as a measure of thrombus formation. PF-4 mRNA levels were measured using the murine primer probe set mPF4_LTS_00086 (forward sequence AGACCCATTTCCTCAAGGTAGAACT, designated herein as SEQ ID NO: 2240; reverse sequence CGCAGCGACGCTCATG, designated herein as SEQ ID NO: 2241; probe sequence TCTTTGGGTCCAGTGGCACCCTCTT, designated herein as SEQ ID NO: 2242). Results are presented as a percentage of PF-4 in ISIS oligonucleotide-treated mice, as compared to the two PBS-treated control groups. As shown in Table 202, treatment with ISIS 482584 resulted in a significant reduction of PF-4 in comparison to the PBS control. Therefore, reduction of PKK by the compound provided herein is useful for inhibiting thrombus formation.

In vivo Effect of Antisense Inhibition of Murine PKK in a Tail Bleeding Assay

Tail-bleeding was measured to observe whether treatment with ISIS 482584 causes excess bleeding or hemorrhage in mice.

Treatment

Groups of 10 male BALB/c mice were treated with 10 mg/kg, 20 mg/kg, or 40 mg/kg of ISIS 482584, administered subcutaneously twice a week for 3 weeks (weekly doses of 20 mg/kg, 40 mg/kg, or 80 mg/kg). A control group of 8 BALB/c mice was treated with PBS, administered subcutaneously twice a week for 3 weeks.

Tail-bleeding Assay

Two days after the final treatment of ISIS oligonucleotides or PBS, mice were placed in a tail bleeding chamber. Mice were anesthetized in the chamber with isoflurane. Then, a small piece of tail (approximately 4 mm from the tip) was cut with sterile scissors. The cut tail was immediately placed in a 15 mL Falcon tube filled with approximately 10 mL of 0.9% NaCl buffer solution warmed to 37° C. The blood was collected over the course of 40 minutes. The saline filled tubes were weighed both before and after bleeding. The results are provided in Table 203.

Treatment with ISIS 482584 did not significantly affect bleeding. These data suggest that the hemorrhagic potential of the compounds provided herein is low. These data taken with the results provided in Example 19 suggest inhibition of PKK with the compounds described herein are useful for providing antithrombotic activity without associated bleeding risk.

In vivo Effect of Antisense Inhibition of Murine PKK in the FeCl 3 Induced Mesenteric Thrombosis Model

ISIS 482584 was evaluated in the FeCl 3 induced mesenteric thrombosis mouse model.

Treatment

A group of 6-8 Swiss-Webster mice was treated with 40 mg/kg of ISIS 482584, administered subcutaneously twice a week for 3 weeks (weekly dose of 80 mg/kg). A control group of 6 Swiss-Webster mice was treated with PBS, administered subcutaneously twice a week for 3 weeks. Two days after the last dose of antisense oligonucleotide or PBS, mice from all groups were anesthetized with 75 mg/kg ketamine mixed with 25 mg/kg xylazine, administered by subcutaneous injection.

Rhodamine 6G dye at a dosage of 5 mg/kg was injected subcutaneously to stain platelets. Alexa-647-labeled anti-fibrinogen antibody at a dosage of 1 mg/kg was injected via tail vein injection to stain fibrin. The abdomen was opened by a middle incision. The visceral mesentery was spread on a glass coverslip and the mesenteric arterioles (70-120 μm) were located by observation under a microscope. Thrombus formation was induced by applying of cotton threads (2×0.3 mm) pre-saturated with 6% FeCl 3 solution directly on the target vessel. After three minutes of exposure, the thread was removed and the color intensities of both the dyes were recorded by fluorescent microscopy (Olympus FluoView 1000 confocal laser scanning microscope) with appropriate filters for 70 min.

The results for platelet aggregation in the control and treatment groups are presented in Table 204, expressed in arbitrary units (a.u.). Platelet aggregation was reduced in mice treated with ISIS 482584 at a dose of 80 mg/kg/week as compared to mice treated with PBS. The results for fibrin formation in the control and treatment groups are presented in Table 205, also expressed in arbitrary units (a.u.). Fibrin formation was reduced in mice treated with ISIS 482584 at a dose of 80 mg/kg/week as compared to mice treated with PBS. Therefore, these results suggest that ISIS 482584 inhibits thrombus formation.

In vivo Effect of Antisense Inhibition of Murine PKK in the Stenosis-induced Inferior Vena Cava Thrombosis Model

ISIS 482584 was evaluated in the stenosis-induced inferior vena cava (IVC) thrombosis model. Reduced blood flow and endothelial damage are hallmarks of this model, also known as the St. Tomas model.

Treatment

Four groups of 6-8 BALB/c mice were treated with 5 mg/kg, 10 mg/kg, 20 mg/kg, or 40 mg/kg of ISIS 482584, administered subcutaneously twice a week for 3 weeks (weekly doses of 10 mg/kg, 20 mg/kg, 40 mg/kg, or 80 mg/kg). A control group of 8 BALB/c mice was treated with PBS, administered subcutaneously twice a week for 3 weeks. Two days after the last dose of antisense oligonucleotide or PBS was administered, mice from all groups were anesthetized with 2.5% inhalant isoflurane. The IVC of the mice was exposed via a midline abdominal incision below the left renal vein, and was separated from the abdominal aorta by blunt dissection. A 6-0 silk tie (Ethicon, UK) was placed behind the blood vessel just below the left renal vein and a metal 4-0 suture (Ethicon, UK) was placed longitudinally over the IVC to tie the silk tie on top. The metal suture was then removed. Two neurovascular surgical clips (Braun Medical Inc, PA) were placed at two separate positions below the ligation for 20 seconds each, after which they were removed. The abdominal cavity contents were then replaced and the abdomen was closed. After 24 hrs, the IVC was exposed and checked for thrombi formation. All thrombi formed were collected and fixed in 10% formalin for 24 hrs.

