Skip to main content
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2013 Nov 2.
Published in final edited form as: J Org Chem. 2012 Oct 25;77(21):9889–9892. doi: 10.1021/jo301834p

Automated Solid-Phase Synthesis of RNA Oligonucleotides Containing a Non-bridging Phosphorodithioate Linkage via Phosphorothioamidites

Nan-Sheng Li 1,, John K Frederiksen 1,, Joseph A Piccirilli 1,2
PMCID: PMC3500914  NIHMSID: NIHMS416770  PMID: 23050987

Abstract

graphic file with name nihms416770f4.jpg

This work describes a general method for the synthesis of oligoribonucleotides containing a site-specific non-bridging phosphorodithioate linkage via automated solid-phase synthesis using 5′-O-DMTr-2′-O-TBS-ribonucleoside 3′-N,N-dimethyl-S-(2,4-dichlorobenzyl) phosphorothioamidites (2a–2d). The 3′-phosphorothioamidites (2a–2d) can be conveniently prepared in good yields (86–99%) via a one-pot reaction from the corresponding 5′-O-DMTr-2′-O-TBS-ribonucleosides (1a–1d).


Oligonucleotides containing phosphoromonothioate (PS) linkages are among the most thoroughly characterized and widely used oligonucleotide analogues.1 Both DNA and RNA PS oligonucleotides are incisive mechanistic probes that have helped to elucidate the catalytic mechanisms of both protein and nucleic acid enzymes.27 PS linkages also confer nuclease resistance to oligodeoxynucleotides and oligoribonucleotides, increasing their half-life as antisense agents.810 More recently, several investigations have studied the effects of PS linkages on the uptake and activity of small interfering RNA (siRNA) oligonucleotides.1115

Less well investigated are phosphorodithioate (PS2) oligonucleotides, in which sulfur replaces both non-bridging phosphate oxygens. Unlike PS oligonucleotides, dithioate analogues are achiral at phosphorous and therefore sterochemically mimic their all-oxygen counterparts more closely than monothioates. PS2 oligonucleotides also do not require stereochemical resolution following synthesis, as do monothioates when species with defined stereochemistry are desired.16

Deoxynucleoside phosphorodithioate dimers have been prepared in several ways. One approach involves the reaction of a 5′-O-protected-2′-deoxynucleoside with a dithiophosphorylating agent, followed by coupling of the product to a 3′-O,N-protected-2′-deoxynucleoside.1719 A second approach, used to prepare dithymidine phosphorodithioate, involves the coupling reaction of the triethylammonium salt of a 3′-thymidine phosphonodithioate20,21 or phosphorodithioate22,23 to a 3′-O-protected thymidine. A third approach involves reaction of the corresponding di-deoxynucleoside phosphoramidites (B1-O-P(NPr-i2)-O-B2) with 4-chlorobenzylmercaptan (or hydrogen sulfide) followed by sulfurization.2426 For construction of oligodeoxynucleotides by solid-phase synthesis, 3′-phosphoramidites of dinucleoside phosphorodithioates17,2527 enabled access to sequences containing a single PS2 linkage, while 2′-deoxynucleoside 3′-phosphorothioamidites2830 enabled access to sequences containing exclusively PS2 linkages.

Compared to the extensive studies describing the synthesis of PS2-bearing deoxyoligonucleotides, only three reports describe the synthesis of RNA oligonucleotides containing PS2 linkages.3133 In 1990, Petersen et al. reported a solution synthesis of ribonucleoside phosphorodithioate dimers using 5′-O-DMTr-2′-O-TBS-ribonucleoside 3′-phosphorothioamidites (Figure 1, I) with tetrazole as activator.32 In 1996, Greef et al. synthesized RNA containing exclusively PS2 linkages via a solid-phase approach that employed ribonucleoside 3′-H-phosphonothioates (Figure 1, II) with diphenylchlorophosphate as an activator and 2,4-dichlorobenzylthiosuccinimide as a sulfurizing reagent.31 This method provides an effective approach for synthesizing oligoribonucleotides containing exclusively PS2 linkages, but synthesis of RNA containing individual PS2 linkages would require multiple activators and likely render the approach difficult to automate. Use of this method to prepare a 12-nt RNA containing a single phosphorodithioate linkage required a manual coupling step.34 Following initial submission of this manuscript, Yang et al. reported a similar solid-phase synthesis of RNAs containing a single or multiple PS2 linkages using ribonucleoside 3′-pyrrolidino-S-[β-benzoylmercapto)ethyl]-3′-phosphorothioamidites (Figure 1, III).33 Here we wish to report the preparation and use of the 3′-phosphorothioamidites (Figure 1 and Scheme 1, 2a–2d) for automated solid-phase synthesis of RNA oligonucleotides containing single PS2 linkages. The 3′-phosphorothioamidites (2a–2d) are prepared with greater convenience and efficiency (86–99% yield) compared to the preparation of 3′-H-phosphonothioates (II) (two steps, 35–77% yield).31 It is also more convenient than the synthesis of the 3′-phosphorothioamidites (I) from Petersen et al.’s procedure32 or the 3′-phosphorothioamidites (III) from Yang et al.’s procedure,33 since all of the starting materials used for the preparation of 2a–2d are commercially available.

