Structurally constrained phosphonate internucleotide linkage impacts oligonucleotide-enzyme interaction, and modulates siRNA activity and allele specificity

Ken Yamada; Samuel Hildebrand; Sarah M Davis; Rachael Miller; Faith Conroy; Ellen Sapp; Jillian Caiazzi; Julia F Alterman; Loic Roux; Dimas Echeverria; Matthew R Hassler; Edith L Pfister; Marian DiFiglia; Neil Aronin; Anastasia Khvorova

doi:10.1093/nar/gkab1126

. 2021 Nov 25;49(21):12069–12088. doi: 10.1093/nar/gkab1126

Structurally constrained phosphonate internucleotide linkage impacts oligonucleotide-enzyme interaction, and modulates siRNA activity and allele specificity

Ken Yamada ^1,^✉, Samuel Hildebrand ², Sarah M Davis ³, Rachael Miller ^4,⁵, Faith Conroy ^6,⁷, Ellen Sapp ⁸, Jillian Caiazzi ⁹, Julia F Alterman ¹⁰, Loic Roux ¹¹, Dimas Echeverria ¹², Matthew R Hassler ¹³, Edith L Pfister ¹⁴, Marian DiFiglia ¹⁵, Neil Aronin ^16,¹⁷, Anastasia Khvorova ^18,^19,^✉

¹ RNA Therapeutics Institute, University of Massachusetts Medical School, 368 Plantation Street, Worcester, MA 01605, USA

² RNA Therapeutics Institute, University of Massachusetts Medical School, 368 Plantation Street, Worcester, MA 01605, USA

³ RNA Therapeutics Institute, University of Massachusetts Medical School, 368 Plantation Street, Worcester, MA 01605, USA

⁴ RNA Therapeutics Institute, University of Massachusetts Medical School, 368 Plantation Street, Worcester, MA 01605, USA

⁵ Department of Medicine, University of Massachusetts Medical School, Worcester, MA, USA

⁶ RNA Therapeutics Institute, University of Massachusetts Medical School, 368 Plantation Street, Worcester, MA 01605, USA

⁷ Department of Medicine, University of Massachusetts Medical School, Worcester, MA, USA

⁸ Department of Neurology, Harvard Medical School and MassGeneral Institute for Neurodegenerative Disease, Charlestown, MA, USA

⁹ RNA Therapeutics Institute, University of Massachusetts Medical School, 368 Plantation Street, Worcester, MA 01605, USA

¹⁰ RNA Therapeutics Institute, University of Massachusetts Medical School, 368 Plantation Street, Worcester, MA 01605, USA

¹¹ RNA Therapeutics Institute, University of Massachusetts Medical School, 368 Plantation Street, Worcester, MA 01605, USA

¹² RNA Therapeutics Institute, University of Massachusetts Medical School, 368 Plantation Street, Worcester, MA 01605, USA

¹³ RNA Therapeutics Institute, University of Massachusetts Medical School, 368 Plantation Street, Worcester, MA 01605, USA

¹⁴ Department of Medicine, University of Massachusetts Medical School, Worcester, MA, USA

¹⁵ Department of Neurology, Harvard Medical School and MassGeneral Institute for Neurodegenerative Disease, Charlestown, MA, USA

¹⁶ RNA Therapeutics Institute, University of Massachusetts Medical School, 368 Plantation Street, Worcester, MA 01605, USA

¹⁷ Department of Medicine, University of Massachusetts Medical School, Worcester, MA, USA

¹⁸ RNA Therapeutics Institute, University of Massachusetts Medical School, 368 Plantation Street, Worcester, MA 01605, USA

¹⁹ Program in Molecular Medicine, University of Massachusetts Medical School, Worcester, MA, USA

^✉

To whom correspondence should be addressed. Tel: +1 774 455 3638; Email: Anastasia.Khvorova@umassmed.edu

^✉

Correspondence may also be addressed to Ken Yamada. Tel: +1 774 455 3636; Email: Ken.Yamada@umassmed.edu

PMCID: PMC8643693 PMID: 34850120

Abstract

Oligonucleotides is an emerging class of chemically-distinct therapeutic modalities, where extensive chemical modifications are fundamental for their clinical applications. Inter-nucleotide backbones are critical to the behaviour of therapeutic oligonucleotides, but clinically explored backbone analogues are, effectively, limited to phosphorothioates. Here, we describe the synthesis and bio-functional characterization of an internucleotide (E)-vinylphosphonate (ⁱE-VP) backbone, where bridging oxygen is substituted with carbon in a locked stereo-conformation. After optimizing synthetic pathways for ⁱE-VP-linked dimer phosphoramidites in different sugar contexts, we systematically evaluated the impact of the ⁱE-VP backbone on oligonucleotide interactions with a variety of cellular proteins. Furthermore, we systematically evaluated the impact of ⁱE-VP on RNA-Induced Silencing Complex (RISC) activity, where backbone stereo-constraining has profound position-specific effects. Using Huntingtin (HTT) gene causative of Huntington's disease as an example, ⁱE-VP at position 6 significantly enhanced the single mismatch discrimination ability of the RISC without negative impact on silencing of targeting wild type htt gene. These findings suggest that the ⁱE-VP backbone can be used to modulate the activity and specificity of RISC. Our study provides (i) a new chemical tool to alter oligonucleotide-enzyme interactions and metabolic stability, (ii) insight into RISC dynamics and (iii) a new strategy for highly selective SNP-discriminating siRNAs.

Graphical Abstract

INTRODUCTION

Oligonucleotide therapeutics, such as small interfering RNAs (siRNAs), are a promising class of human therapeutics that directly and specifically modulate the expression of disease-causing messenger RNA (mRNA) (1–4). Several oligonucleotide drugs have been FDA-approved, and multiple compounds are in the late stages of clinical development (3,4). The clinical efficacy and in vivo stability of these drugs depends on chemical modifications to the nucleotide backbone (i.e. phosphodiester linkages and sugars), which is the major driving force behind the formation of specific nucleic acid-protein complexes (1,2).

Various chemically-modified phosphate backbones, including peptide nucleic acid and morpholino oligomers for splice-switching oligonucleotides, have been developed to modulate the functionality of oligonucleotides for biological and clinical applications (1–4). However, phosphorothioate (PS) is currently the only inter-nucleotide backbone modification applied to siRNAs and RNaseH-activating antisense oligonucleotides (ASOs) in the clinic (5–10). Unlike the other backbone variants, PS modifications have minimal structural impact and thus allow oligonucleotides to maintain efficient interactions with their effector proteins—e.g., RNA-induced Silencing Complex (RISC) and RNaseH (5). Although the widely-used PS modification contributes to metabolic stability and bio-distribution properties of siRNAs and ASOs in vivo, it can cause toxicity dependent on sequence, position and combination of 2′-substituents, potentially limiting the oligonucleotide therapeutic index (5,11–15). To reduce the chance of PS-related toxicity one of options could be reducing the numbers of PS where PS is converted to canonical phosphodiester backbone compromising metabolic stability. Thus, developing alternative, metabolically-stable phosphodiester bond analogues with bio-compatible configurations would be expected to offer therapeutic advances to the oligonucleotides.

One chemical option for a close-to-natural backbone modification is vinylphosphonate, or (E)-VP, a phosphate mimic that replaces the oxygen-phosphorus bond of the phosphodiester linkage with a carbon-phosphorus bond (16). While (E)-VP is not likely to alter the charge or molecular size of the backbone, this modification does restrict the torsion angle to 180^o, which may have an impact on protein recognition to provide the desired metabolic impact (17–20). Modifications that induce structural constraint exerts beneficial effects when applied to the sugars of oligonucleotides. For instance, 2′-fluoro and LNA (locked nucleic acid) sugar modifications can increase nuclease stability and target binding affinity while maintaining compatibility with oligonucleotide effector proteins (7). Synthesis and bio-chemical properties of internucleotide (E)-VP backbone (ⁱE-VP) in a context of single and double stranded DNA oligonucleotides were thoroughly investigated by Hayes, et al. where it was found that the backbone structural constraining induces inhibition of DNA helicase activity and DNA-binding nuclease activity (21–25). However, the impact of ⁱE-VP backbone on RNA oligonucleotides has not established yet in a context of RNA therapeutics research.

Phosphonate and (E)-VP modifications were first explored during the 1980–1990s for the synthesis of different nucleotide-based antiviral agents. Incorporating phosphonate analogues maintained the desired biological interactions, while enhancing metabolic stability (16,26,27). More recently, the impact of a 5′-terminal (E)-VP modification was studied in the context of single-stranded and double-stranded RNAs. Introduction of (E)-VP on the 5′ end enhanced in vivo efficacy of single-stranded, hydrophobically-modified, and GalNAc-conjugated siRNAs (28–35). The enhanced efficacy was attributed to a combination of increased phosphatase resistance and favorable shape-fitting of 5′-(E)-VP structure in the MID domain of human Ago2 – a protein in RISC (28–35). To date, the biological impact of (E)-VP on therapeutic RNA modalities has been evaluated only as a 5′ terminal modification of oligonucleotides or as a monomer, partially because of synthetic limitations. Optimizing synthetic methods to achieve an ‘inter-nucleotide’ (E)-VP or (ⁱE-VP) backbone in the context of therapeutic oligonucleotides will enable investigation into how ⁱE-VP-induced constraint of the backbone impacts biological behavior (Figure 1A, B).

Figure 1. — (A) Structures of canonical phosphodiester backbone and (B) ⁱE-VP backbone.

Here, we report an optimized synthetic pathway for sufficient production of ⁱE-VP-containing dimer phosphoramidites in the context of different sugar modifications. Using dimer phosphoramidite building blocks, we then synthesized panels of ⁱE-VP-modified oligonucleotides, and evaluated the impact of ⁱE-VP on nuclease and reverse transcriptase activity, and on siRNA efficacy and specificity.

MATERIALS AND METHODS

General experimental methods

The NMR spectra were recorded using Brucker 500 MHz spectrometer. ¹H, ¹³C, ¹⁹F and ³¹P NMR spectra were recorded at 500 MHz (¹H-NMR, 500 MHz; ¹³C-NMR, 125 MHz; ³¹P-NMR, 202 MHz). The chemical shifts were measured from tetramethylsilane (0 ppm), CDCl₃ (7.26 ppm), DMSO-d₆ (2.49 ppm) and CD₃CN-d₃ (1.93 ppm) for ¹H-NMR spectra, CDCl₃ (77.0 ppm), DMSO-d₆ (39.7 ppm) and CD₃CN-d₃ (1.30 ppm) for ¹³C-NMR spectra, and 85% H₃PO₄ for ³¹P-NMR spectra as external standards. High-resolution electrospray ionization mass spectrometry (HR-ESI-MS) analysis for the monomers and dimer nucleos(t)ides were performed on a Thermo Scientific Orbitrap Velos Pro mass spectrometer in the positive ion mode. The mass analysis of oligonucleotides was conducted by LC–MS on an Agilent 6530 accurate-mass Q-TOF LC/MS (Agilent technologies, Santa Clara, CA). Thin-layer chromatography (TLC) analysis was conducted using silica gel-coated aluminium-backed TLC plates (0.20 mm thickness) containing F-254 UV indicator (Silicycle Inc., Canada). Column chromatography was performed with silica column (Flash Column Silica-CS-Agela; 12–330 g; 40–60 μm), using the CombiFlash Rf200 (Teledyne Isco, Inc.) Companion Chromatograph. The synthesis of modified oligonucleotides was performed using MerMaid-12 DNA/RNA synthesizer (Bio automation, USA). Analytical anion-exchange HPLC and reverse phase HPLC were performed on Agilent 1260 Infinity Analytical SFC System combined with an Agilent 1100 series quaternary pump with a degasser. Purified oligonucleotides were desalted by Sephadex G-25 (GE Healthcare).

Synthesis of ⁱE-VP-linked dimer phosphoramidites

3-N-Benzoyl-5′-O-[bis(4-methoxyphenyl)phenylmethyl]-2′-deoxy-2′-fluoro-3′-O-(tert-butyldiphenylsilyl)-uridine (2a)

Anhydrous solution of compound 1a (16.6 g, 20.8 mmol) in pyridine (100 ml) was added anhydrous N,N-diisopropylethylamine (6.5 ml, 37.4 mmol) and benzoyl chloride (3.6 ml, 31.2 mmol). After the mixture was stirred for 4 h at rt, excess pyridine was evaporated and diluted with CH₂Cl₂. The organic solution was washed by sat. aq. NaHCO₃. The organic layer was collected, dried over MgSO₄, filtered and evaporated. Obtained crude material was purified by silica gel column chromatography (hexane-ethyl acetate, 4:1 to 1:1) yielding compound 2a as a slightly yellow foam (14.5 g, 78%); ¹H NMR (500 MHz, CDCl₃) δ 7.88–7.87 (m, 2H), 7.84 (d, 1H, J = 8.3 Hz), 7.67–7.58 (m, 5H), 7.48–7.45 (m, 4H), 7.39–7.32 (m, 4H), 7.25–7.23 (m, 3H), 7.18–7.17 (m, 2H), 7.12–7.07 (m, 4H), 6.80–6.75 (m, 4H), 6.08 (dd, 1H, J_HH = 1.5 Hz, J_HF = 15.2 Hz), 5.14, (d, 1H, J_HH= 8.3 Hz), 4.59 (ddd, 1H, J_HH = 3.7, 1.5 Hz, J_HF = 51.9 Hz), 4.43 (ddd, 1H, J_HH = 7.4, 4.0 Hz, J_HF = 19.1 Hz), 4.24–4.23 (m, 1H), 3.79 (s, 6H), 3.62 (dd, 1H, J_HH = 11.2, 2.0 Hz), 3.35 (dd, 1H, J_HH = 11.1, 2.0 Hz), 1.00 (s, 9H); ¹³C NMR (126 Hz, CDCl₃) δ 168.4, 161.8, 158.72, 158.66, 148.9, 143.9, 139.4, 135.71. 135.70, 135.1, 134.8, 134.7, 132.3, 132.2, 131.3, 130.4, 130.2, 130.1, 129.1, 128.2, 128.0,127.91, 127.89, 127.2, 113.19, 113.16, 102.2, 92.5 (d, J_CF = 194.4 Hz), 87.7 (d, J_CF = 34.5 Hz), 87.2, 82.4, 70.0 (d, J_CF = 15.4 Hz), 60.7, 60.4, 55.2, 26.6; ¹⁹F NMR (470 MHz, CDCl₃) δ -201.48 (ddd, J = 52.0, 17.3, 17.3 Hz); HRMS (ESI) calcd. for C₅₃H₅₁FN₂O₈SiNa⁺ [M + Na]⁺m/z 913.3291, found m/z 913.3294.

3-N-Benzoyl-2′-deoxy-2′-fluoro-3′-O-(tert-butyldiphenylsilyl)-uridine (3a)

Compound 2a (14.5 g, 16.3 mmol) was dissolved into 3% trichloroacetic acid/CH₂Cl₂ solution (200 ml) containing triethylsilane (8.0 ml, 50.1 mmol) and stirred for 1 h at rt. After the solution was washed by sat. aq. NaHCO₃ three times, collected organic layer was dried over MgSO₄, filtered, and evaporated. Obtained crude material was purified by silica gel column chromatography (hexane/ethyl acetate, 4:1 to 3:7) yielding compound 3a as a white foam (8.67 g, 91%); ¹H NMR (500 MHz, CDCl₃) δ 7.89–7.88 (m, 2H), 7.68–7.64 (6H, m), 7.51–7.45 (m, 4H), 7.42–7.38 (m, 4H), 5.93 (dd, 1H, J_HH= 2.9 Hz, J_HF = 15.1 Hz), 5.73 (d, 1H, J_HH = 8.2 Hz), 4.74 (ddd, 1H, J_HH = 4.1, 3.2 Hz, J_HF = 52.2 Hz), 4.31 (ddd, 1H, J_HH = 5.8, 4.7, J_HF = 15.4 Hz), 4.11–4.09 (m, 1H), 3.82–3.79 (m, 1H), 3.39 (ddd, 1H, J_HH = 12.1, 5.6, 1.5 Hz), 1.64 (br-s, 1H), 1.11 (s, 9H); ¹³C NMR (126 Hz, CDCl₃) δ 168.3, 161.8, 149.0, 140.5, 135.7, 135.2, 132.8, 132.3, 131.3, 130.5, 130.4, 130.3, 129.2, 128.02, 127.96, 102.4, 91.8 (d, J_CF = 91.8 Hz), 89.5 (d, J_CF = 33.6 Hz), 69.5 (d, J_CF = 69.5 Hz), 60.3, 26.8; ¹⁹F NMR (470 MHz, CD₃CN) δ -203.20 (ddd, J = 52.0, 17.4, 17.4 Hz); HRMS (ESI) calcd. for C₃₂H₃₃FN₂O₆SiNa⁺ [M + Na]⁺m/z 611.1984, found m/z 611.1988.

6′-Dibromo-(5′-deoxy-5′-methylidene)-3′-O-(tert-butyldiphenylsilyl)-2′-deoxy-2′-fluoro-uridine (5a)

Anhydrous solution of compound 3a (6.5 g, 11.0 mmol) was added IBX (7.7 g, 27.6 mmol) and stirred for 2 h at 85°C. After cooling the mixture in an ice bath, the precipitate in the solution was filtered off through celite. Collected eluent was evaporated, co-evaporated with anhydrous CH₃CN three times under argon atmosphere, and obtained compound 4a as a white foam was used without further purification. In a separate flask, anhydrous CH₂Cl₂ (25 ml) solution containing CBr₄ (7.3 g, 22.1 mmol) was added PPh₃ (11.6 g, 44.2 mmol) at 0°C and stirred for 0.5 h at 0°C. To this solution, anhydrous CH₂Cl₂ solution (25 ml) of compound 4a was added dropwise (10 min) at 0°C and stirred for 2 h at rt. After diluting with CH₂Cl₂, the organic solution was washed by aq. sat. NH₄Cl, dried over MgSO₄, filtered, and evaporated. Obtained material was dissolved into minimum amount of diethyl ether and added dropwise to excess diethyl ether solution under vigorously stirring at 0°C. Precipitate in solution was filtered off through celite and eluents was evaporated. Obtained crude material was purified by silica gel column chromatography (hexane/ethyl acetate, 9:1 to 1:1) yielding compound 5a as a white foam (4.3 g, 52% in 2 steps). ¹H NMR (500 MHz, CDCl₃) δ 7.68–7.84 (m, 2H), 7.70–7.65 (m, 3H), 7.60–7.58 (m, 2H), 7.52–7.49 (m, 2H), 7.42–7.36 (m, 4H), 7.31–7.28 (m, 2H), 7.09 (d, 1H, J = 8.2 Hz), 6.25 (d, 1H, J = 8.9 Hz), 5.75 (dd, 1H, J_HF = 8.24 Hz), 5.49 (dd, 1H, J_HF = 21.4 Hz), 4.77 (t, 1H, J_HH = 8.5 Hz, J_HF = 8.5 Hz), 4.38 (dd, 1H, J_HH = 4.1 Hz, J_HF = 52.1 Hz), 4.25 (ddd, 1H, J_HH = 8.1, 4.9 Hz, J_HF = 19.4 Hz), 1.10 (s, 9H); ¹³C NMR (126 Hz, CDCl₃) δ 167.9, 161.6, 148.3, 141.4, 135.8, 134.7 (d, J_C-Br = 139.0 Hz), 132.5, 132.2, 131.1, 130.5, 130.3, 130.2, 129.2, 127.9, 102.7, 97.3, 93.3 (d, J_CF = 39.1 Hz), 91.5 (d, J_CF = 190.7 Hz), 82.4, 73.9 (d, J_CF = 16.4 Hz), 26.7; ¹⁹F NMR (470 MHz, CD₃CN) δ -198.53 (ddd, J = 52.0, 20.8, 20.8 Hz); HRMS (ESI) calcd. for C₃₃H₃₂Br₂FN₂O₅Si⁺ [M + H]⁺m/z 741.0426, found m/z 741.0438.