The thrombi were weighed and the results are presented in Table 206, expressed in milligrams. As demonstrated by the results, treatment with increasing doses of ISIS 482584 resulted in corresponding decrease in thrombus weight. The results indicate that antisense inhibition of PKK is useful for inhibiting thrombus formation.

Inhibition of Murine PKK with an Antisense Oligonucleotide Comprising a GalNAc 3 Conjugate Group

ISIS 482584 and ISIS 722059, shown in the table below, were tested for their effects on murine PKK mRNA in vivo.

Subscripts: “e” indicates 2′-MOE modified nucleoside; “d” indicates β-D-2′-deoxyribonucleoside; “s” indicates phosphorothioate internucleoside linkages (PS); “o” indicates phosphodiester internucleoside linkages (PO); and “o” indicates —O—P(═O)(OH)—. Superscript “m” indicates 5-methlcytosines. The structure of “GalNAc 3 -7” is shown in Example 48.

Treatment

Four groups of four C57Bl/6J-Tyr c-2J mice each were treated with 5.0 mg/kg, 10.0 mg/kg, 20.0 mg/kg, or 40.0 mg/kg of ISIS 482584, administered subcutaneously twice a week for 3 weeks (weekly doses of 10.0 mg/kg, 20.0 mg/kg, 40.0 mg/kg, or 80.0 mg/kg). Four groups of four BALB/c mice each were treated with 1.0 mg/kg, 2.0 mg/kg, 4.0 mg/kg, or 8.0 mg/kg of ISIS 722059, administered subcutaneously twice a week for 3 weeks (weekly doses of 2.0 mg/kg, 4.0 mg/kg, 8.0 mg/kg, or 16.0 mg/kg). A control group of four BALB/c mice was treated with PBS, administered subcutaneously twice a week for 3 weeks. Three days after the last dose of antisense oligonucleotide or PBS, mice from all groups were anesthetized with vaporized isoflurane in air at 2.5% for induction followed by 1-2% isoflurane by nosecone for maintenance. This was followed by cervical dislocation. Following euthanasia, liver was collected for RNA analysis.

RNA Analysis

RNA was extracted from liver tissue for real-time PCR analysis of PKK. PKK mRNA levels were measured using the murine primer probe set (forward sequence ACAAGTGCATTTTACAGACCAGAGTAC, designated herein as SEQ ID NO: 2231; reverse sequence GGTTGTCCGCTGACTTTATGCT, designated herein as SEQ ID NO: 2232; probe sequence AAGCACAGTGCAAGCGGAACACCC, designated herein as SEQ ID NO: 2233). Results are presented as percent inhibition of PKK, relative to PBS control. As shown in Table 208 below, Isis 722059, comprising a GalNAc 3 conjugate group, reduced PKK mRNA significantly more potently than the parent antisense oligonucleotide, Isis 482584. This result is consistent with the results in the above examples, in which antisense oligonucleotides comprising a GalNAc 3 conjugate group were significantly more potent than their parent antisense oligonucleotides, for many target genes in both mouse and human. Thus, it is expected that human PKK antisense oligonucleotides comprising a GalNAc 3 conjugate group would likewise reduce human PKK mRNA significantly more potently than their parent antisense oligonucleotides that do not comprise a conjugate group.

Inhibition of Human PKK with an Antisense Oligonucleotide Comprising a GalNAc 3 Conjugate Group

ISIS 546254 and ISIS 721744, shown in the table below, were tested for their effects on human PKK mRNA in vitro.

Subscripts: “e” indicates 2′-MOE modified nucleoside; “d” indicates β-D-2′-deoxyribonucleoside; “s” indicates a phosphorothioate internucleoside linkages (PS); and “o” indicates a phosphodiester internucleoside linkage (PO). Superscript “m” indicates 5-methlcytosines. The structure of “GalNAc 3 -7” is shown in Example 48, and “GalNAc 3 -7 a-o′ ” indicates a GalNAc 3 -7 conjugate group in which the cleavable moiety is —O—P(═O)(OH)—.

Primary human hepatocyte co-cultures that include stromal cells in order to mimic the physiological microenviroment of the liver in vitro (HepatoPac kit HPHU-TX-96S, Hepregen, Medford, Mass.) were used according to the manufacturer's instructions. A concentration of Isis oligonucleotide listed in table below or PBS was added to each well in the absence of any transfection reagent. 96 hours later, cells were lysed and RNA was isolated from the cells. PKK mRNA levels were measured by quantitative real-time PCR using primer probe set RTS3454 and normalized to total RNA content, as measured by RIBOGREEN®. The results are presented in the table below as percent inhibition of PKK mRNA levels, relative to PBS treated cells; and IC 50 values were calculated using a 4 parameter logistic model (JMP Software, Cary, N.C.). The results show that, under free uptake conditions in which no reagents or electroporation techniques were used to artificially promote entry of the oligonucleotides into cells, the oligonucleotide comprising a GalNAc conjugate was significantly more potent than the parent oligonucleotide that does not comprise a GalNAc conjugate.