Skip to main content
NCBI home page
ACS Medicinal Chemistry Letters logoLink to ACS Medicinal Chemistry Letters
. 2024 Jul 16;15(8):1250–1259. doi: 10.1021/acsmedchemlett.4c00140

2′-O-Alkyl-N3-Methyluridine Functionalized Passenger Strand Improves RNAi Activity by Modulating the Thermal Stability

Avijit Sahoo †, Shalini Gupta ‡, Gourav Das †, Atanu Ghosh ‡, Siddharam Shivappa Bagale §, Surajit Sinha ‡, Kiran R Gore †,*
PMCID: PMC11318005  PMID: 39140063

Abstract

graphic file with name ml4c00140_0009.jpg

Herein, we have demonstrated that the siRNA activity could be enhanced by incorporating the guide strand in the RISC complex through thermodynamic asymmetry caused by m3U-based destabilizing modifications. A nuclease stability study revealed that 2′-OMe-m3U and 2′-OEt-m3U modifications slightly improved the half-lives of siRNA strands in human serum. In the in vitro gene silencing assay, 2′-OMe-m3U modification at the 3′-overhang and cleavage site of the passenger strand in anti-renilla and anti-Bcl-2 siRNA duplexes were well-tolerated and exhibited improved gene silencing activity. However, gene silencing activity was attenuated when these modifications were incorporated at position 3 in the seed region of the antisense strand. The molecular modeling studies using these modifications at the seed region with the MID domain of hAGO2 explained that the 2′-alkoxy group makes steric interactions with the amino acid residues of the hAGO2 protein.

Keywords: Small interfering RNA, RNA interference, N3-methyluridine, Gene silencing, Nuclease resistance

In 1998, the discovery of RNA interference (RNAi) marked a significant finding as a naturally occurring defense mechanism for the invasion of target genes mediated by small interfering RNAs (siRNAs).1 siRNAs are 19–21 nucleotides double-stranded RNAs that emerged as powerful tools for silencing the expression of targeted genes.2 The RNA-induced silencing complex (RISC) assembly composed of the guide strand and catalytic argonaute protein cleaves target mRNA and prevents the formation of the target protein.3 The therapeutic platform based on RNA interference (RNAi) holds significant promise in addressing unmet medical needs.4 In the past decade, six RNAi-based therapeutics, including patisiran, givosiran, lumasiran, inclisiran, vutrisiran, and nedosiran have received clinical approval against various diseases.5−7 At the same time, numerous siRNA drug candidates are in different phases of clinical trials. The success of these siRNA drugs was attributed to the rational use of appropriate chemical modifications and efficient delivery platforms.8 In the development of these therapeutic siRNAs, ribose modifications like 2′-deoxy, 2′-fluoro (2′-F), and 2′-O-methyl (2′-OMe), in conjunction with phosphorothioate (PS) linkages at the 5′- and 3′-termini, provide the necessary specificity and metabolic stability results in excellent safety and efficacy profiles for patients.9

So far, sugar and backbone modifications have been commonly employed to enhance the nuclease stability, silencing efficiency, and drug-like properties of these therapeutically appealing oligonucleotides.10−12 However, the exploration of base-modified siRNAs has been limited and primarily remains a subject of research. In 2003, Y. Chiu and T. Rana evaluated RNAi activity using anti-EGFP siRNA duplexes containing 5-bromouridine, 5-iodouridine, 2,6-diaminopurine, and N3-methyluridine modifications. The RNAi activity was abolished when m3U was placed at position 11 in the guide strand.13 Because the safety of the non-natural residues in metabolism remains uncertain, leading to a preference for naturally occurring nucleobases in therapeutic applications. Numerous natural base modifications, such as N6-methyladenosine (m6A), N7-methylguanosine (m7G), 5-methylcytidine (m5C), pseudouridine (Ψ), N3-methylcytidine (m3C), and N3-methyluridine (m3U), play vital roles in cellular processes like mRNA stability, splicing, and localization.14 Recent studies demonstrated that m6A-modified siRNA minimized off-target effects and evaded immune response without disturbing RNAi activity.15,16

Combining stabilizing and destabilizing chemical modification in siRNAs through enhanced stabilization chemistry plus (ESC+) platform has achieved promising gene silencing and mitigated the off-target-mediated toxicity.17,18 The xenobiotic nucleic acid (XNA) modifications, including glycol nucleic acid (GNA) and unlocked nucleic acid (UNA), have exhibited improved guide strand selection in the RISC complex by introducing thermodynamic asymmetry into the siRNA duplex.19,21 Also, the introduction of 5-nitroindole modification at position 15 of the passenger strand has reduced passenger strand mediate off-target effects.20 Recently, it was reported that GNA modification at position 7 in the guide strand mitigated off-target mediated hepatotoxicity in a rodent model.17 These studies demonstrate that the rational use of destabilizing chemical modifications could enhance the safety profile of siRNAs.21

Herein, we assessed the thermodynamic stability, resistance to nuclease degradation, and gene-silencing activity of 2′-O-methyl-N3-methyluridine and 2′-O-ethyl-N3-methyluridine modified siRNAs (Figure 1). The primary aim of this study was to explore how the modulation of the thermodynamic profile through the incorporation of thermally destabilizing modifications in both the passenger and guide strands influence the siRNA potency.

Figure 1.

Figure 1

Open in a new tab

(A) siRNA sequence targeting luciferase mRNA; (B) Mono and dual modifications incorporated in siRNA: Uridine (U), 2′-O-methyluridine (2′-OMe-U), 2′-deoxy-2′-fluorouridine (2′-F-U), 2′-O-methyl-N3-methyluridine (2′-OMe-m3U), 2′-O-ethyl-N3-methyluridine (2′-OEt-m3U).

Our laboratory has reported the synthesis of a series of 2′-alkoxy/fluoro-N3-methyluridine nucleosides including 2′-F-m3U, 2′-OMe-m3U, 2′-OEt-m3U, 2′-OPr-m3U, and 2′-OMOE-m3U.22,23 The incorporation of these modified monomers at the central position in 12-mer and 14-mer oligonucleotides exhibited a substantial reduction in melting temperature (around 8–12 °C per modification). Moreover, nuclease resistance studies showed that 2′-O-alkyl-m3U modifications improved the half-life of oligonucleotides against 3′- and 5′- exonucleases as compared to 2′-fluoro and 2′-OMe modified oligonucleotides. This unique characteristic prompted us to investigate the compatibility of these nucleoside analogues with the RNA-induced silencing complex (RISC) and possess the ability to regulate gene expression through the RNAi pathway. We envisioned that the destabilizing 2′-O-alkyl-m3U modification could improve the loading of the intended guide strand in the RISC complex through thermodynamic asymmetry, leading to improved RNAi activity. In this report, we have investigated the effect of 2′-OMe-m3U and 2′-OEt-m3U modifications on hybridization properties, nuclease stability, and RNAi activity when incorporated at the overhang, seed region, and cleavage site of siRNA duplex.