(E)/(Z)-6′-Bromo-(5′-deoxy-5′-methylidene)-2′-deoxy-2′-fluoro-3′-O-(tert-butyldiphenylsilyl)-uridine (6a-E and 6a-Z)

Anhydrous solution of compound 5a (4.2 g, 5.66 mmol) in DMF (25 ml) was added dimethyl phosphite (2.09 ml, 22.6 mmol) and triethylamine (1.58 ml, 11.3 mmol) at 0°C, and then stirred overnight at rt. After the solution was diluted with ethyl acetate, the organic solution was washed with aq. sat. NH₄Cl and brine. Then the organic solution was dried over MgSO₄, filtered and evaporated. Obtained crude material was purified repeatedly by silica gel column chromatography (hexane/ethyl acetate, 9:1 to 1:1) until all isomeric-pure compounds were collected separately, giving compound 6a-E (1.95 g, 52%); ¹H NMR (500 MHz, CDCl₃) δ 7.87–7.85 (m, 2H), 7.89–7.85 (m, 3H), 7.61–7.59 (m, 2H), 7.52–7.48 (m, 2H), 7.45–7.32 (m, 6H), 7.08 (d, 1H, J = 8.2 Hz), 6.49 (d, 1H, J = 13.7 Hz), 5.99 (dd, 1H, J = 13.7, 8.1 Hz), 5.75 (d, 1H, J = 8.2 Hz), 5.63 (d, 1H, J_HF = 19.8 Hz), 4.43 (dd, 1H, J_HF = 52.6 Hz, J_HH = 4.3 Hz), 4.42 (t, 1H, J_HH = 8.0 Hz), 4.07 (ddd, J_HF = 19.5 Hz, J_HH = 7.8, 4.7 Hz), 1.08 (s, 9H); ¹³C NMR (126 Hz, CDCl₃) δ 148.4, 140.4, 135.8, 135.7, 135.3, 133.3, 132.3, 132.4, 132.1, 131.1, 130.5, 130.4, 130.3, 129.2, 127.95, 127.93, 112.4, 102.7, 91.7 (d, J_CF = 36.3 Hz), 91.6 (d, J_CF = 191.6 Hz), 82.8, 73.9 (d, J_CF = 16.4 Hz), 26.7, 19.1; ¹⁹F NMR (470 MHz, CD₃CN) δ – 199.04 (ddd, J = 52.0, 20.8, 20.8 Hz); HRMS (ESI) calcd. for C₃₃H₃₃BrFN₂O₅Si⁺ [M + H]⁺m/z 663.1321, found m/z 663.1329; and 6a-Z (0.58 g, 15%); ¹H NMR (500 MHz, CDCl₃) δ 7.87–7.85 (m, 2H), 7.68–7.65 (m, 3H), 7.61–7.59 (m, 2H), 7.52–7.48 (m, 2H), 7.42–7.39 (m, 2H), 7.34–7.29 (m, 4H), 7.12 (d, 1H, J = 8.2 Hz), 6.51 (d, 1H, J = 7.4 Hz), 5.96 (dd, 1H, J= 8.4 Hz, 7.4 Hz), 5.75 (d, 1H, J = 8.2 Hz), 5.57 (dd, 1H, J_HF = 20.6 Hz, J_HH = 1.2 Hz), 5.04 (dd, 1H, J = 8.2, 8.2 Hz), 4.48 (J_HF = 53.1 Hz, J_HH = 3.5 Hz), 4.24 (ddd, 1H, J_HF = 18.6 Hz, J_HH = 7.8, 4.9 Hz), 1.09 (s, 9H); ¹³C NMR (126 Hz, CDCl₃) δ 168.0, 161.7, 148.4, 141.4, 135.9, 135.8, 135.2, 132.6, 132.5, 131.2, 130.6, 130.5, 130.2, 130.1, 129.2, 127.8, 127.7, 114.5, 102.6, 93.0 (d, J_CF = 37.2 Hz), 91.6 (d, J_CF = 191.6 Hz), 80.3, 74.3 (d, J_CF = 16.4 Hz), 26.7, 19.1; HRMS (ESI) calcd. for C₃₃H₃₃BrFN₂O₅Si ⁺ [M + H]⁺m/z 663.1321, found m/z 663.1331.

5′-O-[Bis(4-methoxyphenyl)phenylmethyl]-2′-O-methyl-uridyl-methylphosphono-[3′(O)→5′(C)]-3-N-benzoyl-5′-deoxy-5′-methylidene-3′-O-(tert-butyldiphenylsilyl)-2′-deoxy-2′-fluoro-uridine (7a)

Anhydrous compound 6a-E (1.95 g, 2.94 mmol) and Pd(OAc)₂ (125 mg, 0.59 mmol) and [1,1′-Bis(diphenylphosphino)ferrocene]dichloropalladium (II) (0.65 g, 1.18 mmol) were purged with argon, and then dissolved into anhydrous THF (50 ml). After adding propylene oxide (2.06 ml, 29.4 mmol), compound S2a (2.07 g, 3.24 mmol) was added in one portion and stirred at for 4 h at 70°C. After removing solvent under reduced pressure, the crude mixture was purified by silica gel column chromatography (hexane-ethyl acetate, 50:50 to 0:100) and obtained fractions containing compound 7a were further purified by silica gel column chromatography (CH₂Cl₂/MeOH, 100:0 to 95:5) yielding compound 7a as a mixture of diastereo-isomers (2.04 g, 57%); ³¹P NMR (202 MHz, CDCl₃) δ 18.3, 18.2; ¹⁹F NMR (470 MHz, CDCl₃) δ –197.45 (ddd, J = 52.1, 20.8, 20.8 Hz), –197.69 (ddd, J = 52.0, 20.8, 20.8 Hz); HRMS (ESI) calcd. for C₆₅H₆₆FN₄O₁₅PSiNa ⁺ [M + Na]⁺m/z 1243.3908, found m/z 1243.3915; For ¹H and ¹³C NMR analysis, part of diastereoisomers were purified by preparative TLC. Least polar isomer ¹H NMR (500 MHz, CDCl₃) δ 8.98 (br-s, 1H), 7.88–7.84 (m, 3H), 7.69–7.64 (m, 3H), 7.58–7.64 (m, 2H), 7.51–7.48 (m, 2H), 7.40–7.29 (m, 10H), 7.26–7.23 (m, 5H), 7.01 (d, 1H, J = 8.0 Hz), 6.85–6.83 (m, 4H), 6.04 (d, 1H, J = 3.5 Hz), 5.89 (ddd, 1H, J = 18.8, 17.2, 1.4 Hz), 5.73 (d, 1H, J = 8.2 Hz), 5.59 (dd, 1H, J_HF = 20.4 Hz, J_HH = 1.3 Hz), 5.24 (d, 1H, J = 8.2 Hz), 5.06 (dd, 1H, J = 11.7, 6.1 Hz), 4.43–4.38 (m, 1H), 4.32–4.22 (m, 2H), 4.14–4.07 (m, 2H), 3.78 (s, 6H), 3.69 (d, J_PH = 11.3 Hz), 3.61 (s, 3H), 3.60–3.55 (m, 1H), 3.42–3.41 (m, 1H), 1.08 (s, 9H); ¹³C NMR (125 MHz, CDCl₃) δ 167.9, 162.8, 161.5, 158.79, 158.75, 150.1, 148.4, 147.4 (d, J_CP = 6.4 Hz), 144.1, 140.9, 139.6, 134.7, 135.7, 135.3, 134.92, 134.87, 132.3, 132.0, 130.4, 130.34, 130.28, 130.2, 130.1, 129.2, 128.2, 128.11, 128.06, 127.98, 127.95, 127.89, 127.3, 118.0 (d, J_CP = 190.8 Hz), 113.3, 103.0, 102.4, 92.1 (d, J_CF = 7.3 Hz), 91.1 (d, J_CF = 163.5 Hz), 87.3, 86.8, 82.8, 82.4, 81.7, 81.6 (d, J_CF = 15.4 Hz), 74.3 (d, J_CF = 17.3 Hz), 71.6 (d, J_CP = 4.6 Hz), 61.2, 58.6, 58.5, 55.2 52.9, 52.7 (d, J_CP = 6.3 Hz), 45.8, 26.7, 19.1; most polar isomer ¹H NMR (500 MHz, CDCl₃) δ 8.47 (br-s, 1H), 7.87–7.84 (m, 3H), 7.65–7.64 (m, 3H), 7.59–7.56 (m, 2H), 7.51–7.48 (m, 2H), 7.42–7.23 (m, 15H), 7.90 (d, 1H, J = 8.2 Hz), 6.87–6.85 (m, 4H), 6.73 (ddd, 1H, J = 22.7, 17.1, 5.8 Hz), 6.02 (ddd, 1H, J = 18.6, 7.0, 1.1 Hz), 6.00 (d, 1H, J = 3.3 Hz), 5.78 (d, 1H, J = 8.2 Hz), 5.73 (dd, 1H, J_HF = 19.3 Hz, J_HH = 1.4 Hz), 5.23 (d, 1H, J = 8.2 Hz), 5.14–5.10 (m, 1H), 4.59–4.56 (m, 1H), 4.37 (ddd, J_HF = 52.3 Hz, J_HH = 3.4, 1.4 Hz), 4.27–4.26 (m, 1H), 4.10–4.07 (m, 1H), 3.98 (dd, 1H, J = 4.7, 3.5 Hz), 3.80 (s, 6H), 3.64 (dd, 1H, J = 11.1, 2.3 Hz), 3.52–3.47 (m, 4H), 3.38 (s, 3H), 1.08 (s, 9H); ¹³C NMR (125 MHz, CDCl₃) δ 167.9, 162.6, 161.6, 158.9, 158.8, 149.9, 148.5, 146.6 (d, J_CP = 6.4 Hz), 143.9, 140.2, 139.6, 135.73, 135.69, 134.9, 134.7, 132.11, 132.09, 131.1, 130.4, 130.4, 130.3, 130.24, 130.20, 130.1, 129.2, 128.3, 128.13, 128.07, 127.97, 127.91, 127.4, 120.0 (d, J_CP = 192.5 Hz), 113.3, 103.0, 102.4, 91.4 (d, J_CF = 192.5 Hz), 91.2 (d, J_CF = 33.6 Hz), 87.4, 86.9, 82.5, 82.2, 81.8 (d, J_CP = 7.3 Hz), 74.1 (d, J_CF = 15.5 Hz), 72.0 (d, J_CP = 5.4 Hz), 61.1, 58.7, 58.5, 55.3, 52.2 (d, J_CP = 5.5 Hz), 26.7, 19.2.

5′-O-[Bis(4-methoxyphenyl)phenylmethyl]-2′-O-methyl-uridyl-methylphosphono-[3′(O)→5′(C)]-3-N-benzoyl-2′,5′-dideoxy-2′-fluoro-5′-methylidene-uridine(8a)

Compound 7a (2.0 g, 1.64 mmol) in anhydrous THF (22.5 ml) was added 1.0 M TBAF-THF (2.5 ml, 2.5 mmol) and stirred at ambient temperature for 30 min. After diluting with CH₂Cl₂ (120 ml), the organic layer was washed with brine, dried over MgSO₄, filtered, and then evaporated. Obtained crude material was purified by silica gel column chromatography (1%TEA-CH₂Cl₂/MeOH, 100:0 to 94:6) yielding compound 8a (1.52 g, 94%); ³¹P NMR (202 MHz, CDCl₃) δ 19.0, 18.7; HRMS (ESI) calcd. for C₄₉H₄₈FN₄O₁₅PNa⁺ [M + Na]⁺m/z 1005.2730, found m/z 1005.2717; for¹H and ¹³C NMR analysis, part of diastereoisomers were purified by prep-TLC. Least polar isomer ¹H NMR (500 MHz, CD₃CN) δ 9.08 (br-s, 1H), 7.99–7.97, 7.76–7.73, 7.66 (d, 1H, J = 8.2 Hz), 7.58–7.56 (m, 2H), 7.52 (d, 1H, J = 8.2 Hz), 7.45–7.44 (m, 2H), 7.34–7.31 (m, 7H), 6.92–6.83 (m, 5H), 6.11 (ddd, 1H, J = 20.1, 17.2, 1.4 Hz), 5.88–5.83 (m, 3H), 5.31 (dd, 1H, J = 8.3, 2.2 Hz), 5.16 (dd, 1H, J_HF = 52.5 Hz, J_HH= 4.0 Hz), 5.02 (ddd, 1H, J = 8.0, 5.3, 5.3 Hz), 4.48–4.44 (m, 1H), 4.24–4.22 (m, 1H), 4.21–4.18 (m, 1H), 4.01 (dd, 1H, J = 4.7, 4.7 Hz), 3.94 (d, 1H, J = 7.7 Hz), 3.77 (s, 6H), 3.61 (d, J_PH= 11.3 Hz), 3.43–3.42 (m, 2H), 3.41 (s, 3H); ¹³C NMR (125 MHz, CD₃CN) δ 170.2, 163.7, 163.2, 159.9, 151.4, 150.2, 148.6 (d, J_CP= 5.5 Hz), 145.6, 142.8, 141.0, 136.6, 136.4, 136.3, 132.3, 131.2, 131.1, 130.5, 129.1, 129.0, 128.7, 128.1, 120.9 (d, J_CP = 218.9 Hz), 114.21, 114.20, 103.0, 94.2 (d, J_CF = 185.3 Hz), 91.9 (d, J_CF = 37.2 Hz), 88.1, 87.9, 82.9, 82.8, 82.6 (d, J_CF = 3.6 Hz), 82.5, 82.3,73.7 (d, J_CF = 16.3 Hz), 73.1, 73.0, 62.7, 59.1, 59.0, 56.0, 53.2 (d, J_CP = 5.4 Hz); ¹⁹F NMR (470 MHz, CD₃CN) δ –200.1 (ddd, J = 47.1, 21.2, 21.2 Hz); ³¹P NMR (202 MHz, CD₃CN) δ 18.2; most polar isomer ¹H NMR (500 MHz, CD₃CN) δ 9.22 (br-s, 1H), 7.98–7.96 (m, 2H), 7.76–7.74 (m, 1H), 7.68–7.66 (m, 1H), 7.57–7.54 (m, 2H), 7.48 (d, 1H, J = 8.2 Hz), 7.44–7.42 (m, 2H), 7.34–7.30 (m, 7H), 6.90–6.87 (m, 5H), 6.07 (ddd, 1H, J = 20.1, 17.3, 1.6 Hz), 5.88–5.87 (m, 1H), 5.83–5.79 (m, 1H), 5.33 (dd, 1H, J = 8.2, 2.1 Hz), 5.15 (dd, 1H, J_HF = 53.0 Hz, J_HH= 4.6 Hz), 5.00–4.96 (m, 1H), 4.36–4.32 (m, 1H), 4.20–4.09 (m, 2H), 4.05–4.03 (m, 1H), 3.76 (s, 6H), 3.69 (d, 3H, J_PH = 11.2 Hz), 3.49 (s, 3H), 3.43–3.41 (m, 2H), 3.34 (dd, 1H, J = 11.3, 3.5 Hz); ¹³C NMR (125 MHz, CD₃CN) δ 169.7, 163.4, 162.7, 159.40, 159.38, 150.9, 149.7, 148.6 (d, J_CP = 6.3 Hz), 145.2, 142.5, 140.6, 136.2, 136.02, 135.96, 131.89, 130.9, 130.72, 130.67, 130.1, 128.62, 128.59, 127.7, 119.0 (d, J_CP = 176.1 Hz), 113.80, 113.78, 102.59, 102.55, 93.6 (d, J_CF = 185.3 Hz), 91.5 (d, J_CF = 36.3 Hz), 87.8, 87.4, 82.4, 82.2 (d, J_CF = 6.4 Hz), 81.9, 81.8, 73.2 (d, J_CF = 16.4 Hz), 72.32, 72.27, 62.2, 58.6, 55.5, 53.1 (d, J_CP = 6.4 Hz); ¹⁹F NMR (470 MHz, CD₃CN) δ –200.2 (ddd, J = 52.1, 20.8, 20.8 Hz); ³¹P NMR (202 MHz, CD₃CN) δ 18.1.

5′-O-[Bis(4-methoxyphenyl)phenylmethyl]-2′-O-methyl-uridyl-methylphosphono-[3′(O)→5′(C)]-3-N-benzoyl-2′,5′-dideoxy-2′-fluoro-methylidene-uridine, 3′-[2-cyanoethyl-N,N-bis(1-methylethyl)- phosphoramidite] (9a)

Compound 8a (589.7 mg, 0.6 mmol) was rendered anhydrous by repeated co-evaporation with anhydrous CH₃CN and then dissolved into anhydrous CH₂Cl₂ (6.0 ml). To this solution N,N-diisopropylethylamine (0.31 ml, 1.8 mmol) and 2-cyanoethyl N,N-diisopropylchlorophosphoramidite (0.16 ml, 0.72 mmol) were added at 0°C. After stirring for 30 min at 0°C, the reaction mixture was diluted with excess CH₂Cl₂. The organic layer was repeatedly washed with aq. sat. NaHCO₃, dried over MgSO₄, filtered, and evaporated. The obtained crude material was purified by silica gel column chromatography (1%TEA-CH₂Cl₂/MeOH, from 100:0 to 96:4) yielding compound 9a as a white foam (0.57 g, 80%); ³¹P NMR (202 MHz, CDCl₃) δ 150.3, 151.2, 151.1, 151.0, 18.72, 18.65, 18.55, 18.3; HRMS (ESI) calcd. for C₅₈H₆₅FN₆NaO₁₆P₂⁺ [M + Na]⁺m/z 1205.3809, found m/z 1205.3799.

3-N-Benzoyl-2′-O-methyl-3′-O-(tert-butyldiphenylsilyl)-uridine (3b)

Anhydrous solution of compound 1b (1.35 g, 2.0 mmol) in pyridine (10 ml) was added N,N-diisopropylethylamine (0.63 ml, 3.6 mmol) and benzoyl chloride (0.35 ml, 3.0 mmol), and stirred for 3 h at rt. After diluting with excess CH₂Cl₂, the organic solution was washed with sat. NaHCO₃ aq. and brine. After drying over MgSO₄, filtered and evaporating, obtained crude material containing compound 2b was used for the next reaction without further purification. Obtained crude material containing compound 2b was added 3% trichloroacetic acid in CH₂Cl₂ (25 ml) and triethylsilane (1 ml, 6.26 mmol), and stirred for 1 h at rt. After the reaction mixture was diluted with CH₂Cl₂, the solution was washed with sat. NaHCO₃ aq. three times, dried over MgSO₄, filtered, then evaporated. Obtained crude material was purified by silica gel column chromatography (hexane/ethyl acetate, 4:1 to 1:4) yielding pure compound 3b (0.60 g, 63% in 2 steps); ¹H NMR (500 MHz, DMSO-d₆) δ 8.13 (d, 1H, J = 8.2 Hz), 7.95 (d, 2H, J = 7.3 Hz), 7.81 (t, 1H, J = 7.5 Hz), 7.69–7.68 (m, 2H), 7.64–7.59 (m, 4H), 7.49–7.42 (m, 6H), 5.93 (d, 1H, J = 4.6 Hz), 5.26 (t, 1H, J = 4.6 Hz), 4.36 (dd, 1H, J = 4.6, 4.6 Hz), 4.02–4.00 (m, 1H), 3.65–3.61 (m, 1H), 3.54 (dd, 1H, J = 4.6, 4.6 Hz), 3.09 (s, 3H), 1.03 (s, 9H); ¹³C NMR (126 Hz, DMSO-d₆) δ 169.8, 162.1, 149.5, 141.3, 136.1, 135.9, 135.8, 133.4, 133.2, 131.5, 130.7, 130.52, 130.48, 130.0, 128.4, 128.3, 102.1, 86.7, 85.6, 82.8, 79.7, 70.8, 60.2, 57.8, 27.2, 19.4; HRMS (ESI) m/z calcd for C₃₃H₃₇N₂O₇Si⁺ [M + H]⁺m/z 601.2365, found m/z 601.2372.

6′-Dibromo-(5′-deoxy-5′-methylidene)-2′-O-methyl-3′-O-(tert-butyldiphenylsilyl)-uridine (5b)

Anhydrous solution of compound 3b (0.3 g, 0.5 mmol) in CH₃CN (5 ml) was added IBX (0.35 g, 1.3 mmol) and stirred for 2 h at 85°C. After cooling the solution at 0°C, the precipitate was filtered off by celite-filtration. Obtained eluent containing compound 4b was evaporated, rendered anhydrous by repeated co-evaporation with anhydrous CH₃CN, and used for the next reaction without further purification. Separately prepared anhydrous solution of CBr₄ (0.33 g, 1.0 mmol) in CH₂Cl₂ (5.0 ml) was added triphenylphosphine (524.6 mg, 2.0 mmol) at 0°C in one portion and stirred at 0°C for 30 min. To this solution, compound 4b in anhydrous CH₂Cl₂ (1.5 ml) was added dropwise (10 min) at 0°C and stirred for 2 h at rt. The solution was then diluted with CH₂Cl₂ and washed with sat. NaHCO₃ aq. and brine. After the organic solution was dried over MgSO₄, filtered and evaporated, obtained crude material was purified by silica gel column chromatography (hexane/ethyl acetate, 9:1 to 4:6) yielding compound 5b (0.21 g, 56% in 2 steps); ¹H NMR (500 MHz, CDCl₃) δ 7.88 (d, 2H, J = 7.3 Hz), 7.70–7.62 (5H, m), 7.51–7.38 (m, 9H), 7.08 (d, 1H, J = 8.2 Hz), 6.26 (d, 1H, J = 8.6 Hz), 5.75 (d, 1H, J = 8.2 Hz), 5.68 (d, 1H, J = 0.8 Hz), 4.84 (dd, 1H, J = 8.6, 8.6 Hz), 3.86 (dd, 1H, J = 7.5, 5.0 Hz), 3.30 (s, 3H), 3.18 (br-s, 1H), 1.11 (s, 9H); ¹³C NMR (126 MHz, CDCl₃) δ 168.3, 161.7, 148.6, 138.9, 135.9, 135.8, 134.3, 132.6, 132.4, 131.2, 130.5, 130.4, 130.3, 129.2, 128.0, 127.9, 102.4, 97.5, 90.0, 82.44, 82.39, 74.4, 58.2, 26.7, 19.1; HRMS (ESI) m/z calcd for C₃₄H₃₅Br₂N₂O₆Si⁺ [M + H]⁺m/z 753.0626 [M + H] ⁺, found m/z 753.0637.

(E)/(Z)-6′-Bromo-(5′-deoxy-5′-methylidene)-2′-O-methyl-3′-O-(tert-butyldiphenylsilyl)-uridine (6b-E and 6b-Z)

Anhydrous solution of compound 5b (6.11 g, 8.1 mmol) in DMF (35 ml) was added dimethyl phosphite (2.97 ml, 34.0 mmol) and triethylamine (2.26 ml, 17.0 mmol) at 0°C, and then stirred overnight at rt. After the solution was diluted with ethyl acetate, the organic solution was washed with sat. NH₄Cl aq. and brine. Then the organic solution was dried over MgSO₄, filtered and evaporated, and obtained crude material was purified repeatedly by silica gel column chromatography (hexane/ethyl acetate, 9:1 to 1:1) until all pure isomeric compound were collected separately, giving compound 6b-E (3.0 g, 55%); ¹H NMR (500 MHz, CDCl₃) δ 7.89–7.87 (m, 2H), 7.70–7.62 (m, 5H), 7.51–7.39 (m, 8H), 7.10 (d, 1H, J = 8.3 Hz), 6.47 (dd, 1H, J = 13.6, 0.8 Hz), 6.01 (dd, 1H, J = 13.6, 7.9 Hz), 5.76–5.74 (m, 2H), 4.51 (dd, 1H, J = 7.8, 7.8 Hz), 7.36 (dd, 1H, = 7.8, 4.9 Hz), 3.34 (s, 3H), 3.17 (dd, 1H, J= 4.7, 1.2 Hz), 1.09 (s, 9H); ¹³C NMR (126 MHz, CDCl₃) δ 168.3, 161.7, 148.7, 138.4, 135.9, 135.8, 135.3, 133.8, 132.6, 132.4, 131.2, 130.5, 130.4, 130.3, 129.2, 128.0, 127.9, 112.1, 102.3, 88.9, 82.8, 82.6, 77.2, 74.2, 58.1, 26.8, 19.1; HRMS (ESI) m/z calcd for C₃₄H₃₆BrN₂O₆Si⁺ [M + H] ⁺m/z 675.1521 [M + H] ⁺, found m/z 675.1526 and 6b-Z (1.23 g, 22%); ¹H NMR (500 MHz, CDCl₃) δ 7.89–7.87 (m, 2H), 7.72–7.70 (m, 2H), 7.68–7.63 (m, 3H), 7.51–7.44 (m, 4H), 7.41–7.37 (m, 4H), 7.16 (d, 1H, J = 8.2 Hz), 6.53 (dd, 1H, J= 7.4, 0.6 Hz), 6.03 (dd, 1H, J = 8.5, 7.4 Hz), 5.75–5.73 (m, 2H), 5.12 (t, 1H, J = 8.1 Hz), 3.93 (dd, 1H, J = 6.9, 5.0 Hz), 3.32 (br-s, 1H), 3.26 (s, 3H), 1.10 (s, 9H); ¹³C NMR (126 MHz, CDCl₃) δ 168.3, 161.8, 148.7, 139.3, 135.91, 135.85, 135.22, 132.74, 132.71, 131.2, 130.8, 130.5, 130.23, 130.16, 129.2, 127.78, 127.75, 114.6, 102.2, 90.1, 82.4, 80.6, 77.2, 74.8, 58.1, 26.8, 19.2; C₃₄H₃₆BrN₂O₆Si⁺ [M + H] ⁺m/z 675.1521 [M + H] ⁺, found m/z 675.1529.

5′-O-[Bis(4-methoxyphenyl)phenylmethyl]-2′-deoxy-2′-fluoro-uridyl methylphosphono-[3′(O)→5′(C)] -3-N-benzoyl-5′-deoxy-5′-methylidene-3′-O-(tert-butyldiphenylsilyl)-2′-O-methyl-uridine (7b)

Anhydrous compound 6b-E (2.84 g, 4.20 mmol) and Pd(OAc)₂ (188.6 mg, 0.84 mmol) and [1,1′-Bis(diphenylphosphino)ferrocene]dichloropalladium (II) (931.4 mg, 1.68 mmol) were purged with argon, and then dissolved into anhydrous THF (50 ml). After adding propylene oxide (2.94 ml, 42.0 mmol), compound S2b (3.16 g, 5.04 mmol) was added in one portion and stirred at for 4 h at 70°C. After removing solvent under reduced pressure, the crude mixture was purified by silica gel column chromatography (hexane/ethyl acetate, 50:50 to 0:100) and obtained fractions containing compound 7b were further purified by silica gel column chromatography (1%TEA-CH₂Cl₂/MeOH, 100:0 to 95:5) yielding pure compound 7b as a mixture of diastereoisomers (3.3 g, 64%); HRMS (ESI) calcd. for C₆₅H₆₆FN₄NaO₁₅PSi⁺ [M + Na]⁺m/z 1243.3908 [M + Na]⁺, found m/z 1243.3912. For ¹H, ¹³C, ¹⁹F, and ³¹P NMR analysis, part of diastereoisomers were purified by preparative TLC. Least polar isomer ¹H NMR (500 MHz, CDCl₃) δ 9.12 (br-s, 1H), 7.89–7.85 (m, 3H), 7.77–7.59 (m, 5H), 7.49–7.24 (m, 17H), 7.15 (d, 1H, J = 8.2 Hz), 6.87 (d, 4H, J = 7.8 Hz), 6.84–6.77 (m, 1H), 6.04 (dd, 1H, J_HF = 15.7 Hz, J_HH = 0.9 Hz), 5.98 (ddd, 1H, J = 20.2, 19.1, 1.1 Hz), 5.94 (d, 1H, J = 3.1 Hz), 5.77 (d, 1H, J = 8.2 Hz), 5.25 (d, 1H, J = 8.1 Hz), 5.22–5.07 (m, 2H), 4.67–4.64 (m, 1H), 4.29 (dd, 1H, J = 7.7, 1.8 Hz), 3.91 (dd, 1H, J = 6.4, 2.5 Hz), 3.80 (s, 6H), 3.73 (dd, 1H, J = 11.3, 1.8 Hz), 3.52–3.49 (m, 4H), 3.22 (s, 3H), 3.11 (dd, 1H, J = 3.1, 3.1 Hz), 1.08 (s, 9H); ¹³C NMR (126 MHz, CDCl₃) δ 168.3, 162.6, 161.7, 158.85, 158.83, 149.8, 148.9, 148.8, 143.7, 139.4, 138.5, 135.8, 135.7, 135.2, 134.7, 132.5, 132.4, 131.3, 130.5, 130.4, 130.30, 130.28, 130.24, 129.2, 129.1, 128.3, 128.2, 128.1, 128.0, 127.9, 127.4, 125.3, 117.9 (d, J_CP= 192.5 Hz), 113.4, 102.74, 102.69, 91.8 (d, J_CF = 91.8 Hz), 88.2, 87.7 (d, J_CF = 33.6 Hz), 87.4, 82.5 (d, J_CP = 22.7 Hz), 82.1, 80.7 (d, J_CF = 8.1 Hz), 74.0, 70.8 (dd, J_CP = 4.5 Hz, J_CF = 15.4 Hz), 60.3, 58.0, 55.3, 52.4 (d, J_CP= 5.5 Hz), 26.8, 19.2; ¹⁹F NMR (470 MHz, CD₃CN) δ –199.53 (ddd, J = 52.0, 20.8, 20.8 Hz);³¹P NMR (202 MHz, CDCl₃) δ 19.3; most polar isomer ¹H NMR (500 MHz, CDCl₃) δ 8.9 (br-s, 1H), 7.89–7.87 (m, 2H), 7.83 (d, 1H, J = 8.1 Hz), 7.69–7.59 (m, 5H), 7.49–7.23 (m, 17H), 7.04 (d, 1H, J = 8.3 Hz), 6.85 (d, 2H, J = 3.2 Hz), 6.83 (d, 2H, J = 3.2 Hz), 6.82 (ddd, 1H, J = 23.1, 17.2, 4.3 Hz), 6.13 (dd, 1H, J_HF= 15.7 Hz, J_HH = 2.2 Hz), 5.89 (ddd, 1H, J = 20.0, 17.2, 1.7 Hz), 5.83 (d, 1H, J = 2.6 Hz), 5.73 (d, 1H, J = 8.3 Hz), 5.29–5.28 (m, 2H), 5.19–5.11 (m, 1H), 4.57–4.54 (m, 1H), 4.27–4.25 (m, 1H), 3.87 (dd, 1H, J = 6.9, 5.1 Hz), 3.78 (s, 3H), 3.77 (s, 3H), 3.73 (d, 3H, J_PH = 11.4 Hz), 3.61 (dd, 2H, J = 11.4, 2.1 Hz), 3.43 (dd, 1H, J = 11.3, 2.1 Hz), 3.21 (s, 3H), 3.17–3.15 (m, 1H), 1.08 (s, 9H); ¹³C NMR (125 MHz, CDCl₃) δ 168.2, 162.6, 161.5, 158.81, 158.78, 149.9, 148.7 (d, J_CP = 3.7 Hz), 144.0, 139.5, 138.9, 135.8, 135.7, 135.3, 134.81, 134.76, 132.45, 132.3, 131.2, 130.4, 130.3, 130.2, 130.1, 129.2, 128.08, 128.06, 128.02, 127.9, 127.3, 117.3 (d, J_CP = 192.5 Hz), 113.4, 102.71, 102.68, 92.7, 91.9 (d, J_CF = 195.3 Hz), 89.2, 87.43 (d, J_CF = 33.6 Hz), 87.37, 82.2 (d, J_CP = 21.8 Hz), 82.1, 80.9 (d, J_CF = 8.1 Hz), 74.3, 71.0 (dd, J_CF = 15.5 Hz, J_CP = 4.5 Hz), 60.6, 57.9, 55.22, 55.21, 52.8 (d, J_CP= 5.6 Hz), 26.8, 19.2; ¹⁹F NMR (470 MHz, CD₃CN) δ -199.43 (ddd, J = 52.1, 20.8, 20.8 Hz); ³¹P NMR (202 MHz, CDCl₃) δ 18.7.

5′-O-[Bis(4-methoxyphenyl)phenylmethyl]-2′-deoxy-2′-fluoro-uridyl methylphosphono-[3′(O)→5′(C)]-3-N-benzoyl-5′-deoxy-5′-methylidene-2′-O-methyl-uridine (8b)

Compound 7b (3.3 g, 2.70 mmol) in anhydrous THF (36.5 ml) was added 1.0 M TBAF-THF (4.05 ml, 4.05 mmol) and stirred at ambient temperature for 30 min. After diluting with CH₂Cl₂ (150 ml), the organic layer was washed with brine, dried over MgSO₄, filtered, and then evaporated. Obtained crude material was purified by silica gel column chromatography (1%TEA-CH₂Cl₂/MeOH, 100:0 to 92:8) yielding compound 8b (1.25 g, 47%); HRMS (ESI) calcd. for C₄₉H₄₈FN₄NaO₁₅P⁺ [M + Na]⁺m/z 1005.2730, found m/z 1005.2728; For ¹H, ¹³C, ¹⁹F, ³¹P NMR analysis, part of diastereoisomers were purified by preparative TLC. Least polar isomer ¹H NMR (500 MHz, CDCl₃) δ 9.42 (br-s, 1H), 7.96–7.94 (m, 2H), 7.87 (d, 1H, J = 8.3 Hz), 7.67–7.64 (m, 1H), 7.50 (t, 2H, J = 7.9 Hz), 7.38–7.24 (m, 10H), 7.09 (ddd, 1H, J = 23.8, 17.3, 3.7 Hz), 6.86 (d, 4 H, J = 8.8 Hz), 6.12 (ddd, 1H, J = 21.8, 17.3, 1.7 Hz), 6.06 (dd, 1H, J_HF = 16.3 Hz, J_HH = 1.0 Hz), 5.96 (s, 1H), 5.90 (d, 1H, J = 8.3 Hz), 5.32–5.17 (m, 3H), 4.50–4.48 (m, 1H), 4.34–4.32 (m, 1H), 3.95 (ddd, 1H, J = 14.1, 8.7, 5.5 Hz), 3.82–3.78 (m, 7H), 3.71 (dd, 1H, J = 11.4, 2.0), 3.58 (d, 3H, J_PH = 11.5 Hz), 3.55 (s, 3H), 3.51 (dd, 1H, J = 11.4, 2.5 Hz), 3.27 (d, 1H, J = 9.0 Hz); ¹³C NMR (125 MHz, CDCl₃) δ 168.5, 162.8, 161.9, 158.8, 150.3, 149.0 (d, J_CP = 7.3 Hz), 143.8, 139.4, 138.7, 135.3, 134.9, 134.7, 131.2, 130.6, 130.2, 129.3, 128.3, 128.1, 127.4, 116.8, (d, J_CP = 192.6 Hz), 113.4, 102.9, 102.8, 91.7 (d, J_CF = 191.7 Hz), 88.7, 87.9 (d, J_CF= 33.6 Hz), 87.4, 83.1, 82.4 (d, J_CP = 22.7 Hz), 80.9 (d, J_CF = 8.2 Hz), 72.8, 71.0 (dd, J_CF = 14.5, J_CP = 3.6 Hz), 60.4, 58.9, 55.3, 52.5 (d, J_CP = 6.4 Hz); ¹⁹F NMR (470 MHz, CD₃CN) δ -199.53 (ddd, J = 52.0, 20.8, 20.8 Hz); ³¹P NMR (202 MHz, CDCl₃) δ 19.7; most polar isomer ¹H NMR (500 MHz, CDCl₃) δ 8.63 (br-s, 1H), 7.95–7.93 (m, 2H), 7.81 (d, 1H, J = 8.3 Hz), 7.69–7.66 (m, 1H), 7.53–7.49 (m, 2H), 7.37–7.24 (m, 10H), 6.97 (ddd, 1H, J = 23.4, 17.4, 4.0 Hz), 6.86–6.83 (m, 4H), 6.07 (dd, 1H, J_HF = 16.5 Hz, J_HH = 1.1 Hz), 6.00 (ddd, 1H, J = 19.6, 17.4, 1.7 Hz), 5.86–5.85 (m, 2H), 5.32 (d, 1H, J = 8.3 Hz), 5.28 Hz (ddd, 1H, J_HF = 51.5 Hz, J_HH = 4.2, 1.5 Hz), 5.15–5.08 (m, 1H), 4.31–4.29 (m, 1H), 4.20–4.17 (m, 1H), 3.91–3.87 (m, 1H), 3.84–3.83 (m, 1H), 3.80–3.77 (m, 9H), 3.65 (dd, 1H, J = 11.5, 2.0 Hz), 3.58 (s, 3H), 3.40 (dd, 1H, J = 11.5, 2.6 Hz), 2.90 (d, 1H, J = 9.3 Hz); ¹³C NMR (125 MHz, CDCl₃) δ 168.3, 162.6, 161.6, 158.8, 158.7, 149.8, 148.9, 148.8 (d, J_CP = 148.8 Hz), 144.1, 139.8, 138.8, 135.4, 135.2, 134.9, 131.2, 130.5, 130.3, 130.1, 129.3, 129.1, 128.1, 127.9, 127.8, 127.3, 117.0 (d, J_CP = 192.5 Hz), 113.4, 113.2, 102.9, 102.7, 92.0 (d, J_CF = 92.0 Hz), 89.3, 88.3 (d, J_CF = 33.6 Hz), 87.3, 82.8, 82.4 (d, J_CP = 22.7 Hz), 80.7 (d, J_CF = 8.2 Hz), 72.7, 70.9 (dd, J_CF = 16.4, J_CP = 5.5 Hz), 60.7, 58.9, 55.3, 52.8 (d, J_CP = 52.8 Hz); ¹⁹F NMR (470 MHz, CD₃CN) δ -199.43 (ddd, J = 52.1, 20.8, 20.8 Hz); ³¹P NMR (202 MHz, CDCl₃) δ 19.0.

5′-O-[Bis(4-methoxyphenyl)phenylmethyl]-2′-deoxy-2′-fluoro-uridyl methylphosphono-[3′(O)→5′(C)]-3-N-benzoyl-5′-deoxy-5′-methylidene-2′-O-methyl-uridine, 3′-[2-cyanoethyl-N,N-bis(1-methylethyl)- phosphoramidite] (9b)

Compound 8b (393.2 mg, 0.4 mmol) was rendered anhydrous by repeated co-evaporation with anhydrous CH₃CN and then dissolved into anhydrous CH₂Cl₂ (4.0 ml). To this solution N,N-diisopropylethylamine (0.21 ml, 1.2 mmol) and 2-cyanoethyl N,N-diisopropylchlorophosphoramidite (0.11 ml, 0.48 mmol) were added at 0°C. After stirring for 30 min at 0°C, the reaction mixture was diluted with excess CH₂Cl₂. The organic layer was repeatedly washed with sat. NaHCO₃ aq., dried over MgSO₄, filtered, and evaporated. The obtained crude material was purified by silica gel column chromatography (1%TEA-CH₂Cl₂/MeOH, from 100:0 to 96:4) yielding compound 9b as a white foam (319.6 mg, 68%); ³¹P NMR (202 MHz, CDCl₃) δ 150.7, 150.4, 150.3, 19.9, 19.5, 19.4, 18.8. HRMS (ESI) calcd. for C₅₈H₆₅FN₆NaO₁₆P₂⁺ [M + Na]⁺m/z 1205.3809, found m/z 1205.3799.

5′-O-[Bis(4-methoxyphenyl)phenylmethyl]-2-N-isobutylyl-2′-O-methyl-guanyl methylphosphono-[3′(O)→5′(C)]-5′-deoxy-5′-methylidene-3′-O-(tert-butyldiphenylsilyl)-3-N-benzoyl-2′-deoxy-2′-fluoro-uridine (7c)

Anhydrous compound 6a-E (2.19 g, 3.31 mmol) and Pd(OAc)₂ (148.2 mg, 0.66 mmol) and [1,1′-Bis(diphenylphosphino)ferrocene]dichloropalladium (II) (731.8 mg, 1.32 mmol) were purged with argon, and then dissolved into anhydrous THF (23.0 ml). After adding propylene oxide (2.32 ml. 33.1 mmol), compound S2c (2.72 g, 3.64 mmol) in anhydrous THF (10.0 ml) was added and stirred at 70°C overnight. After removing solvent under reduced pressure, the crude mixture was purified by silica gel column chromatography (ethyl acetate/MeOH, 100:0 to 96:4) yielding pure compound 7c (1.33g, 30%); HRMS (ESI) calcd. for C₇₀H₇₄FN₇O₁₅PSi⁺ [M + H]⁺m/z 1330.4728, found m/z 1330.4744; For ¹H, ¹³C, ³¹P, ¹⁹F NMR analysis, part of diastereoisomers were separated by preparative TLC. Least polar isomer ¹H NMR (500 MHz, CD₃CN) δ 11.9 (br-s, 1H), 10.2 (br-s, 1H), 7.94–7.90 (m, 3H), 7.76–7.73 (m, 1H), 7.67–7.54 (m, 6H), 7.45–7.20 (m, 16H), 6.85–6.76 (m, 4H), 6.17 (ddd, 1H, J = 20.2, 17.3, 1.1 Hz), 5.94 (d, 1H, J = 3.6 Hz), 5.87–5.84 (m, 1H), 5.74 (d, 1H, J = 8.3 Hz), 5.72 (d, J_HF = 22.0 Hz), 4.75 (dd, 1H, J_HF = 52.6 Hz, J_HH = 4.6 Hz), 4.66–4.63 (m, 2H), 4.40–4.33 (m, 2H), 3.77 (s, 3H), 3.76 (s, 3H), 3.54 (d, 3H, J_PH = 11.3 Hz), 3.40 (dd, 1H, J = 11.1, 2.4 Hz), 3.25 (dd, 1H, J = 8.1, 4.1 Hz), 3.22 (s, 3H), 2.63–2.57 (m, 1H), 1.13 (d, 3H, J = 2.6 Hz), 1.12 (d, 3H, J = 2.6 Hz), 1.07 (s, 9H); ¹³C NMR (125 MHz, CD₃CN) δ 180.4, 169.5, 162.7, 159.23, 159.21, 155.9, 149.4, 148.6, 148.2 (d, J_CP = 5.4 Hz), 145.3, 143.5, 139.1, 136.3, 136.24, 136.22, 136.19, 136.1, 133.0, 132.8, 131.8, 130.9, 130.8, 130.5, 130.4, 130.0, 129.5, 128.8, 128.54, 128.46, 128.42, 128.3, 127.5, 125.8, 120.1 (d, J_CP = 189.8 Hz), 113.5, 102.4, 93.3 (d, J_CF = 37.2 Hz), 92.6 (d, J_CF = 188.0 Hz), 86.6, 86.3, 82.7, 82.5, 81.9 (d, J_CF = 5.4 Hz), 81.8 (d, J_CP = 1.9 Hz), 74.8 (d, J_CF = 16.3 Hz), 73.2 (d, J_CP = 4.5 Hz), 62.6, 58.6, 55.5, 53.0 (d, J_CP = 5.4 Hz), 36.5, 26.7, 19.3, 18.8, 18.7; ³¹P NMR (202 MHz, CD₃CN) δ 18.6; ¹⁹F NMR (470 MHz, CD₃CN) δ -196.54, -196.59, -196.65, -196.68; most polar isomer ¹H NMR (500 MHz, CD₃CN) δ 11.9 (br-s, 1H), 10.0 (br-s, 1H), 7.94–8.91 (m, 3H), 7.99–7.76 (m, 1H), 7.68–7.55 (m, 6H), 7.47–7.43 (m, 2H), 7.40–7.33 (m, 8H), 7.24–7.18 (m, 8H), 6.79–6.76 (m, 5H), 6.10–6.13 (m, 1H), 5.97 (d, 1H, J = 3.8 Hz), 5.93 (d, 1H, J = 3.2 Hz), 5.78–5.73 (m, 2H), 4.76–4.63 (m, 2H), 4.53 (br-s, 1H), 4.31–4.24 (m, 2H), 3.76 (s, 6H), 3.69 (d, J_PH = 11.5 Hz), 3.51–3.49 (m, 4H), 3.37–3.35 (m, 1H), 3.25–3.21 (m, 1H), 2.61–2.57 (m, 1H), 1.16–1.14 (m, 6H), 1.07–1.06 (m, 9H); ¹³C NMR (125 MHz, CD₃CN) δ 180.4, 169.6, 162.6, 159.22, 159.21, 155.9, 149.5, 148.5, 147.7 (d, J_CP = 5.4 Hz), 145.4, 143.1, 139.2, 136.4, 136.3, 136.2, 136.14, 136.12, 133.0, 132.8, 131.8, 130.93, 130.92, 130.90, 130.8, 130.54, 130.48,130.0, 128.6, 128.5, 128.3, 127.5, 121.0 (d, J_CP = 215.3 Hz), 113.5, 102.5, 92.6 (d, J_CF = 189.0 Hz), 91.8, 86.7, 86.5, 82.5, 82.3, 82.1 (d, J_CF = 5.5 Hz), 81.7 (d, J_CP= 1.2 Hz), 75.0 (d, J_CF = 14.9 Hz), 73.2 (d, J_CP = 4.6 Hz), 62.5, 58.6, 55.5, 53.40. 53.35, 36.4, 26.7, 19.3, 18.8, 18.7; ³¹P NMR (202 MHz, CD₃CN) δ 18.6; ¹⁹F NMR (470 MHz, CD₃CN) δ -197.4 (ddd, J = 51.7, 20.8, 20.8 Hz).