The 2′-OMe-m3U, 2′-OEt-m3U, 2′-OMe-U, and 2′-F-U phosphoramidites were synthesized employing our previously established protocols.23 The siRNAs targeting Renilla luciferase and endogenous Bcl-2 genes were synthesized by incorporating these modifications at overhangs, seed regions, and cleavage sites using an automated DNA/RNA synthesizer following the conventional phosphoramidite protocols. The siRNA sequences are highlighted in Table S1. Initially, the influence of these modifications on the stability of siRNA duplexes was assessed through UV-thermal melting studies. Tm curves for all duplexes are depicted in Figure S1. Compared to the unmodified anti-renilla siRNA duplex (si-1), the melting temperature consistently decreased upon the single incorporation of 2′-alkoxy-m3U monomers, as outlined in Table 1. Introducing a single monomer (2′-OMe-m3U/2′-OEt-m3U) at the overhang position of the passenger strand led to a Tm decrease of approximately 1 °C (si-2, si-3), while either of these modifications at the seed region of the guide strand resulted in a reduction in thermal stability by approximately 2.7–2.9 °C (si-4, si-5). Meanwhile, the 2′-OMe-m3U modification at the central position of the passenger strand of siRNA duplexes, significantly decreased the Tm by 8 °C (si-6).

Table 1. Sequences and Relative Melting Temperatures (Tm) of Modified siRNAs Targeting Renilla Luciferase and Bcl-2 Genea.

siRNA Strand code siRNA sequences Tm(°C) ΔTm(°C)
si-1 1P 5′-GGCCUUUCACUACUCCUACTT-3′ 78.5 ± 0.2 —
6G 3′-TTCCGGAAAGUGAUGAGGAUGA-5′
si-2 2P 5′-GGCCUUUCACUACUCCUACX1T-3′ 77.5 ± 0.5 –1.0
6G 3′-TTCCGGAAAGUGAUGAGGAUGA-5′
si-3 3P 5′-GGCCUUUCACUACUCCUACX2T-3′ 77.4 ± 0.5 –1.1
6G 3′-TTCCGGAAAGUGAUGAGGAUGA-5′
si-4 1P 5′-GGCCUUUCACUACUCCUACTT-3′ 75.8 ± 0.4 –2.7
7G 3′-TTCCGGAAAGUGAUGAGGAX1GA-5′
si-5 1P 5′-GGCCUUUCACUACUCCUACTT-3′ 75.6 ± 0.5 –2.9
8G 3′-TTCCGGAAAGUGAUGAGGAX2GA-5′
si-6 4P 5′-GGCCUUUCACX1ACUCCUACTT-3′ 70.5 ± 0.4 –8.0
6G 3′-TTCCGGAAAGUGAUGAGGAUGA-5′
si-8 5P 5′-GGCCUfUfUfCACUfACUfCCUfACX1T-3′ 83.6 ± 0.6 +5.1
6G 3′-TTCCGGAAAGUGAUGAGGAUGA-5′
si-9 5P 5′-GGCCUfUfUfCACUfACUfCCUfACX1T-3′ 80.1 ± 0.7 +1.6
7G 3′-TTCCGGAAAGUGAUGAGGAX1GA-5′
si-B1 13P 5′-GUGAAGUCAACAUGCCUGCTT-3′ 75.7 ± 0.1 —
16G 3′-TTCACUUCAGUUGUACGGACG-5′
si-B2 14P 5′-GUGAAGUCAACAUGCCUGX1TT-3′ 67.6 ± 0.5 –8.1
16G 3′-TTCACUUCAGUUGUACGGACG-5′
si-B3 15P 5′-GUGAAGUCAAX1AUGCCUGCTT-3′ 65.4 ± 0.3 –10.3
16G 3′-TTCACUUCAGUUGUACGGACG-5′
Open in a new tab
a

Tm values were recorded using an oligo concentration of 1.0 μM in a buffer containing 10 mM sodium phosphate, 100 mM NaCl, and 0.1 mM EDTA (pH 7.4). ΔTm = (Tm of modified duplex – Tm of corresponding unmodified duplex). Tm values were calculated as the maximum of the first derivative of the melting curve (A260 vs Temp.) and reported as the average of three independent measurements with the estimated standard deviation. Modified nucleotides are represented as follows: 2′-OMe-m3U (X1), 2′-OEt-m3U (X2), 2′-F-U (Uf).

Similarly, the Tm of modified anti-Bcl-2 siRNA duplexes was evaluated by incorporating 2′-OMe-m3U modification at the cleavage site and near the 3′-overhang of the passenger strand (si-B3 and si-B2). We observed a similar trend of melting temperature as observed for anti-renilla siRNAs (Table 1). The observed lower Tm value trend suggests that the N3-methyl group is removing the H-bond donor, potentially disrupting the Watson–Crick hydrogen bonding interactions and intrastrand stacking (see molecular modeling section). This observation correlates with the previously reported m3U or m3C modifications in RNA 14-mer or 12-mer sequences, leading to the weakening of hydrogen bonds, as evidenced by an increase in both average bond distance and fluctuations.14,23

Furthermore, the Tm of siRNA duplexes was increased by 5.1 and 1.6 °C when 2′-OMe-m3U modifications were combined with multiple 2′-F-U modifications in the passenger strand (si-8 and si-9, respectively). Since the 2′-F modification is known to enhance binding affinity in RNA duplexes, these results indicate that incorporating 2′-alkoxy-m3U modifications, along with other 2′-modifications, may effectively counteract the thermal destabilization of siRNA duplexes.12,24

Subsequently, we conducted the CD spectroscopy of siRNA duplexes to compare the global structures of the siRNAs containing these modifications. Despite the significant variations in thermal melting, our analysis of modified siRNA duplexes using CD spectroscopy indicated no significant perturbation in the global duplex geometry (Figure S2). All duplexes, whether modified or unmodified, exhibited a CD maxima around 260 nm, a more pronounced minima at 210 nm, and a less intense minima around 240 nm, indicative of A-type RNA helices. This result suggests that these modifications can effectively modulate thermal stability while preserving the overall integrity of siRNA duplex structures.

One of the primary challenges in nucleic acid-based therapeutics lies in their susceptibility toward endo- and exonucleases present in blood plasma and the cytosol.25 The rational use of chemical modifications in siRNA oligonucleotides proves to be effective in boosting their resistance against nucleases, resulting in prolonged gene silencing.26 Our previous report suggested that introducing a single 2′-alkoxy-m3U modification at either the 3′- or 5′-terminus of oligonucleotide, significantly enhances enzymatic stability in SVPD or PDE-II, respectively.23 This success prompted us to investigate the influence of these modifications on the nuclease-mediated degradation of the modified siRNA strands. To evaluate the impact of these modifications against exonuclease-mediated degradation, we subjected the 3′-penultimate modified passenger strands to snake venom phosphodiesterase (SVPD), a 3′-specific exonuclease.

The oligonucleotides were incubated at 37 °C, and the percentage of full-length products was monitored over 24 h by polyacrylamide gel electrophoresis (PAGE) (Figure S3). Our investigation unveiled that incorporating a single 2′-alkoxy-m3U monomer at the penultimate position of the passenger strand significantly extended the half-life of the oligonucleotide (Figure 2A), compared to the native siRNA strand (1P, t1/2 = 3.2 h). Notably, the half-life of the 2′-OEt-m3U modified passenger strand (3P, t1/2 > 24 h) was longer than that of the 2′-OMe-m3U modified strands (2P, t1/2 = 17 h), possibly due to an increased 2′-O-alkyl chain length. Meanwhile, the strand (5P) modified with 2′-OMe-m3U in combination with 2′-F-U exhibited slightly better half-lives. Subsequently, we investigated the effect of these modifications by incubating the preannealed duplexes (Table S2) in 10% human serum at 37 °C, followed by PAGE analysis (Figure S5). There was no significant improvement in nuclease resistance compared to the unmodified duplex when the duplexes consist of only 2′-alkoxy-m3U modification at the passenger strand or guide strand (si-3, si-6, and si-14 to si-16) (Figure 2B). Surprisingly, the half-lives of si-9 and si-17 were enhanced significantly (>24 h) when the passenger strand was modified extensively using 2′-F-uridines and single 2′-OMe-m3U at the overhang position. The overall increase in duplex binding affinity likely contributes to enhanced serum stability.27

Figure 2.

Figure 2

Open in a new tab

Decay graph of full-length RNA (%) versus incubation time for (A) Passenger strands incubated with 3′-specific exonuclease, SVPD (35 mU/mL) in a buffer containing 50 mM Tris-base (pH 7.2) and 10 mM MgCl2; (B) modified siRNA duplex incubated with 10% human serum in a buffer containing 10 mM sodium phosphate (pH 7.4) and 100 mM NaCl. The determination of half-life was accomplished by fitting the data to first-order kinetics, as demonstrated in Figures S4 and S6.

In recent years, there has been considerable exploration of the utilization of lipophilicity and ionic character to modulate the pharmacokinetic properties of oligonucleotide-based drugs. The lipophilicity of such drugs plays a crucial role in influencing tissue penetration and in vivo half-life.28 In this study, we evaluated the lipophilicity of a series of modified passenger strands (1P-5P) by analyzing their retention times on a C-18 reversed-phase HPLC column. Also, we investigated the impact of position-specific modifications on lipophilicity. All evaluated RNA oligonucleotides exhibited sharp peaks with a diverse range of retention times (Figure 3), signifying a substantial impact of 2′-alkoxy-m3U modifications on the overall lipophilicity of the siRNA.

Figure 3.

Figure 3

Open in a new tab

Normalized HPLC chromatogram of the siRNA passenger strand containing modifications at various positions.

Specifically, the incorporation of a single 2′-OMe-m3U modification at the overhang position of the passenger strand (2P) resulted in an approximate 1 min enhancement in elution time, whereas the incorporation of the 2′-OEt-m3U modification (3P) increased the retention time by approximately 2 min compared to the unmodified passenger strand (1P). Conversely, the placement of the 2′-OMe-m3U modification at the central position of the passenger strand (4P) resulted in an increase in elution time by only 0.7 min. Notably, combining the 2′-OMe-m3U monomer with multiple 2′-F-U functionalizations in the 5P oligo strand has a predominant impact on the overall lipophilicity of the siRNA molecule, and retention time increased by approximately 2.6 min. Consequently, the employment of these modifications in the siRNA strand holds significance for tuning the lipophilicity and ionic character of RNAi-based drugs.

Next, we assessed the effect of chemical modifications on gene silencing efficiency by targeting siRNAs against Renilla luciferase and Bcl-2 as endogenous gene targets. Initially, we transfected the anti-renilla luciferase siRNAs concomitantly with the psiCHECK2 plasmid in HeLa cells using Lipofectamine 2000. The dual luciferase reporter assay (DLRA) was used to analyze the expression of luciferase genes 48 h post-transfection. The scrambled (SC) siRNA was used as the negative control for this in vitro assay, while only the psiCHECK2 plasmid was utilized as the positive control (PC). Gene silencing efficiency was quantified by calculating the ratio of Renilla luciferase light units to firefly luciferase light units, normalized in comparison to the positive control.

In vitro silencing results indicate that the 2′-alkoxy-m3U modifications are well tolerated at different positions of passenger strands (Table 2). The 2′-OMe-m3U at the 3′-overhang of the passenger strand (si-2) was well tolerated and exhibited comparable RNAi activity as unmodified (si-1). In contrast, the 2′-OEt-m3U modification at the penultimate position of the passenger strand (si-3) slightly improved the RNAi activity compared to unmodified siRNA duplexes. However, the inclusion of a single 2′-alkoxy-m3U monomer at position 3 of the seed region in the guide strand (si-4, si-5) led to a significant decrease in RNAi activity (IC50 = 12.5 to 13.7 nM). As expected, the lower activity reflects the possible disruption of hAgo2-RNA interactions caused by the hindered 2′-alkoxy residues, along with the relatively weak binding affinity toward the target mRNA sequence induced by the N3-methyl group. The molecular modeling studies using seed region modified guide strand with MID domain of AGO2 protein demonstrated that the 2′-OMe or 2′-OEt groups make steric interactions with amino acid residues of MID domain which could lead to reduced RNAi activity. (see molecular modeling section below).