5′-O-[Bis(4-methoxyphenyl)phenylmethyl]-2-N-isobutylyl-2′-O-methyl-guanyl methylphosphono-[3′(O)→5′(C)]-3-N-benzoyl-2′,5′-dideoxy-5′-methylidene-2′-fluoro-uridine (8c)

Compound 7c (0.8 g, 0.6 mmol) in anhydrous THF (8.1 ml) was added 1.0 M TBAF-THF (0.9 ml, 0.9 mmol) and stirred at ambient temperature for 30 min. After diluting with CH₂Cl₂ (150 ml), the organic layer was washed with brine, dried over MgSO₄, filtered, and then evaporated. Obtained crude material was purified by silica gel column chromatography (1%TEA-CH₂Cl₂/MeOH, 100:0 to 92:8) yielding compound 8c (0.56 g, 85%); HRMS (ESI) calcd. for C₅₄H₅₆FN₇O₁₅P⁺ [M + H]⁺m/z 1092.3551, found m/z 1092.3558; ³¹P NMR (202 MHz, CD₃CN) δ 19.1, 19.0. For ¹H, ¹³C and ¹⁹F NMR analysis, part of diastereoisomers were purified by preparative TLC, and only least polar isomer was collected as an isomer pure fraction. Least polar isomer ¹H NMR (500 MHz, CD₃CN) δ 12.0 (br-s, 1H), 10.2 (br-s, 1H), 7.98–7.96 (m, 2H), 7.86 (s, 1H), 7.75–7.72 (m, 1H), 7.56–7.51 (m, 3H), 7.34–7.33 (m, 2H), 7.23–7.18 (m, 7H), 6.15 (ddd, 1H, J = 20.2, 17.2, 1.4 Hz), 5.93 (d, 1H, J = 3.9 Hz), 5.86–5.81 (m, 2H), 5.67 (ddd, 1H, J = 8.4, 5.3, 5.3 Hz), 5.18 (dd, 1H, J_HF = 52.8 Hz, J_HH = 4.7 Hz), 4.66 (dd, 1H, J = 4.5 Hz), 4.50–4.48 (m, 1H), 4.32–4.28 (m, 2H), 4.24–4.15 (m, 1H), 3.74 (s, 6H), 3.57 (3H, d, J_PH = 11.5 Hz), 3.39–3.36 (m, 4H), 3.30 (dd, 1H, J = 11.0, 4.4 Hz), 2.59 (sep, 1H, J = 6.9 Hz), 1.14 (d, 3H, J = 6.8 Hz), 1.10 (d, 3H, J = 6.8 Hz); ¹³C NMR (125 MHz, CD₃CN) δ 181.1, 170.2, 163.3, 159.8, 159.7, 156.5, 150.1, 149.4, 149.13, 149.06 (d, J_CP = 6.3 Hz), 145.8, 143.2, 139.7, 136.8, 136.7, 132.3, 131.4, 131.1, 131.0, 130.5, 129.1, 129.0, 128.0, 122.2, 120.1, 114.05, 114.04, 103.0, 94.2 (d, J_CF = 185.3 Hz), 92.4 (d, J_CF = 36.3 Hz), 87.2, 82.8 (d, J_CF = 6.4 Hz), 82.5, 82.3, (dd, J_CP = 2.7, 2.7 Hz), 74.0 (d, J_CP = 4.6 Hz), 73.7 (d, J_CF = 16.7 Hz), 63.4, 59.2, 55.98, 55.96, 53.5 (d, J_CP = 6.3 Hz), 47.3, 36.9, 19.3, 19.2; ¹⁹F NMR (470 MHz, CD₃CN) δ –199.9 (ddd, J = 53.9, 20.9, 20.9 Hz).

5′-O-[Bis(4-methoxyphenyl)phenylmethyl]-2-N-isobutyryl-2′-O-methyl-guanyl-methylphosphono-[3′(O)→5′(C)]-3-N-benzoyl-2′,5′-dideoxy-2′-fluoro-5′-methylidene-uridine, 3′-[2-cyanoethyl N,N-bis(1-methylethyl)- phosphoramidite] (9c)

Compound 8c (557 mg, 0.51 mmol) was rendered anhydrous by repeated co-evaporation with anhydrous CH₃CN and then dissolved into anhydrous CH₂Cl₂ (5.1 ml). To this solution N,N-diisopropylethylamine (0.34 ml, 1.94 mmol) and 2-cyanoethyl N,N-diisopropylchlorophosphoramidite (0.17 ml, 0.77 mmol) were added at 0°C. After stirring for 30 min at 0°C, the reaction mixture was diluted with excess CH₂Cl₂. The organic layer was repeatedly washed with sat. NaHCO₃ aq., dried over MgSO₄, filtered, and evaporated. The obtained crude material was purified by silica gel column chromatography (1% TEA-CH₂Cl₂/Acetone, from 100:0 to 70:30). Obtained compound 9c with a slight amount of reagent residues was dissolved in Et₂O-ethyl acetate (1:1, v/v), repeatedly washed by sat. NaHCO₃ aq., and then washed by brine. Organic layer was dried over MgSO₄, filtered, and evaporated, yielding compound 9c as a white foam (367 mg, 57%); ³¹P NMR (202 MHz, CDCl₃) δ 151.6, 151.5, 151.10, 151.05, 19.5, 19.2, 18.4, 18.3. HRMS (ESI) calcd. for C₆₃H₇₃FN₉O₁₆P₂⁺ [M + H]⁺m/z 1292.4629, found m/z 1292.4643.

Synthesis of inter-nucleotide (E)-vinylphosphonate modified RNA oligonucleotides

Synthesis of all RNA oligonucleotides with or without ⁱE-VP linkages were performed with a MerMade 12 automated RNA synthesizer (BioAutomation) on a 1 μmol UnyLinker support (ChemGenes, co.). All 2′-modified (2′-OMe and 2′-F) phosphoramidites (ChemGenes) and ⁱE-VP-linked dimer phosphoramidites were prepared at a concentration of 0.1 M in anhydrous acetonitrile. Synthesis was conducted on a standard 1.0 μmol scale RNA phosphoramidite synthesis cycle, which consists of (i) detritylation, (ii) coupling, (iii) capping, and (iv) iodine oxidation to phosphate by 0.02 M I₂ in THF-pyridine-H₂O (7:2:1, v/v/v), or sulfurization by 0.1 M DDTT in pyridine:CH₃CN (9:1, v/v). Coupling of phosphoramidites was conducted with a standard protocol for 2-cyanoethyl phosphoramidite using BTT as an activator. For 5′-terminal phosphorylation, bis(2-cyanoethyl)-N,N-diisopropyl phosphoramidite (ChemGenes) was used. For the 5′-termianl (E)-VP modification, 5′-vinyl tetraphosphonate-(pivaloyloxymethyl)-2′-O-methyl-uridine 3′-CE phosphoramidite (ChemGenes) was used. For 3′-cholesterol modified RNA oligonucleotide synthesis, cholesterol 3′-lcaa CPG 500Å (ChemGenes) was used. For 3′-GalNAc conjugated passenger strand synthesis, GalNAc-TEG CPG 500Å (Primetech, ALC.) was used. In the case of ⁱE-VP modified RNA, RNA on solid support was first treated with TMSBr-pyridine-CH₂Cl₂ (3:1:18, v/v/v) for 1 h at ambient temperature in RNA synthesis column. Solid support was washed by water (1 ml × 3), CH₃CN (1 ml × 3) and CH₂Cl₂ (1 ml × 3) by flowing solutions through the synthesis column, and then dried under vacuum. After transferring the solid support to screw-capped sample tube, base treatment by NH₄OH-EtOH (3:1, v/v) for 48 h at 26°C was conducted. For non-ⁱE-VP modified oligonucleotides, base treatment was conducted after chain elongation of oligonucleotides. For 5′-termianl (E)-VP modified guide strands, base treatment for the deprotection was conducted by NH₄OH (3% diethylamine) at 35°C for 20 h. Crude oligonucleotides without cholesterol conjugates were purified by standard anion exchange HPLC, whereas oligonucleotides with cholesterol-conjugate were purified by reversed-phase HPLC. All obtained purified oligonucleotides were desalted by Sephadex G-25 (GE Healthcare) and characterized by electrospray ionization mass spectrometry (ESI-MS) analysis.

General procedures for nuclease digestion tests

Oligonucleotides were incubated in buffers with or without Terminator^TM 5′-P-dependent 5′-exonuclease, 5′-P-independent 5′-exonuclease of phosphodiesterase II from bovine spleen (BSP), or 3′-exonuclease of snake venom phosphodiesterase I (SVPD). An aliquot of the reaction mixtures was taken at each time point and quenched by adding 95% formamide containing 20 mM EDTA, subsequently frozen by liquid nitrogen, and then kept at –80°C. Samples were heated at 95°C for 1 min, iced, and then immediately applied to the 20% formamide 20% polyacrylamide gel containing 7 M urea. After gel electrophoresis at 500 V for 2 h, gels were stained with SYBR^® Gold Nucleic Acid Gel Stain (Thermo Fisher Scientific), and visualized by Typhoon FLA 9000 (GE Healthcare).

Terminator 5′-P dependent 5′-exonuclease stability test

2.5 μM oligonucleotide (50 pmol) were incubated in RNase-free water, or with 0.17 U or 3.3U of Terminator™ (EpiCentre) exonuclease at 37°C in buffer A (EpiCentre, provided with Terminator™ enzyme).

BSP stability test

10 μM oligonucleotide was incubated at 37°C in RNase-free water or 30 mM NaOAc (pH 6.0) buffer containing 0.25 U/ml BSP.

SVPD stability test

17.5 μM oligonucleotide (50 pmol) was incubated in RNase-free water or in a 10 mM Tris–HCl buffer (pH 8.0) containing 2.0 mM MgCl₂ and 4 mU/ml (or 10mU/ml) of SVPD (Sigma-Aldrich).

Primer extension assays

36-nt DNA adapter with a 5′ phosphate (Integrated DNA Technologies, Inc.) was adenylated using the 5′ DNA Adenylation Kit (NEB Cat# E2610) according to manufacturer's protocol. This adenylated DNA was then ligated to sequence-matched oligonucleotides with or without ⁱE-VP modification (^VPG6 for ON5 and ^CtrlG4 for ON7, or ^VPG10 for ON6 and ^CtrlG1 for ON8, respectively) using RNA Ligase 2, truncated K227Q (NEB Cat# M0351) according to manufacturer's protocol. Ligation products were run on a 15% Polyacrylamide gel containing 7 M urea, gel extracted into 300 mM NaCl, and ethanol precipitated. The marker was prepared in the same way. The ligated marker was then hydrolyzed in 50 mM NaHCO₃, pH 9 at 90°C for 20 min. The reaction was then put on ice and neutralized by adding an equal volume of 50 mM Tris–HCl pH 7. Reverse transcription reactions were performed with AMV Reverse Transcriptase (NEB Cat# M0277) and SuperScript II^TM (ThermoFisher Cat# 11904018) according to the manufacturer's protocol, using 10 pmol of a 5′-FAM labeled primer (Integrated DNA Technologies, Inc.), 5′-FAM-d-[CCTTGGCACCCGAGAATTCCA]-3′, complementary to the DNA adapter. These reactions were ethanol precipitated, resuspended in a 47.5% Formamide 10 mM EDTA loading dye, heated to 95°C for 2 min and put on ice. All reactions were then run on a 15% polyacrylamide gel containing 7 M urea and fluorescence was imaged on a Typhoon FLA 7000 (GE Healthcare).

Thermo stability assay

1 μM guide strand and 1 μM complementary sense strand were annealed in a 10 mM sodium phosphate buffer (pH 7.2) containing 100 mM NaCl and 0.1 mM EDTA by heating at 95°C for 1 min and cooled down gradually to room temperature. T_m measurement was performed with temperature controller. Both the heating and cooling curves were measured over a temperature range from 20 to 95°C at 1.0°C/min for three times.

In vitro screen

siRNA passive delivery

Cells were plated in Dulbecco's modified Eagle's medium containing 6% FBS at 8,000 cells per well in 96-well cell culture plates. siRNAs were diluted to twice the final concentration in OptiMEM (Carlsbad, CA; 31985–088), and 50 μl diluted siRNAs were added to 50 μl of cells, resulting in 3% FBS final. Cells were incubated for 72 h at 37°C and 5% CO₂. The maximal dose in the in vitro dose response assays was 1.5 μM compound.

Quantitative analysis of target mRNA

mRNA was quantified from cells using the QuantiGene 2.0 assay kit (Affymetrix, QS0011). Cells were lysed in 250 μl diluted lysis mixture composed of one part lysis mixture (Affymetrix, 13228), two parts H₂O and 0.167 μg/μl proteinase K (Affymetrix, QS0103) for 30 min at 55°C. Cell lysates were mixed thoroughly, and 40 μl of each lysate was added per well of a capture plate with 40 μl diluted lysis mixture without proteinase K and 20 μl diluted probe set. Probe sets for human HTT and Hypoxanthine Phosphoribosyltransferase (HPRT) (Affymetrix; #SA-50339, SA-10030) were diluted and used according to the manufacturer's recommended protocol. Datasets were normalized to HPRT.

Creating bar graph

Data were analyzed using GraphPad Prism 7 software (GraphPad Software, Inc., San Diego, CA). Concentration-dependent IC₅₀ curves were fitted using a log (inhibitor) versus response -variable slope (four parameters). For each cell treatment plate, the level of knockdown at each dose was normalized to the mean of the control (untreated) group. The lower limit of the curve was set to less than 5, and the upper limit of the curve was set to greater than 95. To create the bar graph, the percent difference was calculated by subtracting the IC₅₀ value for each compound from the IC₅₀ value for each corresponding control compound, dividing by the IC₅₀ value for the control compound, and multiplying by 100. If the percent difference was less than -500%, the percent difference was artificially set to -500%. The lower limit of the graph was cut at -300%.

Reporter plasmid preparation

Two plasmids were designed containing a 40mer sequence flanking the SNP rs362273 of the human huntingtin (htt) gene, in a psiCheck-2 vector (Promega cat.# C8029). Each plasmid is identical except for the base at the SNP site. PsiCheck-2 construct made via restriction cloning with XHO1 and NotI sites. Constructs were transformed into OneShot Top10 chemically competent cells. Successfully transformed cells were identified by ampicillin selection. Plasmids subsequently isolated from cells by Qiagen HiSpeed Plasmid Maxi Kit (ref. #12663).

Cell treatment and reporter assay

HeLa cells grown and maintained in Gibco DMEM (ref.# 11965-092) with 1% p/s and 10% heat inactivated FBS. Three days prior to treatment, two 10 cm² dishes were plated with 2 × 10⁶ HeLa cells. The following day, DMEM is replaced with Gibco OptiMEM (ref. #31985–070) and 6 μg of reporter plasmid is added to cells using Invitrogen Lipofectamine 3000 (ref. #L3000-015), following the manufacturer's protocol. Cells are left in OptiMEM/lipofectamine overnight to allow for maximum reporter plasmid transfection.

The following day, siRNA was diluted in Opti-MEM and added to 96-well white wall clear bottom tissue culture plate, in triplicate, for each reporter plasmid for passive uptake. For a dose response curve, an 11 point 1:1 dilution series was performed with final siRNA treatment concentrations of 1.5, 0.7500, 0.375, 0.1875, 0.0938, 0.0469, 0.0234, 0.0117, 0.00585, 0.002925 and 0.0014625 μM. HeLa cells transfected with reporter plasmids the prior night were resuspended in DMEM with 6% heat inactivated FBS (no p/s) at 0.15 × 10⁶ cells/ml and added to plate containing siRNA.

For lipid-mediated uptake dose response curves, a seven point 1:2 dilution series was performed for final siRNA treatment concentrations of 5, 1.67, 0.55, 0.185, 0.0617, 0.0206 and 0.0069 nM. Hela cells transfected with reporter plasmids the prior night were resuspended in DMEM with 10% FBS (no p/s) at 0.0833 × 10⁶ cells/ml and 90 μl of the cell suspension was added to 96-well plates containing 10 μl siRNA formulated with RNAiMax in OptiMEM per well.

Cells were lysed after 72 h of treatment (100% confluency) with 1x Passive Lysis Buffer from Dual-Luciferase Assay System Pack (Promega ref. #E1960). Following lysis, luminescence was read after addition of 50 μl Luciferase Assay Reagent II (Promega ref. #E1960), then read a second time after addition of 50 μl/well of Stop and Glow reagent (Promega ref. #E1960). Absorbances normalized to untreated controls and graphed on a log scale.

Evaluation of compounds efficacy and discriminative power in vivo in BAC-HD mice

BAC HD heterozygous female mice were purchased from The Jackson Laboratory (Tg(HTT*97Q)LXwy/JChdi). All animals were maintained and used according to the Institutional Animal Care and Use Committee guidelines of University of Massachusetts Medical School (docket #20210018). Mice were housed and maintained with a maximum of five to a cage until they aged to necessary time point. Mice were injected subcutaneously with 10 mg/kg of ^CtrlD18 and ^VPD24 at ∼12 weeks of age. The BAC HD heterozygous mice were euthanized two weeks after subcutaneous GalNAc conjugated compounds injections. At this point mice were deeply anesthetized with tribromoethanol, and liver, kidneys, and spleens were collected fresh. Half of the tissues were flash frozen on dry ice and stored at -80°C for protein analysis and the others were placed in RNAlater at 4°C for mRNA analysis.

Automatic western detection of mutant and wt huntingtin

For analysis of HTT protein expression in mouse livers WES by ProteinSimple was used. 2 mm-size tissue punches were homogenized at 75 μl of buffer (10 mM HEPES pH 7.2, 250 mM sucrose, 1 mM EDTA buffer with protease and phosphatase inhibitors) and stored at −80°C. Protein amount was determined using a Bradford Assay. Samples were diluted in 0.1 × sample buffer (ProteinSimple) to ∼0.2 μg/μl. Anti-HTT antibody (36) was diluted 1:50 in antibody diluent (ProteinSimple). This antibody detects both mutant and wt huntingtin protein, which are separated by weight due to the expansion of the CAG repeat in mutant (∼103 copies) versus wt (∼7 copies) proteins. The anti-vinculin (Sigma, V9131) antibody was diluted 1:3000 in diluent and used as a normalization control. Assay was performed as described by the manufacture instructions using the 66–440 kDa plate (SM-W008).

RESULTS

Synthesis of ⁱE-VP-dimer phosphoramidites

The three features defining the utility of synthetic oligonucleotide modifications are stabilization, efficacy, and specificity. Sugar and backbone modifications need to be used in combination to achieve these optimal properties. Therefore, we set out to optimize the synthetic pathways enabling oligonucleotide backbone ⁱE-VP incorporation in the context of the two most commonly-used sugar modifications: 2′-OMe and 2′-F.

Dimer nucleotide phosphoramidite synthesis of 2′-OMe-Ur-(ⁱE-VP)-2′-F-Ur (9a), 2′-F-Ur-(ⁱE-VP)-2′-OMe-Ur (9b), and 2′-OMe-G-(ⁱE-VP)-2′-F-Ur (9c) was carried out using the following synthetic approach (Scheme 1). Synthesis began with commercially-available 5′-O-DMTr protected nucleosides (purchased from Hongene Biotechnology Co., Ltd.), followed by protection of 3′-OH with TBDPS and 3-N with benzoyl (Bz) groups, and eventual 5′-O-DMTr group deprotection. After oxidation of 5′-OH by 2-iodoxybenzoic acid (IBX) converting to aldehyde (4a and 4b), Wittig olefination was carried out to obtain dibromo olefin compounds (5a and 5b) (38). Use of a 3-N-Bz protecting group was essential for the efficiency of this Wittig olefination. Without this protective group, the reaction mixture yielded a highly polar compound that could not be converted to target material even after prolonged reaction time and heating (data not shown). Our result differs from a previous report demonstrating successful synthesis of a DNA variant, ⁱE-VP-containing dimer (dT-ⁱE-VP-dT), without nucleobase protection (38). This difference is likely due to our use of 2′-OMe and 2′-F sugar modifications, both of which are essential for generating functionally-active siRNAs (39,40). Use of published protocols (no nucleobase protection) for the synthesis of 2′-OMe- and 2′-F-containing ⁱE-VP dimers resulted in extremely low yields, likely due to the C3′-endo sugar conformation of 2′-OMe and 2′-F nucleosides, of which the 5′-aldehyde residue may orient proximal to an unprotected uracil base, inducing nucleobase-involving side reactions.

Scheme 1. — Synthesis of ⁱE-VP-linked dimer phosphoramidites (9a, 9b, and 9c). *Reagents and conditions*: (i) TBDPSCl, Imidazole/DMF, rt, overnight, 1a: 99%, 1b: 95%; (ii) BzCl, DIPEA/CH₂Cl₂, rt, 4 h, 2a: 78%, 2b: N.D.^a; (iii) 3% trichloroacetic acid, triethylsilane/CH₂Cl_2, rt, 1 h, 3a: 91%, 3b; 63% (2 steps); (iv) IBX/CH₃CN, 85°C, 2 h; (v) CBr₄, PPh₃/CH₂Cl₂, 0°C then rt, 2 h, 5a: 52% in 2 steps, 5b: 56% in 2 steps; (vi) dimethylphosphite, triethylamine/DMF, rt, overnight, **6a-E**: 52%, **6b-E**: 55%, **6a-Z**: 15%, **6b-Z**: 22%; (vii) 5′-O-DMTr-uridine-3′-H-phosphonate methyl ester (2′-OMe: **S2a** or 2′-F: **S2b**) or 5′-O-DMTr-2-N-(*iso*-butyryl)- 2′-OMe-guanosine-3′-H-phosphonate methyl ester (**S2c)**, Pd(OAc)₂, dppf, propylene oxide/THF, 70°C or 75°C, 4 h-overnight, 7a: 57%, 7b: 64%, 7c: 30%; (viii) 0.1 M TBAF/THF, rt, 0.5 h, 8a: 94%, 8b: 47%, 8c: 85%; (IX) 2-cyanoethyl N,N-diisopropylchlorophosphoramidite, DIPEA/CH₂Cl₂, 0°C then rt, 0.5 h, 9a: 80%, 9b: 68%, 9c: 57%; ^aNot determined (see materials and methods).