Table 2. In Vitro Potency of Modified siRNAs Targeting Renilla Luciferase Genea.

siRNA Strand code siRNA sequences Position of modification IC50(nM)
si-1 1P 5′-GGCCUUUCACUACUCCUACTT-3′ Unmodified 0.99 ± 0.16
6G 3′-TTCCGGAAAGUGAUGAGGAUGA-5′
si-2 2P 5′-GGCCUUUCACUACUCCUACX1T-3′ Overhang of passenger strand 0.78 ± 0.12
6G 3′-TTCCGGAAAGUGAUGAGGAUGA-5′
si-3 3P 5′-GGCCUUUCACUACUCCUACX2T-3′ Overhang of passenger strand 0.56 ± 0.17
6G 3′-TTCCGGAAAGUGAUGAGGAUGA-5′
si-4 1P 5′-GGCCUUUCACUACUCCUACTT-3′ Seed region of guide strand 12.45 ± 5.86
7G 3′-TTCCGGAAAGUGAUGAGGAX1GA-5′
si-5 1P 5′-GGCCUUUCACUACUCCUACTT-3′ Seed region of guide strand 13.74 ± 8.18
8G 3′-TTCCGGAAAGUGAUGAGGAX2GA-5′
si-6 4P 5′-GGCCUUUCACX1ACUCCUACTT-3′ Cleavage site of passenger strand <0.5
6G 3′-TTCCGGAAAGUGAUGAGGAUGA-5′
si-7 2P 5′-GGCCUUUCACUACUCCUACX1T-3′ Combination on both strands 29.44 ± 11.46
9G 3′-TUfCCGGAAAGUOMeGAUfGAGGAUOMeGA-5′
si-8 5P 5′-GGCCUfUfUfCACUfACUfCCUfACX1T-3′ Combination on passenger strand 0.75 ± 0.84
6G 3′-TTCCGGAAAGUGAUGAGGAUGA-5′
si-9 5P 5′-GGCCUfUfUfCACUfACUfCCUfACX1T-3′ Combination on both strands N.A.
7G 3′-TTCCGGAAAGUGAUGAGGAX1GA-5′
si-10 5P 5′-GGCCUfUfUfCACUfACUfCCUfACX1T-3′ Combination on both strands N.A.
9G 3′-TUfCCGGAAAGUOMeGAUfGAGGAUOMeGA-5′
(SC) 17P 5′-UAAGGCUAUGAAGAGAUACTT-3′ — N.A.
18G 3′-TTAUUCCGAUACUUCUCUAUGA-5′
Open in a new tab
a

In vitro RNAi efficacy (IC50) of unmodified and modified siRNAs was assessed 48 h post-transfection. HeLa cells were transfected with psiCHECK2 plasmid along with the siRNAs. The reported IC50 values represent the mean values ± SEM derived from three biological replicates. Modified nucleotides are as follows: 2′-OMe-m3U (X1), 2′-OEt-m3U (X2), 2′-F-U (Uf), 2′-OMe-U (UOMe). SC denotes scrambled siRNA. N.A. = not applicable. For fitted IC50 curves, see Figure S7.

Furthermore, the incorporation of single 2′-OMe-m3U modifications at positions 10 and 11 at the cleavage site of the passenger strand (si-6 and si-13) exhibited higher RNAi activity than the native siRNA duplex (Figure 4 and Figure S8). To validate the synergistic effect of the 2′-OMe-m3U modification at the cleavage site, we employed 2′-O-methyluridine modified siRNA duplexes (si-11 and si-12) as a control (Table S3). The silencing data indicate that 2′-O-methyluridine is well tolerated at the overhang (si-11) and the cleavage site (si-12) of the passenger strand, which has very similar activity to the native duplex. Whereas, 2′-OMe-m3U modification at the cleavage site of the passenger strand improved the RNAi activity around 2-fold. These findings indicate that introducing destabilizing modification at the cleavage site of the passenger strand facilitates effective gene silencing activity by regulating the RISC assembly.

Figure 4.

Figure 4

Open in a new tab

RNAi activity of unmodified and modified siRNAs after 48 h of incubation. (A) Sequence of the model siRNA, X represents the modified position; (B) Dose-dependent RNAi activity of siRNA duplexes. The concentrations used are 0.5, 1, 10, 20, 50, and 100 nM; (C) Relative gene silencing activity of duplexes at 10 nM concentration. HeLa cells were transfected with psiCHECK2 plasmid and siRNAs. Normalized luciferase activity was measured by dividing Renilla luciferase light units by Firefly luciferase light units and normalized with respect to the positive control (PC, no siRNA). SC denotes scrambled siRNA.

Furthermore, the combination of multiple 2′-OMe and 2′-F modifications at the seed region and cleavage site of the guide strand along with 2′-OMe-m3U modification at the 3′-overhang of the passenger strand (si-7) was tested. The RNAi activity was reduced considerably (IC50 = 29.4 nM) due to the presence of 2′-OMe moiety at the seed region and the cleavage site in the guide strand. However, when the passenger strand comprising several 2′-F modifications and 2′-OMe-m3U at 3′-overhang along with the native guide strand (si-8) was transfected, it demonstrated similar activity as the unmodified siRNA duplex (si-1). Notably, the other two modified siRNA duplexes (si-9 and si-10), combining multiple 2′-F and 2′-OMe modifications with our dual modification, also significantly diminished RNAi activity. The reduced RNAi activity for si-9 and si-10 could be due to steric clashes of 2′-OMe moiety at the seed region with amino acid residues of AGO2 protein.