Hirao reduction of the dibromo compound produced (E)-bromo olefin nucleoside (6a-E and 6b-E) as a major product and (Z)-vinyl olefin nucleoside (6a-Z and 6b-Z) as a minor product (38,41). Purification of both stereo-isomers was carried out by repeated silica gel column chromatography (Supplementary Figures S1, S2). The palladium-mediated cross-coupling condensation between (E)-bromo olefin nucleoside (6a-E and 6b-E) and nucleoside-H-phosphonate methyl ester (S2a, S2b, or S2c) yielded a P-OMe-protected ⁱE-VP-linked nucleotide dimer (38). After 3′-O-deprotection and phosphitylation, ⁱE-VP linked dimer phosphoramidites (9a, 9b, and 9c) were synthesized without deprotection of the 3-N-Bz group, which was eventually removed in the final step: base treatment of oligonucleotides. (See supplementary data for details about oligonucleotide synthesis).

Oligonucleotide synthesis

All oligonucleotides used were synthesized on MerMade 12 (BioAutomation) synthesizer in 1 μmol scale using universal UNYLinker support or cholesterol 3′-lcaa CPG (ChemGenes). All commercially available 2′-modified nucleoside phosphoramidites and ⁱE-VP-dimer phosphoramidites were coupled under standard conditions on solid support using 5-(benzylthio)-1-H-tetrazole (BTT) as an activator (see supporting information for details). The efficiencies, equivalent use, and time of reaction for dimer phosphoramidites were similar to monomeric assembly, indicating that this strategy can be used to synthesize ⁱE-VP-modified oligonucleotides with similar yields and efficiency (Supplementary Figures S3, S4). The obtained ⁱE-VP-modified siRNAs were first treated with TMSBr to deprotect the methyl group on ⁱE-VP, followed by standard base deprotection to release oligonucleotides from the solid support (34). Control siRNAs with identical sequence and chemical modification pattern, but without ⁱE-VP modification, were prepared to enable direct comparison of ⁱE-VP incorporation on efficacy independent of sequence and modification pattern (Supplementary Tables S1–S3).

ⁱ E-VP incorporation can positively or negatively impact oligonucleotide-enzyme interactions

To evaluate the impact of ⁱE-VP incorporation on oligonucleotide interactions with nucleases, we synthesized oligonucleotide variants with or without ⁱE-VP incorporated at 5′ and 3′-terminal backbone linkages (ON1-4, Supplementary Table S3). ON1 and 2 where chemically phosphorylated to meet substrate requirements for 5′-phosphate-dependent 5′-exonuclease. We also sought to evaluate how stereo constraint of the backbone impacts the ability of reverse transcriptases (RTs) to synthesize complementary DNA because oligonucleotides can serve as a template for polymerase reactions. To achieve this, we synthesized template oligonucleotide variants with or without ⁱE-VP incorporated between nucleotide positions 6–7 (ON5 and 6, respectively) or 10–11 (ON7 and 8, respectively), which positions located at middle positions in ⁱE-VP modified template oligonucleotide (Supplementary Table S3). For synthesis of all oligonucleotides, an automated RNA synthesizer was used to assemble phosphoramidites on a solid support (see Materials and Methods). After incubating with a variety of exonucleases, full-length oligonucleotide and degradation products were resolved on denaturing gels and visualized by SYBR-Gold. Interestingly, ⁱE-VP-containing ON2 behaved as a better substrate for 5′-P-dependent 5′-exonuclease compared to an oligonucleotide of identical sequence and sugar modification pattern, but with a canonical (non-modified) phosphodiester linkage (ON1) (Figure 2A). At the conditions tested, >70% of ⁱE-VP-containing ON2 was degraded at 60 min, while the majority (>50%) of ON1 remained intact. This result is likely due to the locked torsion angle of ⁱE-VP inducing a better fit in the 5′-P-dependent 5′-exonuclease active site (50). By contrast, ⁱE-VP incorporation at the 5′-terminal backbone linkage provided full stabilization against the 5′-exonuclease BSP (bovine spleen phosphodiesterase II). This is because BSP cleaves at P-O5′ of the phosphodiester linkage, which is replaced with the uncleavable P-C(sp²) bond in the ⁱE-VP structure (ON4, Figure 1B and 2A).

Figure 2. — Impact of incorporating ⁱE-VP into oligonucleotide backbone on nuclease and reverse transcriptase activity. (A) Nuclease digestion test using P-O3′ bond-cleaving 5′-phosphate (P)-dependent 5′-exonuclease (Terminator™, 8.5 mU/μl), P-O5′-cleaving 5′-exonuclease (BSP II, 0.25 U/ml), and P-O3′-bond cleaving snake venom phosphodiesterase I (SVPD, 4 mU/ml). R = PO₃^– for **ON1** and 2, R = H for **ON3** and 4. (B) Reverse transcription by Superscript II and AMV RT using ⁱE-VP-modified RNA template. *Full-length, extended product. **Truncated product stopped at position of ⁱE-VP (6–7 or 10–11) on templates. ***Minor truncated products stopped at phosphorothioate modification on templates.

ⁱ E-VP modification at 3′-terminal backbone linkages provided significant stabilization against the P-O3′-cleaving 3′-exonuclease, snake venom phosphodiesterase I (SVPD). At conditions tested, >90% of the control oligonucleotide (ON1) was degraded within an hour, while ⁱE-VP-modified compound (ON2) showed no detectable degradation, even after 8 h, suggesting the ⁱE-VP structure is highly incompatible with the SVPD active site (Figure 2A).

As a positive control for all nuclease stability experiments, we synthesized PS-containing oligonucleotides. As expected, PS linkages exhibited high resistance against all exonucleases tested (Supplementary Figures S7–S9). However, iE-VP’s protection against SVPD was relatively less than that observed with PS. This became more evident when stabilization effects were evaluated at exaggerated SVPD concentrations and assayed using ⁱE-VP/PS mixed sequences (Supplementary Figures S7 and S8). Altogether, our results from the exonuclease digestion assays demonstrate that backbone stereo-constraining can either disturb or enhance activity of exonucleases in enzyme-specific manner.

To investigate the impact of the ⁱE-VP modification on primer extension by RTs, ⁱE-VP-containing oligonucleotides (^VPG6 and ^VPG10 for ON5 and 6, respectively) and control oligonucleotides (^CtrlG4 and ^CtrlG1, for ON7 and 8, respectively) were ligated to a DNA adapter and hybridized to a 5′-FAM-labeled primer (Figure 2B). When Avian Myeloblastosis Virus reverse transcriptase (AMV-RT) was used, primer extension was quantitatively terminated upon encountering the ⁱE-VP-modified inter-nucleotide position. Truncated band position was shifted between ON5 (ⁱE-VP modification between nucleotides 6–7) and ON7 (ⁱE-VP between nucleotides 10–11) by four nucleotides, confirming specificity of the observed phenomena. Our results suggest that AMV-RT cannot accommodate the stereo-constrained backbone for reverse transcription. In contrast, Superscript II (a genetically-engineered Moloney Murine Leukemia Virus (MMLV) RT variant with better processivity and tolerance for modifications) efficiently utilized ⁱE-VP-containing templates with no termination products observed at the site of ⁱE-VP incorporation (Figure 2B) (51). Collectively, these findings demonstrate that, though it is a minimal structural change, structure constraint of the ⁱE-VP backbone induces huge impact on enzymatic polymerization reaction in enzyme specific manner.

Design and synthesis of chemically-modified siRNAs

To evaluate the impact of ⁱE-VP incorporation on siRNA RISC forming proteins, we designed and synthesized a panel of modified siRNA guide strands, in which individual positions of the backbone were modified with ⁱE-VP (^VPG1-20), as follows:

siRNA conjugate and chemical modification pattern design

Because conjugate-mediated delivery is the dominant strategy for fully chemically modified siRNA delivery in the clinic (8–10,37), we elected to use cholesterol-conjugated fully chemically modified asymmetric siRNA in our studies (Figure 3) (48). The chemical modification pattern consisted of alternating 2′-OMe and 2′-F sugar modifications with terminal PS stabilization for both guide and passenger strands, a configuration that is widely utilized for conjugated siRNAs (1,8,40,42). These compounds are readily internalized by many cell types without requiring formulation, making functional in vitro evaluation straightforward (30). When a similar chemical modification pattern was applied to siRNAs conjugated to GalNAc or hydrophobic moieties, robust in vivo efficacy was achieved (29,37, 43–47).

Figure 3. — Position-dependent impact of ⁱE-VP backbone on siRNA efficacy and impact of ⁱE-VP backbone on thermal stability of siRNA duplex. Silencing efficacy change was defined by eq 1. Red bar graphs indicate siRNAs having neutral or enhanced silencing efficacy compared to control siRNAs. Black bar graphs indicate ⁱE-VP-modified siRNAs showed lower efficacy than that of control siRNAs. See Table 1 and supporting information for sequences, and Supplementary Table S7 for IC₅₀ values used for the calculation of efficacy changes. *Efficacy change that exceed –300% was set as an artificial difference to -300% in the graph. Comparative T_m of siRNA duplex containing ⁱE-VP and corresponding control siRNA duplex was measured in 10 mM Sodium phosphate buffer (pH 7.0) containing 100 mM NaCl and 0.1 mM EDTA. ΔT_m = T_m^VP – T_m^Ctrl. T_m values were determined in triplicate experiments.

siRNA sequence design

We selected previously-validated, highly-potent sequences targeting the huntingtin (HTT) gene (48). The panel of these siRNAs contain a uridine-uridine dimeric sequence at each nucleotide position on the guide strand (Table 1).

Table 1.

Sequence of ⁱE-VP modified guide strands and control guide strands

^VPGuide strands	Sequence (5′ → 3′)^a	Calcd. mass [M-H]^–	Observed mass	Correspond Passenger strands^b	^VPsiRNA #	Correspond ^CtrlsiRNA^c #
^VPG1	P-UvpU#AAUCUCUUUAC#U#G#A#U#A#U#A	6599.4	6599.7	P1	^VPD1	^CtrlD1
^VPG2	P-U#UvpUUUAAAUCCUG#A#G#A#A#G#A#A	6724.5	6724.8	P2	^VPD2	^CtrlD2
^VPG3	P-U#U#UvpUUAAAUCCUG#A#G#A#A#G#A#A	6739.6	6740.7	P2	^VPD3	^CtrlD2
^VPG4	P-U#U#UUvpUAAAUCCUG#A#G#A#A#G#A#A	6740.6	6740.7	P2	^VPD4	^CtrlD2
^VPG5	P-U#C#UCUvpUUACUGAU#A#U#A#A#U#U#A	6614.4	6615.7	P3	^VPD5	^CtrlD3
^VPG6	P-U#A#UGUUvpUUCACAU#A#U#U#G#U#C#A	6631.4	6631.6	P4	^VPD6	^CtrlD4
^VPG7	P-U#G#AAUGUvpUCACGC#A#G#U#G#G#G#C	6802.6	6802.8	P5	^VPD7	^CtrlD5
^VPG8	P-U#A#UCAGCUvpUUUCC#A#G#G#G#U#C#G	6700.5	6700.7	P6	^VPD8	^CtrlD6
^VPG9	P-U#U#AAUCUCUvpUUAC#U#G#A#U#A#U#A	6615.4	6615.7	P1	^VPD9	^CtrlD1
^VPG10	P-U#U#AAUCUCUUvpUAC#U#G#A#U#A#U#A	6615.4	6615.7	P1	^VPD10	^CtrlD1
^VPG11	P-U#U#AACGUCAGUvpUC#A#U#A#A#A#C#C	6675.5	6675.7	P7	^VPD11	^CtrlD7
^VPG12	P-U#C#CACUAUGUUUvpU#C#A#C#A#U#A#U	6590.4	6590.7	P8	^VPD12	^CtrlD8
^VPG13	P-U#C#CAAAUACUGGUvpU#G#U#C#G#G#U	6708.5	6708.7	P9	^VPD13	^CtrlD9
^VPG14	P-U#C#CGGUCACAACA#UvpU#G#U#G#G#U	6707.5	6707.7	P10	^VPD14	^CtrlD10
^VPG15	P-U#A#UGUUUUCACAU#A#UvpU#G#U#C#A	6615.4	6615.7	P14	^VPD15	^CtrlD4
^VPG16	P-U#U#UGGUAGCUGAA#A#G#UvpU#C#U#U	6710.4	6710.7	P11	^VPD16	^CtrlD11
^VPG17	P-U#U#AAUCUCUUUAC#U#G#A#UvpU#U#A	6576.3	6576.6	P1	^VPD17	^CtrlD12
^VPG18	P-U#C#UCUUUACUGAU#A#U#A#A#UvpU#A	6599.4	6599.7	P3	^VPD18	^CtrlD3
^VPG19	P-U#U#AAUCUCUUUAC#U#G#A#U#A#UvpU	6576.3	6576.7	P1	^VPD19	^CtrlD13
^VPG20	P-UvpU#AAUCUCUUUAC#U#G#A#UvpU#UvpU	6513.2	6513.7	P1	^VPD20	^CtrlD14

Open in a new tab

^aUppercase and underlined uppercase represent 2′-OMe and 2′-F, respectively. 5′-end phosphate is represented as ‘P’. Inter-nucleotide phosphorothioate and (E)-vinylphosphonate linkages are indicated by ‘#’ and vp, respectively.

^b3′-end tetraethylene glycol (Teg) -linked cholesterol (Chol) conjugated 15 mer passenger strands (P1–P14) to prepare ^VPsiRNAs (^VPD1–^VPD20) consists with (i) complementary sequence through 5′-end to 15th nucleotide of guide strands and (ii) with fully chemical modification pattern: 5′-N#N#NNNNNNNNNNN#N#N-Teg-Chol-3′ (N and N represent 2′-OMe and 2′-F, respectively. See supporting information for each of passenger strand sequences).

^cComparative control siRNA duplexes corresponding to each ^VPsiRNA. Sequences shown in supporting information.

Synthesis of siRNA panel

The antisense or ‘guide’ strand of the siRNA duplex is more sensitive to modifications than the sense or ‘passenger’ strand (49). Therefore, we sought to evaluate the impact of incorporating ⁱE-VP into the guide strand. We synthesized different guide strands with a single incorporation of ⁱE-VP at a specific inter-nucleotide position (^vpG1-19) or with multiple terminal ⁱE-VP insertions (^vpG20), which were then used to create our screening panel (Table 1). Control guide strands without ⁱE-VP consisting with identical chemical modifications were also synthesized (Supplementary Table S1).

Position-dependent impact of ⁱE-VP incorporation on siRNA efficacy

Using our full panel of ⁱE-VP-modified siRNAs (^VPD1–20) and matched controls (^CtrlD1–14), we systematically evaluated the position-dependent impact of ⁱE-VP in guide strands on siRNA efficacy (Table 1). Specifically, we treated HeLa cells with our siRNA library and measured target HTT gene expression by QuantiGene assay after 72 h (see Materials and Methods). We tested compound efficacy at 7 concentrations to calculate IC₅₀ values (Supplementary Table S7). IC₅₀ of ⁱE-VP-modified siRNAs (^VPD1–20) and control siRNAs (^CtrlD1–14) of identical sequence were used to define ‘efficacy change’ based on the following equation:

(1)

As summarized in Figure 3, the structurally-constraining ⁱE-VP backbone either positively or negatively affected RISC efficacy in a position-specific manner. Incorporating ⁱE-VP was well tolerated at guide termini positions 1–2 (^vpD1) and 19–20 (^vpD19), and enhanced RISC activity. ⁱE-VP was also well tolerated at: (i) ‘seed region’ positions 5–6 (^vpD5) and 6–7 (^vpD6), which are essential for RISC recognition of target mRNA; and (ii) position 10–11 (^vpD10), which is next to the mRNA cleavage site. ⁱE-VP incorporation at all other inter-nucleotide positions resulted in decreased RISC activity.

To exclude the possibility that observed differences in efficacy between ⁱE-VP-containing and control siRNA were due to a change in siRNA duplex thermal stability (T_m), we measured the impact of ⁱE-VP on the T_m of ^VPD1–20 (Supplementary Figure S5). We observed very limited contribution of T_m [ΔT_m = T_m (^VPsiRNA) – T_m (^CtrlsiRNA) = +4.8 to –3.7°C)], and no correlation with functional outcomes (Figure 3 and S6). This suggests that changes in the thermodynamic profile of siRNA duplexes do not significantly contribute to functional differences (Figure 3).

Stereo-constraint of the backbone with ⁱE-VP enhances SNP-based discrimination

Structural limitation is a major driving force of the specificity in inter-molecular recognition. Our finding that structurally-constraining ⁱE-VP is well tolerated at the end of the seed region prompted us to evaluate whether structurally-constraining the backbone could enhance specificity of the RISC-target mRNA complex, particularly in the context of mismatch-based target sequence discrimination. Mismatch-based discrimination by siRNAs is a promising therapeutic strategy for allele-specific silencing (52,53). For instance, in Huntington's disease (HD), the ability to target heterozygous single nucleotide polymorphisms (SNPs) and selectively silence only the mutant allele is clinically preferred (54). We recently developed and optimized SNP-specific siRNA targeting a common HD heterozygous SNP (rs362273) (Conroy, F., Miller, R.A. et al. 2021, in revision). Although this compound is potent and shows reasonable SNP discrimination (∼60X selectivity towards single mismatch), there is still detectable off-target activity in vitro. The injection of this compound in the CNS supports efficient discrimination in most of the regions of the brain but shows a trend towards reduction of the wild-type allele silencing in vivo, particularly in regions with high levels of oligonucleotide accumulation. Both sugar and nucleobase modifications of antisense oligonucleotides were explored as an approach to improve SNP recognition (68), however, the impact of structural constraint of internucleotide backbone of siRNAs on their SNP recognition ability has not been investigated. We hypothesized that a locked ⁱE-VP backbone in the context of a mismatched base pair could alter local RISC structure to induce non-functional RNA-Ago2 accommodation and limit off-target recognition and silencing.

To test this hypothesis, we synthesized a panel of siRNAs targeting HTT SNP rs362273. In these siRNA guide strands, nucleotide position 6 is complementary to the SNP site and position 11 is a secondary mismatch introduced to enhance single mismatch discrimination (Conroy, F., Miller, R.A. et al. 2021, in revision). Guide strands were synthesized with stereo-constraining ⁱE-VP located 5′-upstream to the primary SNP mismatch (^VPG21-23) along with matched controls (^CtrlG15-17) (Supplementary Tables S4, S5). To enable synthesis of these guide strands, we further prepared 2′-OMe-G-(ⁱE-VP)-2′-F-Ur dimer phosphoramidite (9c) using synthetic intermediate 6a-E (Scheme 1). The sequence and chemical modification patterns of SNP-targeting siRNAs (^vpD21-23), corresponding control siRNAs (^CtrlD15-17), and Passenger strands (P15-17) are shown in Supplementary Tables S4-S6. We evaluated siRNA efficacy using a luciferase-based reporter system with a single nucleotide difference in the target region (‘target’ mutant vs ‘mismatch (mm)-target’ wild-type) at seven concentrations to generate IC₅₀ values. Incorporating ⁱE-VP next to the primary SNP site did not negatively impact ‘on-target’ activity – IC₅₀ values of ^CtrlD15 and ^VPD21 were 2.64 nM and 1.74 nM, respectively (Figure 4A). This RISC compatibility of ⁱE-VP at position 5–6 was consistent with our efficacy screening results in HeLa cells (Figure 3), despite the different sequence contexts (ⁱE-VP backbone between 2′-OMe-U/2′-F-U dimer versus 2′-OMe-G/2′-F-U dimer at position 5–6). Importantly, ^VPD21 completely lost the ability to silence mm-target mRNA at all concentrations tested (Figure 4C for passive uptake and 4D for lipid mediated uptake). Similar results were generated in the context of different secondary mismatches (^VPD22 and ^VPD23, Supplementary Figure S10), suggesting that stereo-constraining the backbone next to a SNP can be used as a chemical strategy to enhance mismatch discrimination by siRNAs.

Figure 4. — SNPs discrimination properties of ⁱE-VP-modified siRNA. (A) Sequences of siRNAs (**^CtrlD15** and **^VPD21**), target mRNAs with or without single mismatch (mm) base (target and mm-target, respectively), and IC₅₀ values. (B) Conceptual scheme of ON and OFF-target silencing of **^VPD21**. (C) Passive uptake of siRNA and dose response of **^VPD21** and **^CtrlD15** in the presence of target or mm-target mRNAs. (D) Lipid-mediated uptake and dose response of **^VPD21** and **^CtrlD15** in the presence of target or mm-target mRNAs. ^a^CtrlD15 and **^VPD21** consist of **^CtrlG15**/**P15** and **^VPG21**/**P15**, respectively; ^bUppercase and underlined uppercase represent 2′-OMe and 2′-F, respectively. 5′-end phosphate is represented as ‘5′-P’. Inter-nucleotide phosphorothioate and (E)-vinylphosphonate linkages are indicated by ‘#’ and vp, respectively. Passenger strands have 3′-end tetraethylene glycol (Teg)-linked cholesterol (Chol) conjugate. ^cIC₅₀ was calculated based on duplicate experiments. ^dN.D. = Not determined.