After using a reporter system for in vitro gene silencing, we also tested the gene silencing activity using siRNAs modified with 2′-OMe-m3U against Bcl-2 as an endogenous chromosomal gene target in HeLa cells. Bcl-2 protein has emerged as a promising anticancer target due to its pivotal role in regulating cellular apoptosis.29 The 2′-OMe-m3U modification was incorporated at the cleavage site (si-B3) and first base-pair region of the 3′-overhang (si-B2) of the passenger strand (Table 1). The silencing data illustrated that at low concentration (25 nM), the unmodified siRNA (si-B1) showed poor activity; however, both chemically modified siRNAs, si-B2 and si-B3, exhibited better gene silencing (Figure 5). Here, si-B2 bearing 2′-OMe-m3U at the first base-pair position from 3′-end enhanced RNAi activity at 25 nM and 50 nM concentrations. Moreover, the modification at the cleavage site (si-B3) was also well tolerated and exhibited similar Bcl-2 protein knockdown as a native duplex. This increase in RNAi activity for si-B2 could be attributed to the thermodynamic asymmetry at the 5′-termini of the guide strand, which increases the RISC loading of the intended guide strand.30 In summary, our modification strategy demonstrated efficient gene silencing in exogenous and endogenous gene systems.

Figure 5.

Figure 5

Open in a new tab

(A) Sequence of siRNA duplex targeting Bcl-2 gene (Table 1), X= 2′-OMe-m3U; (B) Western Blot and (C) Densitometric estimation of Bcl-2 protein levels in HeLa cells after incubation with siRNAs (modified and unmodified). β-actin was used as a loading control, and data represent n = 3, mean ± SEM. Data was analyzed by One-way ANOVA (Dunett’s test), and p < 0.05 was considered significant. SC denotes scrambled siRNA.

A prevailing concern associated with chemically modified oligonucleotides is the potential toxicity if they were to be excised by any exonucleases and endonucleases present in the human body.31 Neverthless, our modification strategy contains the naturally modified m3U nucleobase, which appears to be nontoxic in the body. To address this concern, we conducted a cell viability assay for each modified nucleoside within the tested library, assessing concentrations up to 1.0 mM in HeLa cells. Notably, no significant reduction in cell viability was observed within this concentration range (Figure 6). However, it is important to emphasize that 1.0 mM concentration surpasses physiological conditions, where the typical concentration of bases and nucleosides in human plasma and other extracellular fluids ranges from 0.4 to 6 μM.32 Therefore, it is reasonably anticipated that these chemically modified nucleosides, if liberated during siRNA hydrolysis by nucleases and phosphatases, should not pose toxicity concerns for human cells.

Figure 6.

Figure 6

Open in a new tab

Normalized cell viability assay (MTT) over a range of tested concentrations. The experiments were performed in triplicates and data are presented as the mean ± SEM. HeLa cells were exposed to varying concentrations of each nucleoside analogue, ranging from 0.1 μM to 1000 μM.

Computational studies were carried out to investigate the effect of the 2′-OMe-m3U modification on the structural aspects of siRNA duplex as well as various interactions between modified siRNA and proteins including hAGO2 and 3′-exonucleases. MD simulations of 500 ns were carried out on the si-1 (unmodified) and si-6 (2′-OMe-m3U modified) duplexes using AMBER20.33 It was observed from the pseudorotation wheels (Figure S30) that both the unmodified (U11) and 2′-OMe-m3U (U*11) sugars predominantly maintain the North-type conformations during the simulations (P = 20.45° in the unmodified sugar; P = 16.84° in the modified sugar of the average structure) with glycosidic torsion angles of ∼-160.0° (Figure S31, Table S4). In the modified siRNA (si-6), 2′-OMe-m3U (U*11) and its complementary nucleotide (A31) adjust themselves to accommodate the N3-methyl group. Hence, we observed significant deviations in backbone torsion angles at the modified region in si-6 compared to si-1 (Table S4). Also, changes in the base-pair parameters were observed relative to the unmodified duplex (Table S5). As expected, the N3-methyl group significantly disrupts the Watson–Crick hydrogen bonds at the modified region (Figure 7, Table S6). Surprisingly, we observed noncanonical O4–C2 (14.61%) and canonical O4–N6 (6.95%) hydrogen bonds with the complementary nucleotides. Moreover, it was observed that the average C1′-C1′ distance and interstrand phosphate distance of the considered base-pair were increased by ∼0.80 Å and ∼0.13 Å, respectively, in the presence of the modified nucleotide (Table S7).

Figure 7.

Figure 7

Open in a new tab

Major cluster representative structures of the (A) unmodified and (B) 2′-OMe-m3U modified siRNA duplexes from the 500 ns MD trajectory; 2′-OMe-m3U is denoted by U*11.

Furthermore, the modification alters the base-pair step parameters at the modified region (Table S8), leading to the disruption in intrastrand stacking interactions (Figure 7). Stacking energies were calculated for 2′-OMe-m3U (U*11) and its complementary nucleotide (A31) from the major cluster average structures obtained from the simulation. The trinucleotide segment containing the modification (shown in Figure S32) was extracted from the siRNA duplex for the calculations. It was observed that the intrastrand stacking interactions at the modified region were disturbed in si-6. The total intrastrand stacking energy in the trinucleotide system for si-1 is −20.96 kcal/mol, whereas it is −15.03 kcal/mol in the case of si-6 (Table S9). The backbone and all-atom RMSD values were also observed to increase for the modified duplex (Figure S33–S36). The backbone RMSD values for si-1 and si-6 are 3.78 Å (±0.79) and 3.87 Å (±1.13) respectively, whereas the all-atom RMSD values are 3.44 Å (±0.71) and 3.61 Å (±1.00) respectively. On the other hand, in the trinucleotide system, backbone RMSD values for si-1 and si-6 are calculated to be 0.86 Å (±0.19) and 1.13 Å (±0.41) respectively, while the all-atom RMSD values are 0.98 Å (±0.16) and 1.67 Å (±0.39) respectively. Overall, the disruption of W–C H-bonding and nucleobase stacking interactions, higher C1′-C1′ distance, and RMSD values validate the decrease in thermal melting temperature (Tm) of the siRNA duplex in the presence of the modification. However, the modified siRNA duplex was found to maintain the A-type geometry. It is interesting to note that the rational use of the destabilizing modification (2′-OMe-m3U) in combination with 2′-fluoro has improved the thermal stability of si-8 and si-9 (Table 1).