Stereo-constraint of the backbone with ⁱE-VP enhances SNP-based discrimination in vivo in BAC-HD mice

Incorporation of ⁱE-VP next to the mismatch site significantly enhances discrimination between mutant and wt allele, and, most importantly, induces no negative impact on on-target activity. The two-mismatch containing compound shows reasonable discrimination in vivo in mice brain (Conroy, F., Miller, R.A. et al. 2021, in revision). In CNS delivery, the relative accumulation of compounds in tissues is low. To test the hypothesis that incorporation of ⁱE-VP further enhances discrimination in vivo, we needed to select the system where high accumulation of oligonucleotides in tissues can be achieved. Trivalent GalNAc conjugation (30,31,40,44,45) enables robust delivery to liver with high degree of oligonucleotides accumulation and represent a great system to push the limits of discrimination in vivo. The model compound we use in this study targets HD heterozygous SNP (rs362273). This SNP is present in the BAC-HD 103 human transgene, while the mouse gene carries an opposite allele in this location. The mutant HTT protein contains ∼ 103 CAG repeats in the exon one while mouse wt variant has only ∼ 7 CAGs. Thus, the mutant and wt HTT protein differ in size and can be differentially detected by the western blot.

To evaluate single mismatched SNP discrimination ability of ⁱE-VP modified siRNA in liver, trivalent-GalNAc conjugated siRNAs that contains ⁱE-VP backbone at position 5 of internucleotide, secondary mismatch base at position 11, and 5′-termianl (E)-VP in the guide strand was prepared (^VPD24). For the control GalNAc-siRNA (^CtrlD18), guide strand without ⁱE-VP was also prepared (Supplementary Table S8).

BAC-HD mice were treated with ^CtrlD18 and ^VPD24 and level of HTT mutant and wt expression were evaluated at two weeks post injection relative to control. While both ^CtrlD18 and ^VPD24 compounds shows efficient silencing of the mutant protein (>80%, P < 0.0001), the level of wt (non-target) expression was affected differently between two variants. The ^CtrlD18 induce partial ∼60% silencing of the wt, while the level of silencing induced by ^VPD24 did not reach a significance threshold (Figure 5). This data indicates that stereo-constraining of the backbone next to the SNP site allows for significant increase in the degree of SNP discrimination in vivo, even in the tissue where siRNAs accumulated efficiently.

Figure 5. — Introduction of ⁱE-VP next to the SNP site enhances discrimination based on a single nucleotide difference in the target site. GalNAc conjugated siRNAs with (**^VPD24**) and without (**^CtrlD18)** the introduction of the ⁱE-VP next to the SNP site (**^VPD24** and **^CtrlD18**, respectively), were injected at 10 mg/kg dose level SC at ∼12 weeks BAC-HD mice. The level of mutant and WT huntingtin proteins was evaluated using western blot (Protein Simple) in mice livers at two weeks post injection. N-5,6, one-way Anova with Bonferroni correction for multiple comparisons. (A) Level of mutant protein expression. Both compounds induce robust silencing. (B) Level of WT protein expression. **^CtrlD18** induces WT protein silencing which is significantly reduced with **^VPD24** compound.

DISCUSSION

Oligonucleotide structure is responsible for highly-specific interactions with a wide range of biological partners. The vast majority of nucleic acid-protein interactions involve individual backbone contacts (55), where the structural configuration and the angle of the backbone are significant contributors to the efficiency and specificity of molecular recognition. Here, we describe the synthesis and functional characterization of a novel backbone modification, ⁱE-VP, where the bridging oxygen is substituted with a carbon and locked torsion angle at 180^o. We find that stereo-constraining the backbone can decrease or enhance efficiency of enzymatic recognition. Addition of ⁱE-VP to the chemical portfolio of oligonucleotide modifications will enable the use of stereo constraint as a way to modulate efficacy and specificity of a vast range of complex RNP structures, ranging from RISC to aptamer-protein complexes to antiviral inhibitors to CRISPR/Cas genome editing technology.

In the synthesis of ⁱE-VP-linked dimer phosphoramidites, nucleobase protection was essential for efficient Wittig olefination to yield dibromo-olefin derivatives (5a, 5b). Using a base-labile acyl group (benzoyl, Bz) for uracil base protection (3a, 3b) is likely applicable to other nucleosides, such as 4-N-Ac (or Bz)-cytidine, 6-N-Bz-adenosine, and 2-N-iBu-guanosine, which are protected with base-labile acyl groups commonly used for phosphoramidite solid-phase oligonucleotide synthesis, thus it would not require any of nucleobase-specific additional deprotection step (38).

Substitution of P-O5′ of the phosphodiester linkage with a non-cleavable P-C (sp²) bond resulted in full stabilization against a P-O5′-cleaving 5′-exonuclease. Structural rigidity can both positively and negatively affect the binding affinity of enzymes to oligonucleotides and the formation of intermediate structure in enzymatic cleavage reaction. Partial stabilization against a P-O3′-cleaving 3′-exonuclease was likely due to ⁱE-VP-driven structural incompatibility with the enzyme active site. Likewise, the lack of resistance towards P-O3′-cleaving 5′-P-dependent 5′-exonuclease may be due to ⁱE-VP-induced structural fitting in the nuclease active site (50). Stability of ⁱE-VP backbone at the first internucleotide position in the presence of PS at 3′-downstrem of ⁱE-VP (^VPG20) was also tested (Table 1). Rapid cleavage of ^VPG20 was also observed, whereas ^CtrlG14 containing PS at first internucleotide position showed significant resistance against this nuclease (Supplementary Figure S9). Thus, iE-VP can be used to stabilize the 5′ and 3′ end of oligonucleotides, except for the first internucleotide position of 5′-P-bearing siRNA guide strands. Currently, PS is the dominant modification used in the clinic for terminal oligonucleotide protection (5–11). Current clinically-advanced lead oligonucleotides use partially-modified PS bonds to avoid toxicity (56). In these compounds, however, decreasing PS content and increasing PO content reduces metabolic stability and negatively affects duration of effect and efficacy. Combining ⁱE-VP with PS might generate compounds with lower toxicity while maintaining stability and potency. As for any new modification, the impact of ⁱE-VP incorporation on oligonucleotide toxicity must be systematically evaluated in vivo. Enzyme specific impact of ⁱE-VP was also observed with reverse transcriptases. These results imply that ⁱE-VP is useful biology tool to modulate interaction with various enzymes.

Backbone variants that change RNA duplex shape or charge recognition cannot be applied to siRNAs because those modifications disrupt siRNA-Ago2 interactions (57). This trend is especially true for the siRNA guide strand, which is selectively loaded into RISC and is much more sensitive to modifications than the passenger strand (1,49). Therefore, we evaluated the impact of incorporating ⁱE-VP into the guide strand specifically. Published crystal structures of Ago2 in complex with a fully modified guide strand (similar modification pattern used in this study) with or without cleaved substrate show that backbone torsion angles (P-O5′-C5′-C4′) vary from 140° to 177° in the majority of inter-nucleotide backbones in positions 1–14 (58,62,63). This range suggests that the incorporation of ⁱE-VP, expected to restrict the torsion angle to 180°, should not cause major disturbances in structure at most positions. Experimentally, this assumption was not the case. ⁱE-VP was poorly tolerated at many internal regions, except for positions between 5–7 and 10–12. There was no correlation between tolerance of ⁱE-VP, RISC function, and backbone torsion angles from currently-available crystal structures, suggesting that the observed negative effects are related to structural interference during dynamic changes in RISC formation, or in the target-induced fitting required for mRNA cleavage. This finding is consistent with the current understanding that RISC undergoes a significant conformational change to catalyse mRNA cleavage (59).

The tolerance of stereo-angle fixation at the end of the seed region (positions between 5–7) and near a mRNA cleaving site (positions 10–11) suggests that these regions do not undergo a significant conformational change or are weakly/flexibly recognized by Ago2. Indeed, tolerance at positions 5–7 is consistent with structural flexibilities previously observed in RISC crystal structure (62), and in single molecule kinetics analysis of RISC (61). In both analyses, chemical modification and mismatches in this region had minimal impact on siRNA potency and binding affinity/kinetics of the seed with target mRNA (60,61). Definitive conclusions regarding tolerance of ⁱE-VP in the cleavage site are still up for debate, but our findings are consistent with the current understanding that backbone positions 10 and 11 form a kinked structure that can accommodate dynamic conformational changes during target mRNA binding (61). This implies that molecular positioning of the backbone in this region is relatively unrestricted and structurally-constrained ⁱE-VP was tolerated in this region.

Some enhancement in activity was observed with ⁱE-VP incorporation at the guide strand termini (positions 1 and 2). According to crystal structures, the defined torsion angle of the phosphodiester backbone in this region is 158^o (58). Thus, the ⁱE-VP-induced shift to 180° should require some conformational adaptation. It is possible that structural fixation of the relatively flexible duplex terminus may support stable formation of RISC or RISC-substrate complex; and thus, enable sustained RISC function. This conclusion is supported by previous reports showing that incorporation of glycol nucleic acid (GNA) or unlocked-nucleic-acid (UNA) – nucleoside derivatives that induce structural flexibility – in positions 1 and/or 2 significantly decreases RISC activity (64,65). We also observed enhanced RISC activity with ⁱE-VP incorporation at terminal position 19–20 (PAZ binding region). Analysis of the PAZ domain in RISC crystal structure indicates a reasonable fit of ⁱE-VP in this region (66). Fixation of an optimal fit in PAZ may slightly enhance RISC activity (IC₅₀ for ^VPD19 and ^CtrlD13: 45.5 and 82.8 nM, respectively, Supplementary Table S7). Collectively, our findings from experiments incorporating position-specific ⁱE-VP into the siRNA guide strand not only inform strategies for using ⁱE-VP to manipulate RISC efficacy, but also provide structural insight into our current understanding of RISC dynamics.

In this study, we also demonstrate an ⁱE-VP-driven enhancement of siRNA sensitivity towards a G to A SNP. This finding has important implications for mismatch-based discrimination strategies to achieve allele-specific silencing in disease treatment (53). While some siRNAs and ASOs have been developed to exhibit mismatch sensitivity, the level a single mismatch discrimination is usually in a range of an order of magnitude of difference in IC₅₀ (44,58,59), and is not sufficient to translate into robust gene discrimination in vivo. Extensive chemical engineering has been employed to continue enhancing ASO mismatch-based recognition (67,68). For siRNAs, incorporation of a GNA modification at position 7 in the seed has been explored to limit potential microRNA-like target recognition (which does not require perfect complementarity) by siRNAs in vivo and enhance their overall specificity (69). In our study, stereo-constraining the backbone next to the primary SNP significantly enhanced siRNA ability to discriminate between mRNA alleles based on a single mismatch. This discrimination enabled productive silencing of target Mut-HTT, without any silencing of the wild type-HTT allele, generating a compound with complete selectivity in vitro based on a single mismatch.

The observed enhanced mismatch sensitivity is likely driven by the structural restriction of the functional accommodation. Normally, the structure of the RISC-substrate complex can accommodate mismatches. This accommodation requires an induced fit at the SNP-proximal backbones, which is fully blocked by the introduction of structurally-constrained ⁱE-VP. Future studies should confirm that similar enhancement can be observed with mismatches of different nucleotide composition.

Enhancement in SNP-based discrimination essential for translation of the SNP-selective compounds towards clinical evaluation, specifically for indications where preservation of wt allele expression is essential. Oligonucleotides do not distribute uniformly in many tissues, specifically CNS, resulting in highly significant differences in degree of accumulation across brain regions (47,56). This bias complicates ability to achieve allele specific silencing in tissue with highly different accumulation levels. Here we further demonstrate, that introduction of ⁱE-VP next to the mismatch allows significant enhancement of selectivity at conditions where delivery level are saturated. We intentionally treated animals with high dose, 10 mg/kg, which results in saturating level of huntingtin targeting compounds accumulation in liver. While the control compounds, containing two mismatches shows reasonable discrimination in the brain, it silencing both mutant and wild type variant in liver in vivo. Incorporation of the ⁱE-VP significantly reduced the level of wt HTT protein silencing without compromising efficacy toward mutant protein. Enhancing siRNA selectivity and abolishing single mismatch silencing by ⁱE-VP modification could pave the path towards developing fully allele-specific siRNAs for devastating CNS disorders, including HD.

In summary, we describe a modified high-yielding procedure enabling incorporation of ⁱE-VP backbone modifications into 2′-OMe and 2′-F modified oligonucleotides. The modified synthetic protocol for ⁱE-VP includes nucleobase protection with base labile benzoyl group, which was essential for successful adaptation of this chemistry for 2′-OMe and F modified nucleosides. Using our newly-synthesized ⁱE-VP-modified oligonucleotides, we determined that stereo constraint of the oligonucleotide backbone can inhibit or enhance protein recognition, depending on the enzyme in question and the position of the ⁱE-VP within the oligonucleotide. ⁱE-VP provided significant 3′ terminal stabilization. Our systematic evaluation of ⁱE-VP incorporation on in vitro siRNA efficacy demonstrates that this modification enhances activity when incorporated at termini and is tolerated in certain internal positions of the guide strand. The ability to internally modify siRNA with ⁱE-VP will expand the tool box of backbone modifications available for modulating clinical siRNA properties. Incorporation of structurally-constraining ⁱE-VP next to a SNP-mismatch base pair located in the seed region results in significant enhancement of SNP-mismatch discrimination. This novel strategy could pave the way towards the development of highly-discriminatory allele-specific siRNAs, which could potentially treat neurodegenerative disorders in the future. While ⁱE-VP fixes the torsion angle at 180^o, it has minimal impact on backbone size, charge, and geometry, making it a useful modification available for, not just siRNAs, but also other oligonucleotides that require extensive protein interactions, such as RNase-H dependent ASOs and CRISPR-RNAs.

DATA AVAILABILITY

The data that support this study are available from the corresponding authors upon reasonable request.

Supplementary Material

gkab1126_Supplemental_File

Click here for additional data file.^{(17.9MB, pdf)}

ACKNOWLEDGEMENTS

We thank all Khvorova Lab members for intensive discussions from various point of views, Dr Akbar Ali for maintaining infrastructure for NMR, Dr Yanglan Tan for measurement of HRMS, Dr Emily Haberlin for reviewing the manuscript.

Notes

Present address: Anastasia Khvorova, RNA Therapeutics Institute, University of Massachusetts Medical School, Worcester, MA 01605, USA.

Contributor Information

Ken Yamada, RNA Therapeutics Institute, University of Massachusetts Medical School, 368 Plantation Street, Worcester, MA 01605, USA.

Samuel Hildebrand, RNA Therapeutics Institute, University of Massachusetts Medical School, 368 Plantation Street, Worcester, MA 01605, USA.

Sarah M Davis, RNA Therapeutics Institute, University of Massachusetts Medical School, 368 Plantation Street, Worcester, MA 01605, USA.

Rachael Miller, RNA Therapeutics Institute, University of Massachusetts Medical School, 368 Plantation Street, Worcester, MA 01605, USA; Department of Medicine, University of Massachusetts Medical School, Worcester, MA, USA.

Faith Conroy, RNA Therapeutics Institute, University of Massachusetts Medical School, 368 Plantation Street, Worcester, MA 01605, USA; Department of Medicine, University of Massachusetts Medical School, Worcester, MA, USA.

Ellen Sapp, Department of Neurology, Harvard Medical School and MassGeneral Institute for Neurodegenerative Disease, Charlestown, MA, USA.

Jillian Caiazzi, RNA Therapeutics Institute, University of Massachusetts Medical School, 368 Plantation Street, Worcester, MA 01605, USA.

Julia F Alterman, RNA Therapeutics Institute, University of Massachusetts Medical School, 368 Plantation Street, Worcester, MA 01605, USA.

Loic Roux, RNA Therapeutics Institute, University of Massachusetts Medical School, 368 Plantation Street, Worcester, MA 01605, USA.

Dimas Echeverria, RNA Therapeutics Institute, University of Massachusetts Medical School, 368 Plantation Street, Worcester, MA 01605, USA.

Matthew R Hassler, RNA Therapeutics Institute, University of Massachusetts Medical School, 368 Plantation Street, Worcester, MA 01605, USA.

Edith L Pfister, Department of Medicine, University of Massachusetts Medical School, Worcester, MA, USA.

Marian DiFiglia, Department of Neurology, Harvard Medical School and MassGeneral Institute for Neurodegenerative Disease, Charlestown, MA, USA.

Neil Aronin, RNA Therapeutics Institute, University of Massachusetts Medical School, 368 Plantation Street, Worcester, MA 01605, USA; Department of Medicine, University of Massachusetts Medical School, Worcester, MA, USA.

Anastasia Khvorova, RNA Therapeutics Institute, University of Massachusetts Medical School, 368 Plantation Street, Worcester, MA 01605, USA; Program in Molecular Medicine, University of Massachusetts Medical School, Worcester, MA, USA.

SUPPLEMENTARY DATA

Supplementary Data are available at NAR Online.

FUNDING

NIH/NINDS [R01 NS104022 for A.K.]; S10 [OD020012 for A.K.]; MIRA [R35 GM131839 for A.K.]; NIH CREATE [U01 NS114098 for N.A.]; CHDI Foundation [RecID A-5038 for N.A.]. The open access publication charge for this paper has been waived by Oxford University Press – NAR Editorial Board members are entitled to one free paper per year in recognition of their work on behalf of the journal.

Conflict of interest statement. None declared.

REFERENCES

1. Khvorova A., Watts J.K.. The chemical evolution of oligonucleotide therapies of clinical utility. Nat. Biotechnol. 2017; 35:238–248. [DOI] [PMC free article] [PubMed] [Google Scholar]
2. Crooke S.T., Witztum J.L., Bennett C.F., Baker B.F.. RNA targeted therapeutics. Cell Metab. 2018; 27:714–739. [DOI] [PubMed] [Google Scholar]
3. Levin A.A. Treating disease at the RNA level with oligonucleotides. N. Engl. J. Med. 2019; 380:57–70. [DOI] [PubMed] [Google Scholar]
4. Shen X., Corey D.R.. Chemistry, mechanism and clinical status of antisense oligonucleotides and duplex RNAs. Nucleic Acids Res. 2018; 46:1584–1600. [DOI] [PMC free article] [PubMed] [Google Scholar]
5. Eckstein F. Phosphorothioates, essential components of therapeutic oligonucleotides. Nucleic Acid Ther. 2014; 24:374–387. [DOI] [PubMed] [Google Scholar]
6. Egli M., Manoharan M.. Re-engineering RNA molecules into Therapeutic Agents. Acc. Chem. Res. 2019; 52:1036–1047. [DOI] [PubMed] [Google Scholar]
7. Deleavey G.F., Damha M.J.. Designing chemically modified oligonucleotides for targeted gene silencing. Chem. Biol. 2012; 19:937–954. [DOI] [PubMed] [Google Scholar]
8. Khvorova A. Oligonucleotide therapeutics—a new class of cholesterol-lowering drugs. N. Engl. J. Med. 2017; 376:4–7. [DOI] [PubMed] [Google Scholar]
9. Ray K.K., Landmesser U., Leiter L.A., Kallend D., Dufour R., Karakas M., Hall T., Troquay R.P., Turner T., Visseren F.L.et al.. Inclisiran in patients at high cardiovascular risk with elevated LDL cholesterol. N. Engl. J. Med. 2017; 376:1430–1440. [DOI] [PubMed] [Google Scholar]
10. Fitzgerald K., White S., Borodovsky A., Bettencourt B.R., Strahs A., Clausen V., Wijngaard P., Horton J.D., Taubel J., Brooks A.et al.. Highly durable RNAi therapeutic inhibitor of PCSK9. N. Engl. J. Med. 2017; 376:41–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
11. Soutschek J., Akinc A., Bramlage B., Charisse K., Constien R., Donoghue M., Elbashir S., Geick A., Hadwiger P., Harborth J.et al.. Therapeutic silencing of an endogenous gene by systemic administration of modified siRNAs. Nature. 2004; 432:173–178. [DOI] [PubMed] [Google Scholar]
12. Wang S., Allen N., Vickers T.A., Revenko A.S., Sun H., Liang X.H., Crooke S.T.. Cellular uptake mediated by epidermal growth factor receptor facilitates the intracellular activity of phosphorothioate-modified antisense oligonucleotides. Nucleic Acids Res. 2018; 46:3579–3594. [DOI] [PMC free article] [PubMed] [Google Scholar]
13. Iannitti T., Morales-Medina J.C., Palmieri B.. Phosphorothioate oligonucleotides: effectiveness and toxicity. Curr. Drug Targets. 2014; 15:663–673. [DOI] [PubMed] [Google Scholar]
14. Crooke S.T., Seth P.P., Vickers T.A., Liang X.. The interaction of phosphorothioate-containing RNA targeted drugs with proteins is a critical determinant of the therapeutic effect of these agents. J. Am. Chem. Soc. 2020; 142:14754–14771. [DOI] [PubMed] [Google Scholar]
15. Shen W., Hoyos C.L.D., Sun H., Vickers T.A., Liang X., Crooke S.T.. Acute hepatotoxicity of 2′ fluoro-modified 5-10-5 gapmer phosphorothioate oligonucleotides in mice correlates with intracellular protein binding and the loss of DBHS proteins. Nucleic Acids Res. 2018; 46:2204–2217. [DOI] [PMC free article] [PubMed] [Google Scholar]
16. Clercq E.D., Holy A., Rosenberg I.. 1988; Therapeutical Application of Phosphonylmethoxyalkyl Adenines. US Patent 4724233A. [Google Scholar]
17. Engel R. Phosphonates as analogues of natural phosphates. Chem. Rev. 1977; 77:349–367. [Google Scholar]
18. Hecker S.J., Erion M.D.. Prodrugs of phosphates and phosphonates. J. Med. Chem. 2008; 51:2328–2345. [DOI] [PubMed] [Google Scholar]
19. Hayes V.E., Ternan N.G., McMullan G.. Organophosphonate metabolism by a moderately halophilic bacterial isolate. FEMS Microbiol. Lett. 2000; 186:171–175. [DOI] [PubMed] [Google Scholar]
20. Zhao Z., Caruthers M.H.. Synthesis and preliminary biochemical studies with 5′-deoxy-5′-methylidyne phosphonate linked thymidine oligonucleotides. Tetrahedron Lett. 1996; 37:6239–6242. [Google Scholar]
21. Abbas S., Hayes C.J., Worden S.. The ‘Hirao reduction’ revisited: a procedure for the synthesis of terminal vinyl bromides by the reduction of 1,1-dibromoalkenes. Tetrahedron Lett. 2000; 41:3215–3219. [Google Scholar]
22. Bertram R., Hayes C.J., Soultanas P.. Vinylphosphonate internucleotide linkages inhibit the ictivity of PcrA DNA helicase. Biochemistry. 2002; 41:7725–7731. [DOI] [PMC free article] [PubMed] [Google Scholar]
23. Abbas S., Hayes C.J.. An improved procedure for the synthesis of vinylphosphonate-linked nucleic acid. Tetrahedron Lett. 2000; 41:4513–4517. [Google Scholar]
24. Garcia P.L., Bradley G., Hayes C.L., Krintel S., Soultanas P., Janscak P.. RPA alleviates the inhibitory effect of vinylphosphonate internucleotide linkage on DNA unwinding by BLM and WRN helicases. Nucleic Acids Res. 2004; 32:3771–3778. [DOI] [PMC free article] [PubMed] [Google Scholar]
25. Doddridge Z., Bertram R.D., Hayes C.L., Soultanas P.. Effect of vinylphosphonate internucleotide linkages on the cleavage specificity of exonuclease III and on the activity of DNA polymerase I. Biochemistry. 2003; 42:3239–3246. [DOI] [PMC free article] [PubMed] [Google Scholar]
26. Starrett J.E. Jr, Tortolani D.R., Russell J., Hitchcock M.J., Whiterock V., Martin J.C., Mansuri M.M.. Synthesis, oral bioavailability determination, and in vitro evaluation of prodrugs of the antiviral agent 9-[2-(phosphonomethoxy)ethyl]adenine (PMEA). J. Med. Chem. 1994; 37:1857–1864. [DOI] [PubMed] [Google Scholar]
27. Navé J.F., Taylor D., Tyms S., Kenny M., Eggenspiller A., Eschbach A., Dulworth J., Brennan T., Piriou F., Halazy S.. Synthesis, antiviral activity and enzymatic phosphorylation of 9-phosphonopentenyl derivatives of guanine. Antiviral Res. 1995; 27:301–316. [DOI] [PubMed] [Google Scholar]
28. Lima W.F., Prakash T.P., Murray H.M., Kinberger G.A., Li W., Chappell A.E., Li C.S., Murray S.F., Gaus H., Seth P.P.et al.. Single-stranded siRNAs Activate RNAi in Animals. Cell. 2012; 150:883–894. [DOI] [PubMed] [Google Scholar]
29. Matsuda S., Keiser K., Nair J.K., Charisse K., Manoharan R.M., Kretschmer P., Peng C.G., Kel’in A.V., Kandasamy P., Willoughby J.L.S.et al.. siRNA conjugates carrying sequentially assembled trivalent N-acetylgalactosamine linked through nucleosides elicit robust gene silencing in vivo in hepatocytes. ACS Chem. Biol. 2015; 10:1181–1187. [DOI] [PubMed] [Google Scholar]
30. Prakash T.P., Kinberger G.A., Murray H.M., Chappell A., Riney S., Graham M.J., Lima W.F., Swayze E.E., Seth P.P.. Synergistic effect of phosphorothioate, 5′-vinylphosphonate and GalNAc modifications for enhancing activity of synthetic siRNA. Bioorg. Med. Chem. Lett. 2016; 26:2817–2820. [DOI] [PubMed] [Google Scholar]
31. Parmar R., Willoughby J.L., Liu J., Foster D.J., Brigham B., Theile C.S., Charisse K., Akinc A., Guidry E., Pei Y.et al.. 5′-(E)-Vinylphosphonate: a stable phosphate mimic can improve the RNAi activity of siRNA-GalNAc conjugates. ChemBioChem. 2016; 17:985–989. [DOI] [PubMed] [Google Scholar]
32. Elkayam E., Parmar R., Brown C.R., Willoughby J.L., Theile C.S., Manoharan M., Joshua-Tor L.. siRNA carrying an (E)-vinylphosphonate moiety at the 5′-end of the guide strand augments gene silencing by enhanced binding to human Argonaute-2. Nucleic Acids Res. 2017; 45:3528–3536. [DOI] [PMC free article] [PubMed] [Google Scholar]
33. Prakash T.P., Lima W.F., Murray H.M., Li W., Kinberger G.A., Chappell A.E., Gaus H., Seth P.P., Bhat B., Crooke S.T.et al.. Identification of metabolically stable 5′-phosphate analogs that support single strand siRNA activity. Nucleic Acids Res. 2015; 43:2993–3011. [DOI] [PMC free article] [PubMed] [Google Scholar]
34. Haraszti R.A., Roux L., Coles A.H., Turanov A.A., Alterman J.F., Echeverria D., Godinho B.M.D.C., Aronin N., Khvorova A.. 5′-Vinylphosphonate improves tissue accumulation and efficacy of conjugated siRNAs in vivo. Nucleic Acids Res. 2017; 45:7581–7592. [DOI] [PMC free article] [PubMed] [Google Scholar]
35. Parmar R.G., Brown C.R., Matsuda S., Willoughby J.L.S., Theile C.S., Charissé K., Foster D.J., Zlatev I., Jadhav V., Maier M.A.et al.. Facile synthesis, geometry, and 2′-substituent-dependent in vivo activity of 5′-(E)- and 5′-(Z)-vinylphosphonate-modified siRNA conjugates. J. Med. Chem. 2018; 61:734–744. [DOI] [PubMed] [Google Scholar]
36. DiFiglia M., Sapp E., Chase K., Schwarz C., Meloni A., Young C., Martin E., Vonsattel J.P., Carraway R., Reeves S.A.et al.. Huntingtin is a cytoplasmic protein associated with vesicles in human and rat brain neurons. Neuron. 1995; 14:1075–1081. [DOI] [PubMed] [Google Scholar]
37. Biscans A., Coles A., Haraszti R., Echeverria D., Hassler M., Osborn M., Khvorova A.. Diverse lipid conjugates for functional extra-hepatic siRNA delivery in vivo. Nucleic Acids Res. 2019; 47:1082–1096. [DOI] [PMC free article] [PubMed] [Google Scholar]
38. Abbas S., Bertram R.D., Hayes C.J.. Commercially available 5′-DMT phosphoramidites as reagents for the synthesis of vinylphosphonate-linked oligonucleic acids. Org. Lett. 2001; 3:3365–3367. [DOI] [PubMed] [Google Scholar]
39. Allerson C.R., Sioufi N., Jarres R., Prakash T.P., Naik N., Berdeja A., Wanders L., Griffey R.H., Swayze E.E., Bhat B.. Fully 2′-modified oligonucleotide duplexes with improved in vitro potency and stability compared to unmodified small interfering RNA. J. Med. Chem. 2005; 48:901–904. [DOI] [PubMed] [Google Scholar]
40. Foster D.J., Brown C.R., Shaikh S., Trapp C., Schlegel M.K., Qian K., Sehgal A., Rajeev K.G., Jadhav V., Manoharan M.et al.. Advanced siRNA designs further improve in vivo performance of GalNAc-siRNA conjugates. Mol. Ther. 2018; 26:708–717. [DOI] [PMC free article] [PubMed] [Google Scholar]
41. Hirao T., Masunaga T., Ohshiro Y., Agawa T.. Reduction of gem-dibromides with diethyl phosphite. J. Org. Chem. 1981; 46:3745–3747. [Google Scholar]
42. Hassler M.R., Turanov A.A., Alterman J.F., Haraszti R.A., Coles A.H., Osborn M.F., Echeverria D., Nikan M, Salomon W.F., Roux L.et al.. Comparison of partially and fully chemically-modified siRNA in conjugate-mediated delivery in vivo. Nucleic Acids Res. 2018; 46:2185–2196. [DOI] [PMC free article] [PubMed] [Google Scholar]
43. Ly S., Navaroli D.M., Didiot M.C., Cardia J., Pandarinathan L., Alterman J.F., Fogarty K., Standley C., Lifshitz L.M.et al.. Visualization of self-delivering hydrophobically modified siRNA cellular internalization. Nucleic Acids Res. 2017; 45:15–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
44. Nair J.K., Willoughby J.L.S., Chan A., Charisse K., Alam M.R., Wang Q., Hoekstra M, Kandasamy P., Kel’in A.V., Milstein S.et al.. Multivalent N-acetylgalactosamine-conjugated siRNA localizes in hepatocytes and elicits robust RNAi-mediated gene silencing. J. Am. Chem. Soc. 2014; 136:16958–16961. [DOI] [PubMed] [Google Scholar]
45. Østergaard M.E., Yu J., Kinberger G.A., Wan W.B., Migawa M.T., Vasquez G., Schmidt K., Gaus H.J., Murray H.M., Low A.et al.. Effcient synthesis and biological evaluation of 5′-GalNAc conjugated antisense oligonucleotides. Bioconjug. Chem. 2015; 26:1451–1455. [DOI] [PubMed] [Google Scholar]
46. Rajeev K.G., Nair J.K., Jayaraman M., Charisse K., Taneja N., O'Shea J., Willoughby J.L., Yucius K., Nguyen T., Shulga-Morskaya S.et al.. Hepatocyte-specific delivery of siRNAs conjugated to novel non-nucleosidic trivalent N-acetylgalactosamine elicits robust gene silencing in vivo. Chembiochem. 2015; 16:903–908. [DOI] [PubMed] [Google Scholar]
47. Alterman J.F., Godinho B.M.D.C., Hassler M.R., Ferguson C.M., Echeverria D., Sapp E., Haraszti R.A., Coles A.H., Conroy F., Miller R.et al.. A divalent siRNA chemical scaffold for potent and sustained modulation of gene expression throughout the central nervous system. Nat. Biotechnol. 2019; 37:884–894. [DOI] [PMC free article] [PubMed] [Google Scholar]
48. Alterman J.F., Hall L.M., Coles A.H., Hassler M.R., Didiot M.C., Chase K., Abraham J., Sottosanti E., Johnson E., Sapp E.et al.. Hydrophobically modified siRNAs silence Hutingtin mRNA in primary neurons and mouse brain. Mol. Ther. Nucleic Acids. 2015; 4:e266. [DOI] [PMC free article] [PubMed] [Google Scholar]
49. Prakash T.P., Allerson C.R., Dande P., Vickers T.A., Sioufi N., Jarres R., Baker B.F., Swayze E.E., Griffey R.H., Bhat B.. Positional effect of chemical modifications on short interference RNA activity in mammalian cells. J. Med. Chem. 2005; 48:4247–4253. [DOI] [PubMed] [Google Scholar]
50. Jinek M., Coyle S.M., Doudna J.A.. Coupled 5′ nucleotide recognition and processivity in Xrn1-mediated mRNA decay. Mol. Cell. 2011; 41:600–608. [DOI] [PMC free article] [PubMed] [Google Scholar]
51. Stovall G.M., Bedenbaugh R.S., Singh S., Meyer A.J., Hatala P.J., Elliington A.D., Hall B.. In vitro selection using modified or unnatural nucleotides. Curr. Protoc. Nucleic Acid Chem. 2014; 56:9.6.1. [DOI] [PMC free article] [PubMed] [Google Scholar]
52. Ross C.A., Aylward E.H., Wild E.J., Langbehn D.R., Long J.D., Warner J.H., Scahill R.I., Leavitt B.R., Stout J.C., Paulsen J.S.et al.. Huntington disease: natural history, biomarkers and prospects for therapeutics. Nat. Rev. Neurol. 2014; 10:204–216. [DOI] [PubMed] [Google Scholar]
53. Pfister E.L., Zamore P.D.. Huntington's disease: silencing a brutal killer. Exp. Neurol. 2009; 220:226–229. [DOI] [PMC free article] [PubMed] [Google Scholar]
54. Pfister E.L., Kennington L., Straubhaar J., Wagh S., Liu W., DiFiglia M., Landwehrmeyer B., Vonsattel J-P., Zamore P.D., Aronin N.. Five siRNAs targeting three SNPs in Huntingin may provide therapy for three-quarters of Huntington's disease patients. Curr. Biol. 2009; 19:774–778. [DOI] [PMC free article] [PubMed] [Google Scholar]
55. von Hippel P. H. From “Simple” DNA-protein interactions to the macromolecular machines of gene expression. Annu. Rev. Biophys. Biomol. Struct. 2007; 36:79–105. [DOI] [PMC free article] [PubMed] [Google Scholar]
56. Tabrizi S.J., Leavitt B.R., Landwehrmeyer G.B., Wild E.J., Saft C., Barker R.A., Blair N.F., Craufurd D., Priller J., Rickards H.et al.. Phase 1–2a IONIS-HTTRx study site teams, targeting Huntingtin expression in patients with Huntington's disease. N. Engl. J. Med. 2019; 380:2307–2316. [DOI] [PubMed] [Google Scholar]
57. Chiu Y.L., Rana T.M.. siRNA function in RNAi: a chemical modification analysis. RNA. 2003; 9:1034–1048. [DOI] [PMC free article] [PubMed] [Google Scholar]
58. Schirle N.T., Sheu-Gruttadauria J., MacRae I.J.. Structural basis for MicroRNA targeting. Science. 2014; 346:608–613. [DOI] [PMC free article] [PubMed] [Google Scholar]
59. Preall J.B., Sontheimer E.J.. RNAi: RISC gets loaded. Cell. 2005; 123:543–545. [DOI] [PubMed] [Google Scholar]
60. Kawamata T., Tomari Y.Making RISC. . Trends Biochem. Sci. 2010; 35:368–376. [DOI] [PubMed] [Google Scholar]
61. Salomon W.E., Jolly S.M., Moore M.J., Zamore P.D., Serebrov V.. Single-molecule imaging reveals that Argonaute reshapes the binding properties of its nucleic acid guides. Cell. 2015; 162:84–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
62. Schirle N.T., Kinberger G.A., Murray H.F., Lima W.F., Prakash T.P., MacRae I.J.. Structural analysis of human Argonaute-2 bound to a modified siRNA guide. J. Am. Chem. Soc. 2016; 138:8694–8697. [DOI] [PMC free article] [PubMed] [Google Scholar]
63. Wang Y., Sheng G., Juranek S., Tuschl T., Patel D.J.. Structure of the guide-strand-containing argonaute silencing complex. Nature. 2008; 456:209–213. [DOI] [PMC free article] [PubMed] [Google Scholar]
64. Schlegel M.K., Foster D.J., Kel’in A.V., Zlatev I., Bisbe A., Jayaraman M., Lackey J.G., Rajeev K.G., Charisse K., Harp J.et al.. Chirality dependent potency enhancement and structural impact of glycol nucleic acid modification on siRNA. J. Am. Chem. Soc. 2017; 139:8537–8546. [DOI] [PubMed] [Google Scholar]
65. Kenski D.M., Cooper A.J., Li J.J., Willingham A.T., Haringsma H.J., Young T.A., Kuklin N.A., Jones J.J., Cancilla M.T., McMasters E.R.et al.. Analysis of acyclic nucleoside modifications in siRNAs finds sensitivity at position 1 that is restored by 5′-terminal phosphorylation both in vitro and in vivo. Nucleic Acids Res. 2010; 38:660–671. [DOI] [PMC free article] [PubMed] [Google Scholar]
66. Ma J.B., Ye K., Patel D.J.. Structural basis for overhang- specific small interfering RNA recognition by the PAZ domain. Nature. 2004; 429:318–322. [DOI] [PMC free article] [PubMed] [Google Scholar]
67. Østergaard M.E., Kumar P., Nichols J., Watt A., Sharma P.K., Nielsen P., Seth P.P.. Allele-selective inhibition of mutant Huntingtin with 2-thio- and C5- triazolylphenyl-deoxythymidine-modified antisense oligonucleotides. Nucleic Acid Ther. 2015; 25:266–274. [DOI] [PubMed] [Google Scholar]
68. Østergaard M.E., Nichols J., Dwight T.A., Lima W., Jung M.E., Swayze E.E., Seth P.P.. Fluorinated nucleotide modifications modulate allele selectivity of SNP-targeting antisense oligonucleotides. Mol. Ther. Nucleic Acids. 2017; 7:20–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
69. Janas M.M., Schlegel M.K., Harbison C.E., Yilmaz V.O., Jiang Y., Parmar R., Zlatev I., Castoreno A., Xu H., Shulga-Morskaya S.et al.. Selection of GalNAc-conjugated siRNAs with limited off-target-driven rat hepatotoxicity. Nat. Commun. 2018; 9:723. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

gkab1126_Supplemental_File

Click here for additional data file.^{(17.9MB, pdf)}

Data Availability Statement

The data that support this study are available from the corresponding authors upon reasonable request.

[B1] 1. Khvorova A., Watts J.K.. The chemical evolution of oligonucleotide therapies of clinical utility. Nat. Biotechnol. 2017; 35:238–248. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B2] 2. Crooke S.T., Witztum J.L., Bennett C.F., Baker B.F.. RNA targeted therapeutics. Cell Metab. 2018; 27:714–739. [DOI] [PubMed] [Google Scholar]

[B3] 3. Levin A.A. Treating disease at the RNA level with oligonucleotides. N. Engl. J. Med. 2019; 380:57–70. [DOI] [PubMed] [Google Scholar]

[B4] 4. Shen X., Corey D.R.. Chemistry, mechanism and clinical status of antisense oligonucleotides and duplex RNAs. Nucleic Acids Res. 2018; 46:1584–1600. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B5] 5. Eckstein F. Phosphorothioates, essential components of therapeutic oligonucleotides. Nucleic Acid Ther. 2014; 24:374–387. [DOI] [PubMed] [Google Scholar]

[B6] 6. Egli M., Manoharan M.. Re-engineering RNA molecules into Therapeutic Agents. Acc. Chem. Res. 2019; 52:1036–1047. [DOI] [PubMed] [Google Scholar]

[B7] 7. Deleavey G.F., Damha M.J.. Designing chemically modified oligonucleotides for targeted gene silencing. Chem. Biol. 2012; 19:937–954. [DOI] [PubMed] [Google Scholar]

[B8] 8. Khvorova A. Oligonucleotide therapeutics—a new class of cholesterol-lowering drugs. N. Engl. J. Med. 2017; 376:4–7. [DOI] [PubMed] [Google Scholar]

[B9] 9. Ray K.K., Landmesser U., Leiter L.A., Kallend D., Dufour R., Karakas M., Hall T., Troquay R.P., Turner T., Visseren F.L.et al.. Inclisiran in patients at high cardiovascular risk with elevated LDL cholesterol. N. Engl. J. Med. 2017; 376:1430–1440. [DOI] [PubMed] [Google Scholar]

[B10] 10. Fitzgerald K., White S., Borodovsky A., Bettencourt B.R., Strahs A., Clausen V., Wijngaard P., Horton J.D., Taubel J., Brooks A.et al.. Highly durable RNAi therapeutic inhibitor of PCSK9. N. Engl. J. Med. 2017; 376:41–51. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B11] 11. Soutschek J., Akinc A., Bramlage B., Charisse K., Constien R., Donoghue M., Elbashir S., Geick A., Hadwiger P., Harborth J.et al.. Therapeutic silencing of an endogenous gene by systemic administration of modified siRNAs. Nature. 2004; 432:173–178. [DOI] [PubMed] [Google Scholar]

[B12] 12. Wang S., Allen N., Vickers T.A., Revenko A.S., Sun H., Liang X.H., Crooke S.T.. Cellular uptake mediated by epidermal growth factor receptor facilitates the intracellular activity of phosphorothioate-modified antisense oligonucleotides. Nucleic Acids Res. 2018; 46:3579–3594. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B13] 13. Iannitti T., Morales-Medina J.C., Palmieri B.. Phosphorothioate oligonucleotides: effectiveness and toxicity. Curr. Drug Targets. 2014; 15:663–673. [DOI] [PubMed] [Google Scholar]

[B14] 14. Crooke S.T., Seth P.P., Vickers T.A., Liang X.. The interaction of phosphorothioate-containing RNA targeted drugs with proteins is a critical determinant of the therapeutic effect of these agents. J. Am. Chem. Soc. 2020; 142:14754–14771. [DOI] [PubMed] [Google Scholar]

[B15] 15. Shen W., Hoyos C.L.D., Sun H., Vickers T.A., Liang X., Crooke S.T.. Acute hepatotoxicity of 2′ fluoro-modified 5-10-5 gapmer phosphorothioate oligonucleotides in mice correlates with intracellular protein binding and the loss of DBHS proteins. Nucleic Acids Res. 2018; 46:2204–2217. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B16] 16. Clercq E.D., Holy A., Rosenberg I.. 1988; Therapeutical Application of Phosphonylmethoxyalkyl Adenines. US Patent 4724233A. [Google Scholar]

[B17] 17. Engel R. Phosphonates as analogues of natural phosphates. Chem. Rev. 1977; 77:349–367. [Google Scholar]

[B18] 18. Hecker S.J., Erion M.D.. Prodrugs of phosphates and phosphonates. J. Med. Chem. 2008; 51:2328–2345. [DOI] [PubMed] [Google Scholar]

[B19] 19. Hayes V.E., Ternan N.G., McMullan G.. Organophosphonate metabolism by a moderately halophilic bacterial isolate. FEMS Microbiol. Lett. 2000; 186:171–175. [DOI] [PubMed] [Google Scholar]

[B20] 20. Zhao Z., Caruthers M.H.. Synthesis and preliminary biochemical studies with 5′-deoxy-5′-methylidyne phosphonate linked thymidine oligonucleotides. Tetrahedron Lett. 1996; 37:6239–6242. [Google Scholar]

[B21] 21. Abbas S., Hayes C.J., Worden S.. The ‘Hirao reduction’ revisited: a procedure for the synthesis of terminal vinyl bromides by the reduction of 1,1-dibromoalkenes. Tetrahedron Lett. 2000; 41:3215–3219. [Google Scholar]