The interactions of the seed-region modified siRNA guide strand and the MID domain of human argonaute-2 (hAgo2) protein were studied using the crystal structure of hAgo2 in complex with miR-20a (PDB ID: 4F3T).34 The guide strand residues from the 5′-end were changed according to si-5, with the third position modified to 2′-OEt-m3U. All the sugar conformations were kept unaltered, where the initial conformation of the modified residue was C3′-endo. Thereafter, UCSF Chimera35 was used to minimize the hAgo2-siRNA complex. From the minimized structure (Figure 8C, 8D), it was observed that the modified sugar maintains the C3′-endo pucker state and the 2′-OEt group makes a steric clash with a cysteine residue (C793), having the closest distance of 3.30 Å. However, the N3-methyl group did not make any contact with the surrounding residues. A similar interaction was found in the case of 2′-OMe-m3U modified siRNA (si-4), where the 2′-OMe group makes a steric clash with the C793 residue (Figure 8A, 8B); the closest distance is 3.60 Å, which is less than the sum of the van der Waals radii of two methyl groups, i.e., 4.00 Å. The steric interactions exerted by the 2′-OMe and 2′-OEt groups might interrupt the proper binding of MID domain of the hAgo2 protein to the seed region of the modified siRNA. The observations help to understand the reduced activity of siRNAs (si-4 and si-5) in the presence of the 2′-modifications at the third position of the seed region (Table 2). The reduced activity of si-7, si-9, and si-10 can also be explained similarly.

Figure 8.

Figure 8

Open in a new tab

Minimized structure of (A, B) modified guide strand bound to the hAgo2-MID domain (PDB ID: 4F3T). The third position from the 5′-end was modified to 2′-OMe-m3U (U*); (C, D) modified guide strand bound to the hAgo2-MID domain (PDB ID: 4F3T). The third position from the 5′-end was modified to 2′-OEt-m3U (U*); (E, F) modified DNA fragment 3′-d(TPSTT)-5′ bound at the active site of DNA polymerase I Klenow fragment 3′–5′-exonuclease (PDB ID: 1KSP). The sulfur atom of the phosphorothioate bond was replaced by oxygen, whereas the 5′-end dT and the penultimate dT from the 3′-end was modified to cytidine and 2′-OEt-m3U (U*). The water molecules and most of the nonpolar hydrogen atoms were omitted for visual clarity. The black arrow indicates the closest distance between the substituent and the residues. The dashed line indicates CH-π interaction. The phosphorus, nitrogen, oxygen, and hydrogen atoms were highlighted in orange, blue, red, and white colors, respectively. The carbon atoms of the 2′-OEt and 2′-OMe groups are shown in green, whereas the N3-methyl group is shown in teal.

Subsequently, we investigated the interactions between the 2′-OEt-m3U modified oligonucleotide and the 3′-exonuclease. The study utilized the crystal structure of Escherichia coli DNA polymerase I Klenow fragment 3′–5′-exonuclease bound to a DNA trimer (PDB ID 1KSP).36 All the dT nucleotides in the trimer acquire C3′-endo conformation. The sulfur atom of the phosphorothioate bond was replaced with oxygen, and the penultimate dT from the 3′-end was modified to 2′-OEt-m3U (U*). Subsequently, the whole complex was minimized using UCSF Chimera.35 It was evident from the minimized structure (Figure 8E, 8F) that the 2′-OEt group is oriented toward a confined space limited by a Tyr residue (Y423); the closest distance is 3.40 Å, which is below the sum of the van der Waals radii, i.e., 4.00 Å. Interestingly, the methylene group can form CH-π interaction with the phenyl ring. Additionally, a leucine residue (L361) resides in close proximity to the N3-methyl-modified uracil base, with the nearest distance between the isobutyl group of the Leu residue and N3-methyl measuring 4.30 Å. Therefore, the N3-methyl group could prevent the proper binding of the 3′-exonuclease with the oligonucleotide. On the other hand, the closest distance between the 2′-OEt group and phosphodiester linkage is observed to be 4.60 Å. Overall, the steric clashes exerted by the 2′-OEt and N3-methyl groups with the active site residues as well as the close proximity of the 2′-OEt group to the phosphodiester linkage leads to the improved 3′-exonuclease stability of the modified siRNAs.

In conclusion, we have reported the interesting destabilizing modifications 2′-OEt-m3U and 2′-OMe-m3U in siRNAs. Modifications have exhibited a range of melting temperatures based on their position in the siRNA duplex. We also demonstrated that the thermodynamic stability could be modulated using a combination of our destabilizing modifications and 2′-fluoro modification. The nuclease resistance assay illustrated that the modified oligonucleotide improved the half-life considerably against 3′-exonuclease. Notably, gene silencing activity was increased when these modifications were incorporated near the 3′-overhang and positions 10, and 11 in the cleavage site of the passenger strand. The destabilizing modification at the 3′-overhang or cleavage site raised the RNAi activity significantly, possibly due to increasing guide strand selection in RISC assembly. However, the silencing activity dropped down when these modifications were placed at position 3 in the seed region of the guide strand. The molecular modeling studies revealed that H-bond occupancy and intrastrand stacking were hampered due to the presence of the N3-methyl group. Furthermore, 2′-O-alkyl moiety makes steric clashes with amino acid residues of the MID domain, which could be responsible for poor RNAi activity for seed-modified siRNA duplex. Nevertheless, these initial studies render the usefulness of our dual modifications in improving siRNA therapeutic properties. Moreover, future studies involving on-target vs off-target potency and immune response analysis are under progress in our research laboratory.

Acknowledgments

Authors acknowledge the Science and Engineering Research Board (SERB), Department of Science and Technology (DST), Govt. of India (DST-SERB CRG/2022/003222; dated 08.12/2022) for financial support. We thank the Indian Institute of Technology, Kharagpur, for providing the infrastructure and research facilities. We thank Dr. S. Harikrishna for his valuable suggestions in molecular modelling studies. The authors also acknowledge the Supercomputing Facility “PARAM Shakti” at IIT Kharagpur established under the National Supercomputing Mission (NSM), Government of India. A.S. and G.D. acknowledge the Indian Institute of Technology Kharagpur, for the fellowship. S.G. and A.G. thank CSIR, and S.S.B. acknowledges the Indian Institute of Technology Bombay, for the fellowship.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsmedchemlett.4c00140.

  • Experimental procedures, Tm graphs, CD spectra, gel images, fitted IC50 curves, HPLC chromatograms, MALDI-TOF mass spectra, and molecular modeling data (PDF)

Author Contributions

⊥ S.G. and G.D. contributed equally to this work.

The authors declare no competing financial interest.

Supplementary Material

ml4c00140_si_001.pdf (7.6MB, pdf)

References

  1. Fire A.; Xu S.; Montgomery M. K.; Kostas S. A.; Driver S. E.; Mello C. C. Potent and specific genetic interference by double-stranded RNA in Caenorhabditis elegans. Nature. 1998, 391, 806–811. 10.1038/35888. [DOI] [PubMed] [Google Scholar]
  2. Elbashir S. M.; Harborth J.; Lendeckel W.; Yalcin A.; Weber K.; Tuschl T. Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian cells. Nature. 2001, 411, 494–498. 10.1038/35078107. [DOI] [PubMed] [Google Scholar]
  3. Friedrich M.; Aigner A. Therapeutic siRNA: state-of-the-art and future perspectives. BioDrugs. 2022, 36, 549–571. 10.1007/s40259-022-00549-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Watts J. K.; Deleavey G. F.; Damha M. J. Chemically modified siRNA: tools and applications. Drug Discovery Today. 2008, 13, 842–855. 10.1016/j.drudis.2008.05.007. [DOI] [PubMed] [Google Scholar]
  5. Egli M.; Manoharan M. Chemistry, structure and function of approved oligonucleotide therapeutics. Nucleic Acids Res. 2023, 51, 2529–2573. 10.1093/nar/gkad067. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Gangopadhyay S.; Gore K. R. Advances in siRNA therapeutics and synergistic effect on siRNA activity using emerging dual ribose modifications. RNA biol. 2022, 19, 452–467. 10.1080/15476286.2022.2052641. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Guo S.; Zhang M.; Huang Y. Three ‘E’challenges for siRNA drug development. Trends. Mol. Med. 2024, 30, 13–24. 10.1016/j.molmed.2023.10.005. [DOI] [PubMed] [Google Scholar]
  8. Hu B.; Zhong L.; Weng Y.; Peng L.; Huang Y.; Zhao Y.; Liang X. J. Therapeutic siRNA: state of the art. Signal Transduct. Target. Ther. 2020, 5, 101. 10.1038/s41392-020-0207-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Khvorova A.; Watts J. K. The chemical evolution of oligonucleotide therapies of clinical utility. Nat. Biotechnol. 2017, 35, 238–248. 10.1038/nbt.3765. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Gore K. R.; Nawale G. N.; Harikrishna S.; Chittoor V. G.; Pandey S. K.; Höbartner C.; Patankar S.; Pradeepkumar P. I. Synthesis, gene silencing, and molecular modeling studies of 4′-C-aminomethyl-2′-O-methyl modified small interfering RNAs. J. Org. Chem. 2012, 77, 3233–3245. 10.1021/jo202666m. [DOI] [PubMed] [Google Scholar]
  11. Baker Y. R.; Thorpe C.; Chen J.; Poller L. M.; Cox L.; Kumar P.; Lim W. F.; Lie L.; McClorey G.; Epple S.; Singleton D.; McDonough M. A.; Hardwick J. S.; Christensen K. E.; Wood M. J. A.; Hall J. P.; El-Sagheer A. H.; Brown T. An LNA-amide modification that enhances the cell uptake and activity of phosphorothioate exon-skipping oligonucleotides. Nat. Commun. 2022, 13, 4036. 10.1038/s41467-022-31636-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Gangopadhyay S.; Das G.; Gupta S.; Ghosh A.; Bagale S. S.; Roy P. K.; Mandal M.; Harikrishna S.; Sinha S.; Gore K. R. 4′-C-Acetamidomethyl-2′-O-methoxyethyl Nucleic Acid Modifications Improve Thermal Stability, Nuclease Resistance, Potency, and hAgo2 Binding of Small Interfering RNAs. J. Org. Chem. 2024, 89, 3747–3768. 10.1021/acs.joc.3c02506. [DOI] [PubMed] [Google Scholar]
  13. Chiu Y. L.; Rana T. M. siRNA function in RNAi: a chemical modification analysis. Rna. 2003, 9, 1034–1048. 10.1261/rna.5103703. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Mao S.; Haruehanroengra P.; Ranganathan S. V.; Shen F.; Begley T. J.; Sheng J. Base Pairing and Functional Insights into N3-Methylcytidine (m3C) in RNA. ACS Chem. Biol. 2021, 16, 76–85. 10.1021/acschembio.0c00735. [DOI] [PubMed] [Google Scholar]
  15. Rydzik A. M.; Gottschling D.; Simon E.; Skronska-Wasek W.; Rippmann J. F.; Riether D. Epigenetic Modification 6-Methyladenosine Can Impact the Potency and Specificity of siRNA. ChemBioChem. 2021, 22, 491–495. 10.1002/cbic.202000551. [DOI] [PubMed] [Google Scholar]
  16. Imaeda A.; Tomoike F.; Hayakawa M.; Nakamoto K.; Kimura Y.; Abe N.; Abe H. N6-methyl adenosine in siRNA evades immune response without reducing RNAi activity. Nucleosides Nucleotides Nucleic Acids. 2019, 38, 972–979. 10.1080/15257770.2019.1641205. [DOI] [PubMed] [Google Scholar]
  17. Schlegel M. K.; Matsuda S.; Brown C. R.; Harp J. M.; Barry J. D.; Berman D.; Castoreno A.; Schofield S.; Szeto J.; Manoharan M.; Charissé K.; Egli M.; Maier M. A. Overcoming GNA/RNA base-pairing limitations using isonucleotides improves the pharmacodynamic activity of ESC+ GalNAc-siRNAs. Nucleic Acids Res. 2021, 49, 10851–10867. 10.1093/nar/gkab916. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Schlegel M. K; Janas M. M; Jiang Y.; Barry J. D; Davis W.; Agarwal S.; Berman D.; Brown C. R; Castoreno A.; LeBlanc S.; Liebow A.; Mayo T.; Milstein S.; Nguyen T.; Shulga-Morskaya S.; Hyde S.; Schofield S.; Szeto J.; Woods L. B.; Yilmaz V. O; Manoharan M.; Egli M.; Charisse K.; Sepp-Lorenzino L.; Haslett P.; Fitzgerald K.; Jadhav V.; Maier M. A From bench to bedside: Improving the clinical safety of GalNAc–siRNA conjugates using seed-pairing destabilization. Nucleic Acids Res. 2022, 50, 6656–6670. 10.1093/nar/gkac539. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Laursen M. B.; Pakula M. M.; Gao S.; Fluiter K.; Mook O. R.; Baas F.; Langklær N.; Wengel S. L.; Wengel J.; Kjems J.; Bramsen J. B. Utilization of unlocked nucleic acid (UNA) to enhance siRNA performance in vitro and in vivo. Mol. Biosyst. 2010, 6, 862–870. 10.1039/b918869j. [DOI] [PubMed] [Google Scholar]