[B22] 22. Bertram R., Hayes C.J., Soultanas P.. Vinylphosphonate internucleotide linkages inhibit the ictivity of PcrA DNA helicase. Biochemistry. 2002; 41:7725–7731. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B23] 23. Abbas S., Hayes C.J.. An improved procedure for the synthesis of vinylphosphonate-linked nucleic acid. Tetrahedron Lett. 2000; 41:4513–4517. [Google Scholar]

[B24] 24. Garcia P.L., Bradley G., Hayes C.L., Krintel S., Soultanas P., Janscak P.. RPA alleviates the inhibitory effect of vinylphosphonate internucleotide linkage on DNA unwinding by BLM and WRN helicases. Nucleic Acids Res. 2004; 32:3771–3778. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B25] 25. Doddridge Z., Bertram R.D., Hayes C.L., Soultanas P.. Effect of vinylphosphonate internucleotide linkages on the cleavage specificity of exonuclease III and on the activity of DNA polymerase I. Biochemistry. 2003; 42:3239–3246. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B26] 26. Starrett J.E. Jr, Tortolani D.R., Russell J., Hitchcock M.J., Whiterock V., Martin J.C., Mansuri M.M.. Synthesis, oral bioavailability determination, and in vitro evaluation of prodrugs of the antiviral agent 9-[2-(phosphonomethoxy)ethyl]adenine (PMEA). J. Med. Chem. 1994; 37:1857–1864. [DOI] [PubMed] [Google Scholar]

[B27] 27. Navé J.F., Taylor D., Tyms S., Kenny M., Eggenspiller A., Eschbach A., Dulworth J., Brennan T., Piriou F., Halazy S.. Synthesis, antiviral activity and enzymatic phosphorylation of 9-phosphonopentenyl derivatives of guanine. Antiviral Res. 1995; 27:301–316. [DOI] [PubMed] [Google Scholar]

[B28] 28. Lima W.F., Prakash T.P., Murray H.M., Kinberger G.A., Li W., Chappell A.E., Li C.S., Murray S.F., Gaus H., Seth P.P.et al.. Single-stranded siRNAs Activate RNAi in Animals. Cell. 2012; 150:883–894. [DOI] [PubMed] [Google Scholar]

[B29] 29. Matsuda S., Keiser K., Nair J.K., Charisse K., Manoharan R.M., Kretschmer P., Peng C.G., Kel’in A.V., Kandasamy P., Willoughby J.L.S.et al.. siRNA conjugates carrying sequentially assembled trivalent N-acetylgalactosamine linked through nucleosides elicit robust gene silencing in vivo in hepatocytes. ACS Chem. Biol. 2015; 10:1181–1187. [DOI] [PubMed] [Google Scholar]

[B30] 30. Prakash T.P., Kinberger G.A., Murray H.M., Chappell A., Riney S., Graham M.J., Lima W.F., Swayze E.E., Seth P.P.. Synergistic effect of phosphorothioate, 5′-vinylphosphonate and GalNAc modifications for enhancing activity of synthetic siRNA. Bioorg. Med. Chem. Lett. 2016; 26:2817–2820. [DOI] [PubMed] [Google Scholar]

[B31] 31. Parmar R., Willoughby J.L., Liu J., Foster D.J., Brigham B., Theile C.S., Charisse K., Akinc A., Guidry E., Pei Y.et al.. 5′-(E)-Vinylphosphonate: a stable phosphate mimic can improve the RNAi activity of siRNA-GalNAc conjugates. ChemBioChem. 2016; 17:985–989. [DOI] [PubMed] [Google Scholar]

[B32] 32. Elkayam E., Parmar R., Brown C.R., Willoughby J.L., Theile C.S., Manoharan M., Joshua-Tor L.. siRNA carrying an (E)-vinylphosphonate moiety at the 5′-end of the guide strand augments gene silencing by enhanced binding to human Argonaute-2. Nucleic Acids Res. 2017; 45:3528–3536. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B33] 33. Prakash T.P., Lima W.F., Murray H.M., Li W., Kinberger G.A., Chappell A.E., Gaus H., Seth P.P., Bhat B., Crooke S.T.et al.. Identification of metabolically stable 5′-phosphate analogs that support single strand siRNA activity. Nucleic Acids Res. 2015; 43:2993–3011. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B34] 34. Haraszti R.A., Roux L., Coles A.H., Turanov A.A., Alterman J.F., Echeverria D., Godinho B.M.D.C., Aronin N., Khvorova A.. 5′-Vinylphosphonate improves tissue accumulation and efficacy of conjugated siRNAs in vivo. Nucleic Acids Res. 2017; 45:7581–7592. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B35] 35. Parmar R.G., Brown C.R., Matsuda S., Willoughby J.L.S., Theile C.S., Charissé K., Foster D.J., Zlatev I., Jadhav V., Maier M.A.et al.. Facile synthesis, geometry, and 2′-substituent-dependent in vivo activity of 5′-(E)- and 5′-(Z)-vinylphosphonate-modified siRNA conjugates. J. Med. Chem. 2018; 61:734–744. [DOI] [PubMed] [Google Scholar]

[B36] 36. DiFiglia M., Sapp E., Chase K., Schwarz C., Meloni A., Young C., Martin E., Vonsattel J.P., Carraway R., Reeves S.A.et al.. Huntingtin is a cytoplasmic protein associated with vesicles in human and rat brain neurons. Neuron. 1995; 14:1075–1081. [DOI] [PubMed] [Google Scholar]

[B37] 37. Biscans A., Coles A., Haraszti R., Echeverria D., Hassler M., Osborn M., Khvorova A.. Diverse lipid conjugates for functional extra-hepatic siRNA delivery in vivo. Nucleic Acids Res. 2019; 47:1082–1096. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B38] 38. Abbas S., Bertram R.D., Hayes C.J.. Commercially available 5′-DMT phosphoramidites as reagents for the synthesis of vinylphosphonate-linked oligonucleic acids. Org. Lett. 2001; 3:3365–3367. [DOI] [PubMed] [Google Scholar]

[B39] 39. Allerson C.R., Sioufi N., Jarres R., Prakash T.P., Naik N., Berdeja A., Wanders L., Griffey R.H., Swayze E.E., Bhat B.. Fully 2′-modified oligonucleotide duplexes with improved in vitro potency and stability compared to unmodified small interfering RNA. J. Med. Chem. 2005; 48:901–904. [DOI] [PubMed] [Google Scholar]

[B40] 40. Foster D.J., Brown C.R., Shaikh S., Trapp C., Schlegel M.K., Qian K., Sehgal A., Rajeev K.G., Jadhav V., Manoharan M.et al.. Advanced siRNA designs further improve in vivo performance of GalNAc-siRNA conjugates. Mol. Ther. 2018; 26:708–717. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B41] 41. Hirao T., Masunaga T., Ohshiro Y., Agawa T.. Reduction of gem-dibromides with diethyl phosphite. J. Org. Chem. 1981; 46:3745–3747. [Google Scholar]

[B42] 42. Hassler M.R., Turanov A.A., Alterman J.F., Haraszti R.A., Coles A.H., Osborn M.F., Echeverria D., Nikan M, Salomon W.F., Roux L.et al.. Comparison of partially and fully chemically-modified siRNA in conjugate-mediated delivery in vivo. Nucleic Acids Res. 2018; 46:2185–2196. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B43] 43. Ly S., Navaroli D.M., Didiot M.C., Cardia J., Pandarinathan L., Alterman J.F., Fogarty K., Standley C., Lifshitz L.M.et al.. Visualization of self-delivering hydrophobically modified siRNA cellular internalization. Nucleic Acids Res. 2017; 45:15–25. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B44] 44. Nair J.K., Willoughby J.L.S., Chan A., Charisse K., Alam M.R., Wang Q., Hoekstra M, Kandasamy P., Kel’in A.V., Milstein S.et al.. Multivalent N-acetylgalactosamine-conjugated siRNA localizes in hepatocytes and elicits robust RNAi-mediated gene silencing. J. Am. Chem. Soc. 2014; 136:16958–16961. [DOI] [PubMed] [Google Scholar]

[B45] 45. Østergaard M.E., Yu J., Kinberger G.A., Wan W.B., Migawa M.T., Vasquez G., Schmidt K., Gaus H.J., Murray H.M., Low A.et al.. Effcient synthesis and biological evaluation of 5′-GalNAc conjugated antisense oligonucleotides. Bioconjug. Chem. 2015; 26:1451–1455. [DOI] [PubMed] [Google Scholar]

[B46] 46. Rajeev K.G., Nair J.K., Jayaraman M., Charisse K., Taneja N., O'Shea J., Willoughby J.L., Yucius K., Nguyen T., Shulga-Morskaya S.et al.. Hepatocyte-specific delivery of siRNAs conjugated to novel non-nucleosidic trivalent N-acetylgalactosamine elicits robust gene silencing in vivo. Chembiochem. 2015; 16:903–908. [DOI] [PubMed] [Google Scholar]

[B47] 47. Alterman J.F., Godinho B.M.D.C., Hassler M.R., Ferguson C.M., Echeverria D., Sapp E., Haraszti R.A., Coles A.H., Conroy F., Miller R.et al.. A divalent siRNA chemical scaffold for potent and sustained modulation of gene expression throughout the central nervous system. Nat. Biotechnol. 2019; 37:884–894. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B48] 48. Alterman J.F., Hall L.M., Coles A.H., Hassler M.R., Didiot M.C., Chase K., Abraham J., Sottosanti E., Johnson E., Sapp E.et al.. Hydrophobically modified siRNAs silence Hutingtin mRNA in primary neurons and mouse brain. Mol. Ther. Nucleic Acids. 2015; 4:e266. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B49] 49. Prakash T.P., Allerson C.R., Dande P., Vickers T.A., Sioufi N., Jarres R., Baker B.F., Swayze E.E., Griffey R.H., Bhat B.. Positional effect of chemical modifications on short interference RNA activity in mammalian cells. J. Med. Chem. 2005; 48:4247–4253. [DOI] [PubMed] [Google Scholar]

[B50] 50. Jinek M., Coyle S.M., Doudna J.A.. Coupled 5′ nucleotide recognition and processivity in Xrn1-mediated mRNA decay. Mol. Cell. 2011; 41:600–608. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B51] 51. Stovall G.M., Bedenbaugh R.S., Singh S., Meyer A.J., Hatala P.J., Elliington A.D., Hall B.. In vitro selection using modified or unnatural nucleotides. Curr. Protoc. Nucleic Acid Chem. 2014; 56:9.6.1. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B52] 52. Ross C.A., Aylward E.H., Wild E.J., Langbehn D.R., Long J.D., Warner J.H., Scahill R.I., Leavitt B.R., Stout J.C., Paulsen J.S.et al.. Huntington disease: natural history, biomarkers and prospects for therapeutics. Nat. Rev. Neurol. 2014; 10:204–216. [DOI] [PubMed] [Google Scholar]

[B53] 53. Pfister E.L., Zamore P.D.. Huntington's disease: silencing a brutal killer. Exp. Neurol. 2009; 220:226–229. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B54] 54. Pfister E.L., Kennington L., Straubhaar J., Wagh S., Liu W., DiFiglia M., Landwehrmeyer B., Vonsattel J-P., Zamore P.D., Aronin N.. Five siRNAs targeting three SNPs in Huntingin may provide therapy for three-quarters of Huntington's disease patients. Curr. Biol. 2009; 19:774–778. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B55] 55. von Hippel P. H. From “Simple” DNA-protein interactions to the macromolecular machines of gene expression. Annu. Rev. Biophys. Biomol. Struct. 2007; 36:79–105. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B56] 56. Tabrizi S.J., Leavitt B.R., Landwehrmeyer G.B., Wild E.J., Saft C., Barker R.A., Blair N.F., Craufurd D., Priller J., Rickards H.et al.. Phase 1–2a IONIS-HTTRx study site teams, targeting Huntingtin expression in patients with Huntington's disease. N. Engl. J. Med. 2019; 380:2307–2316. [DOI] [PubMed] [Google Scholar]

[B57] 57. Chiu Y.L., Rana T.M.. siRNA function in RNAi: a chemical modification analysis. RNA. 2003; 9:1034–1048. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B58] 58. Schirle N.T., Sheu-Gruttadauria J., MacRae I.J.. Structural basis for MicroRNA targeting. Science. 2014; 346:608–613. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B59] 59. Preall J.B., Sontheimer E.J.. RNAi: RISC gets loaded. Cell. 2005; 123:543–545. [DOI] [PubMed] [Google Scholar]

[B60] 60. Kawamata T., Tomari Y.Making RISC. . Trends Biochem. Sci. 2010; 35:368–376. [DOI] [PubMed] [Google Scholar]

[B61] 61. Salomon W.E., Jolly S.M., Moore M.J., Zamore P.D., Serebrov V.. Single-molecule imaging reveals that Argonaute reshapes the binding properties of its nucleic acid guides. Cell. 2015; 162:84–95. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B62] 62. Schirle N.T., Kinberger G.A., Murray H.F., Lima W.F., Prakash T.P., MacRae I.J.. Structural analysis of human Argonaute-2 bound to a modified siRNA guide. J. Am. Chem. Soc. 2016; 138:8694–8697. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B63] 63. Wang Y., Sheng G., Juranek S., Tuschl T., Patel D.J.. Structure of the guide-strand-containing argonaute silencing complex. Nature. 2008; 456:209–213. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B64] 64. Schlegel M.K., Foster D.J., Kel’in A.V., Zlatev I., Bisbe A., Jayaraman M., Lackey J.G., Rajeev K.G., Charisse K., Harp J.et al.. Chirality dependent potency enhancement and structural impact of glycol nucleic acid modification on siRNA. J. Am. Chem. Soc. 2017; 139:8537–8546. [DOI] [PubMed] [Google Scholar]

[B65] 65. Kenski D.M., Cooper A.J., Li J.J., Willingham A.T., Haringsma H.J., Young T.A., Kuklin N.A., Jones J.J., Cancilla M.T., McMasters E.R.et al.. Analysis of acyclic nucleoside modifications in siRNAs finds sensitivity at position 1 that is restored by 5′-terminal phosphorylation both in vitro and in vivo. Nucleic Acids Res. 2010; 38:660–671. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B66] 66. Ma J.B., Ye K., Patel D.J.. Structural basis for overhang- specific small interfering RNA recognition by the PAZ domain. Nature. 2004; 429:318–322. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B67] 67. Østergaard M.E., Kumar P., Nichols J., Watt A., Sharma P.K., Nielsen P., Seth P.P.. Allele-selective inhibition of mutant Huntingtin with 2-thio- and C5- triazolylphenyl-deoxythymidine-modified antisense oligonucleotides. Nucleic Acid Ther. 2015; 25:266–274. [DOI] [PubMed] [Google Scholar]

[B68] 68. Østergaard M.E., Nichols J., Dwight T.A., Lima W., Jung M.E., Swayze E.E., Seth P.P.. Fluorinated nucleotide modifications modulate allele selectivity of SNP-targeting antisense oligonucleotides. Mol. Ther. Nucleic Acids. 2017; 7:20–30. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B69] 69. Janas M.M., Schlegel M.K., Harbison C.E., Yilmaz V.O., Jiang Y., Parmar R., Zlatev I., Castoreno A., Xu H., Shulga-Morskaya S.et al.. Selection of GalNAc-conjugated siRNAs with limited off-target-driven rat hepatotoxicity. Nat. Commun. 2018; 9:723. [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Structurally constrained phosphonate internucleotide linkage impacts oligonucleotide-enzyme interaction, and modulates siRNA activity and allele specificity

Ken Yamada

Samuel Hildebrand

Sarah M Davis

Rachael Miller

Faith Conroy

Ellen Sapp

Jillian Caiazzi

Julia F Alterman

Loic Roux

Dimas Echeverria

Matthew R Hassler

Edith L Pfister

Marian DiFiglia

Neil Aronin

Anastasia Khvorova

Abstract

Graphical Abstract

Graphical Abstract.

INTRODUCTION

Figure 1.

MATERIALS AND METHODS

General experimental methods

Synthesis of iE-VP-linked dimer phosphoramidites

3-N-Benzoyl-5′-O-[bis(4-methoxyphenyl)phenylmethyl]-2′-deoxy-2′-fluoro-3′-O-(tert-butyldiphenylsilyl)-uridine (2a)

3-N-Benzoyl-2′-deoxy-2′-fluoro-3′-O-(tert-butyldiphenylsilyl)-uridine (3a)

6′-Dibromo-(5′-deoxy-5′-methylidene)-3′-O-(tert-butyldiphenylsilyl)-2′-deoxy-2′-fluoro-uridine (5a)

(E)/(Z)-6′-Bromo-(5′-deoxy-5′-methylidene)-2′-deoxy-2′-fluoro-3′-O-(tert-butyldiphenylsilyl)-uridine (6a-E and 6a-Z)

5′-O-[Bis(4-methoxyphenyl)phenylmethyl]-2′-O-methyl-uridyl-methylphosphono-[3′(O)→5′(C)]-3-N-benzoyl-5′-deoxy-5′-methylidene-3′-O-(tert-butyldiphenylsilyl)-2′-deoxy-2′-fluoro-uridine (7a)

5′-O-[Bis(4-methoxyphenyl)phenylmethyl]-2′-O-methyl-uridyl-methylphosphono-[3′(O)→5′(C)]-3-N-benzoyl-2′,5′-dideoxy-2′-fluoro-5′-methylidene-uridine(8a)

5′-O-[Bis(4-methoxyphenyl)phenylmethyl]-2′-O-methyl-uridyl-methylphosphono-[3′(O)→5′(C)]-3-N-benzoyl-2′,5′-dideoxy-2′-fluoro-methylidene-uridine, 3′-[2-cyanoethyl-N,N-bis(1-methylethyl)- phosphoramidite] (9a)

3-N-Benzoyl-2′-O-methyl-3′-O-(tert-butyldiphenylsilyl)-uridine (3b)

6′-Dibromo-(5′-deoxy-5′-methylidene)-2′-O-methyl-3′-O-(tert-butyldiphenylsilyl)-uridine (5b)

(E)/(Z)-6′-Bromo-(5′-deoxy-5′-methylidene)-2′-O-methyl-3′-O-(tert-butyldiphenylsilyl)-uridine (6b-E and 6b-Z)

5′-O-[Bis(4-methoxyphenyl)phenylmethyl]-2′-deoxy-2′-fluoro-uridyl methylphosphono-[3′(O)→5′(C)] -3-N-benzoyl-5′-deoxy-5′-methylidene-3′-O-(tert-butyldiphenylsilyl)-2′-O-methyl-uridine (7b)

5′-O-[Bis(4-methoxyphenyl)phenylmethyl]-2′-deoxy-2′-fluoro-uridyl methylphosphono-[3′(O)→5′(C)]-3-N-benzoyl-5′-deoxy-5′-methylidene-2′-O-methyl-uridine (8b)

5′-O-[Bis(4-methoxyphenyl)phenylmethyl]-2′-deoxy-2′-fluoro-uridyl methylphosphono-[3′(O)→5′(C)]-3-N-benzoyl-5′-deoxy-5′-methylidene-2′-O-methyl-uridine, 3′-[2-cyanoethyl-N,N-bis(1-methylethyl)- phosphoramidite] (9b)

5′-O-[Bis(4-methoxyphenyl)phenylmethyl]-2-N-isobutylyl-2′-O-methyl-guanyl methylphosphono-[3′(O)→5′(C)]-5′-deoxy-5′-methylidene-3′-O-(tert-butyldiphenylsilyl)-3-N-benzoyl-2′-deoxy-2′-fluoro-uridine (7c)

5′-O-[Bis(4-methoxyphenyl)phenylmethyl]-2-N-isobutylyl-2′-O-methyl-guanyl methylphosphono-[3′(O)→5′(C)]-3-N-benzoyl-2′,5′-dideoxy-5′-methylidene-2′-fluoro-uridine (8c)

5′-O-[Bis(4-methoxyphenyl)phenylmethyl]-2-N-isobutyryl-2′-O-methyl-guanyl-methylphosphono-[3′(O)→5′(C)]-3-N-benzoyl-2′,5′-dideoxy-2′-fluoro-5′-methylidene-uridine, 3′-[2-cyanoethyl N,N-bis(1-methylethyl)- phosphoramidite] (9c)

Synthesis of inter-nucleotide (E)-vinylphosphonate modified RNA oligonucleotides

General procedures for nuclease digestion tests

Terminator 5′-P dependent 5′-exonuclease stability test

BSP stability test

SVPD stability test

Primer extension assays

Thermo stability assay

In vitro screen

siRNA passive delivery

Quantitative analysis of target mRNA

Creating bar graph

Reporter plasmid preparation

Cell treatment and reporter assay

Evaluation of compounds efficacy and discriminative power in vivo in BAC-HD mice

Automatic western detection of mutant and wt huntingtin

RESULTS

Synthesis of iE-VP-dimer phosphoramidites

Scheme 1.

Oligonucleotide synthesis

i E-VP incorporation can positively or negatively impact oligonucleotide-enzyme interactions

Figure 2.

Design and synthesis of chemically-modified siRNAs

siRNA conjugate and chemical modification pattern design

Figure 3.

siRNA sequence design

Table 1.

Synthesis of siRNA panel

Position-dependent impact of iE-VP incorporation on siRNA efficacy

Stereo-constraint of the backbone with iE-VP enhances SNP-based discrimination

Figure 4.

Stereo-constraint of the backbone with iE-VP enhances SNP-based discrimination in vivo in BAC-HD mice

Figure 5.

DISCUSSION

DATA AVAILABILITY

Supplementary Material

ACKNOWLEDGEMENTS

Notes

Contributor Information

SUPPLEMENTARY DATA

Synthesis of ⁱE-VP-linked dimer phosphoramidites

Synthesis of ⁱE-VP-dimer phosphoramidites

ⁱ E-VP incorporation can positively or negatively impact oligonucleotide-enzyme interactions

Position-dependent impact of ⁱE-VP incorporation on siRNA efficacy

Stereo-constraint of the backbone with ⁱE-VP enhances SNP-based discrimination

Stereo-constraint of the backbone with ⁱE-VP enhances SNP-based discrimination in vivo in BAC-HD mice