  20. Zhang J.; Zheng J.; Lu C.; Du Q.; Liang Z.; Xi Z. Modification of the siRNA Passenger Strand by 5-Nitroindole Dramatically Reduces its Off-Target Effects. ChemBioChem. 2012, 13, 1940–1945. 10.1002/cbic.201200349. [DOI] [PubMed] [Google Scholar]
  21. Egli M.; Schlegel M. K.; Manoharan M. Acyclic (S)-glycol nucleic acid (S-GNA) modification of siRNAs improves the safety of RNAi therapeutics while maintaining potency. RNA 2023, 29, 402–414. 10.1261/rna.079526.122. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Sahoo A.; Das G.; Choudhary N. K.; Harikrishna S.; Gore K. R. Synthesis and Structural Analysis of N3-Methyluridine and 2′-Alkoxy/Fluoro-N3-Methyluridine Nucleosides by Using NMR Spectroscopy, X-Ray Crystallography, and Computational Methods. ChemistrySelect 2023, 8, e202301858 10.1002/slct.202301858. [DOI] [Google Scholar]
  23. Sahoo A.; Das G.; Ghosh A.; Shivappa Bagale S.; Kumar Choudhary N.; Harikrishna S.; Sinha S.; Gore K. R. N3-Methyluridine and 2′-O-Alkyl/2′-Fluoro-N3-methyluridine functionalized nucleic acids improve nuclease resistance while maintaining duplex geometry. Bioorg. Med. Chem. 2024, 100, 117616 10.1016/j.bmc.2024.117616. [DOI] [PubMed] [Google Scholar]
  24. Martin-Pintado N.; Deleavey G. F.; Portella G.; Campos-Olivas R.; Orozco M.; Damha M. J.; González C. Backbone FC—H···O Hydrogen Bonds in 2′ F-Substituted Nucleic Acids. Angew. Chem., Int. Ed. 2013, 52, 12065–12068. 10.1002/anie.201305710. [DOI] [PubMed] [Google Scholar]
  25. Corey D. R.; Damha M. J.; Manoharan M. Challenges and opportunities for nucleic acid therapeutics. Nucleic Acid Ther. 2022, 32, 8–13. 10.1089/nat.2021.0085. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Roberts T. C.; Langer R.; Wood M. J. Advances in oligonucleotide drug delivery. Nat. Rev. Drug Discovery 2020, 19, 673–694. 10.1038/s41573-020-0075-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Habibian M.; Harikrishna S.; Fakhoury J.; Barton M.; Ageely E. A.; Cencic R.; Fakih H. H.; Katolik A.; Takahashi M.; Rossi J.; Pelletier J.; Gagnon K. T.; Pradeepkumar P. I.; Damha M. J. Effect of 2′-5′/3′-5′ phosphodiester linkage heterogeneity on RNA interference. Nucleic Acids Res. 2020, 48, 4643–4657. 10.1093/nar/gkaa222. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Iwamoto N.; Butler D. C D; Svrzikapa N.; Mohapatra S.; Zlatev I.; Sah D. W Y; Meena; Standley S. M; Lu G.; Apponi L. H; Frank-Kamenetsky M.; Zhang J. J.; Vargeese C.; Verdine G. L Control of phosphorothioate stereochemistry substantially increases the efficacy of antisense oligonucleotides. Nat. Biotechnol. 2017, 35, 845–851. 10.1038/nbt.3948. [DOI] [PubMed] [Google Scholar]
  29. Czabotar P. E.; Lessene G.; Strasser A.; Adams J. M. Control of apoptosis by the BCL-2 protein family: implications for physiology and therapy. Nat. Rev. Mol. Cell Biol. 2014, 15, 49–63. 10.1038/nrm3722. [DOI] [PubMed] [Google Scholar]
  30. Khvorova A.; Reynolds A.; Jayasena S. D. Functional siRNAs and miRNAs exhibit strand bias. Cell. 2003, 115, 209–216. 10.1016/S0092-8674(03)00801-8. [DOI] [PubMed] [Google Scholar]
  31. Andersson S.; Antonsson M.; Elebring M.; Jansson-Löfmark R.; Weidolf L. Drug metabolism and pharmacokinetic strategies for oligonucleotide-and mRNA-based drug development. Drug discov. Today. 2018, 23, 1733–1745. 10.1016/j.drudis.2018.05.030. [DOI] [PubMed] [Google Scholar]
  32. Traut T. W. Physiological concentrations of purines and pyrimidines. Mol. Cell. Biochem. 1994, 140, 1–22. 10.1007/BF00928361. [DOI] [PubMed] [Google Scholar]
  33. Case D. A.; Belfon K.; Ben-Shalom I. Y.; Brozell S. R.; Cerutti D. S.; Cheatham T. E. III; Cruzeiro V. W. D.; Darden T. A.; Duke R. E.; Giambasu G.. Amber; University of California, San Francisco, 2020.
  34. Elkayam E.; Kuhn C. D.; Tocilj A.; Haase A. D.; Greene E. M.; Hannon G. J.; Joshua-Tor L. The Structure of Human Argonaute-2 in Complex with miR-20a. Cell. 2012, 150, 100–110. 10.1016/j.cell.2012.05.017. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Pettersen E. F.; Goddard T. D.; Huang C. C.; Couch G. S.; Greenblatt D. M.; Meng E. C.; Ferrin T. E. UCSF Chimera—a visualization system for exploratory research and analysis. J. Comput. Chem. 2004, 25, 1605–1612. 10.1002/jcc.20084. [DOI] [PubMed] [Google Scholar]
  36. Brautigam C. A.; Steitz T. A. Structural principles for the inhibition of the 3′-5′ exonuclease activity of Escherichia coli DNA polymerase I by phosphorothioates. J. Mol. Biol. 1998, 277, 363–377. 10.1006/jmbi.1997.1586. [DOI] [PubMed] [Google Scholar]

Associated Data

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

Supplementary Materials

ml4c00140_si_001.pdf (7.6MB, pdf)

Articles from ACS Medicinal Chemistry Letters are provided here courtesy of American Chemical Society

RESOURCES