Figure 1.

Figure 1

Phosphorus reagents used to synthesize RNAs containing PS2 linkages.

Scheme 1.

Scheme 1

The ribonucleoside 3′-phosphorothioamidites 2a–2d were prepared from the corresponding 5′-O-DMTr-2′-O-TBS-protected ribonucleoside derivatives 1a–1d in 86–99% yields according to the literature procedure described for the synthesis of 2d32 (Scheme 1).

Incorporation of phosphorothioamidites 2a–2d into oligonucleotides at specific sites was performed on an Expedite Nucleic Acid Synthesis System (8900) via a modified 1-µM RNA protocol using phosphoramidite chemistry (Scheme 2). The protocol was modified for phosphorothioamidites 2a–2d as follows: (1) The coupling step was repeated once (double coupling); (2) sulfurization with 3-(N,N-dimethylaminomethylidene) amino-3-H-1,2,4-dithiazole-5-thione (DDTT) was executed prior to the standard capping procedure. As expected, coupling of the ribonucleoside phosphorothioamidites 2a–2d was less efficient than for the corresponding 2′-deoxyribonucleoside phosphorothioamidites.29 However, when double coupling was applied, the trityl data showed the coupling yields with 2a–2d were comparable to those of commercially available RNA phosphoramidites. We also found that sulfurization was more efficient with DDTT than with the Beaucage reagent.35,36 For the incorporation of standard nucleotides, all commercially available phosphoramidites could be used except the N2-isobutyrylguanosine phosphoramidite. To minimize their degradation under basic conditions, we chose PAC protected phosphoramidites to allow deprotection of the oligonucleotides by treatment with ammonium hydroxide at 55°C for only two hours. After solid-phase synthesis, the solid supports were deprotected via a three-step protocol as shown in Scheme 2.31 The crude oligonucleotides containing single phosphorodithioate linkages were analyzed and purified by ion exchange HPLC. HPLC profiles of the crude oligonucleotides showed that the phosphorodithioate oligonucleotides were the major products (see Supporting Information). The following oligonucleotides containing single PS2 linkages were prepared and characterized: 5′-ACG UA-O-P(S)2-O-C GUU-3′ (3a), 5′-ACG UAC-O-P(S)2-O-GUU-3′ (3b), 5′-ACG-O-P(S)2-O-UAC GUU-3′ (3c), ACG U-O-P(S)2-O-AC GUU-3′ (3d), 5′-UGG UAA U-O-P(S)2-O-AA GCU GAC GGA CAU-3′ (3e) and 5′-UGG UA-O-P(S)2-O-A UAA GCU GAC GGA CAU-3′ (3f). The purified phosphorodithioate oligonucleotides were also analyzed and confirmed by MALDI-TOF mass spectrometry (Table 1).

Scheme 2.

Scheme 2

Table 1.

Yields and MALDI-TOF MS data of the purified oligonucleotides containing single PS2 linkages (3a–3f).

Oligos Yield (%) MALDI-TOF MS
Calculated Found
3a 52 2846.4 (M+), 2847.4 (MH+), 2890.4 (M−2H+2Na+) 2847.5, 2890.4
3b 58 2846.8, 2888.7
3c 55 2847.9, 2888.8
3d 53 2847.9, 2889.8
3e 33 6788.9 (M+), 6789.9 (MH+), 6811.9 (MNa+), 6832.9 (M−2H+2Na+) 6854.9 (M−3H+3Na) 6789.2, 6790.6, 6811.4, 6812.6, 6829.7, 6853.8
3f 36 6787.4, 6809.6, 6831.2, 6853.2, 6854.7

In summary, in contrast to well developed approaches for preparation of ODNs bearing PS2 linkages, corresponding approaches for RNAs until now have lacked automation or the capacity for site-specific incorporation. To enable convenient access to these modifications, we constructed the four ribonucleoside 3′-phosphorothioamidites (2a–2d) in one pot synthesis from commercially available starting materials and used them to prepare RNA containing site-specific PS2 linkages by automated solid-phase synthesis. Two recent applications in RNAi33 and RNA structure-function analysis37 likely will heighten future interest in these molecules. The current work simplifies access to RNA bearing site-specific PS2 linkages by enabling automated, solid-phase synthesis using ribonucleoside 3′-phosphorothioamidites (2a–2d).38 Following submission of this work, Yang et al demonstrated the use of 3′-pyrrolidino-S-[β-benzoylmercapto)ethyl]-3′-phosphorothioamidites for constructing PS2 RNA, but our approach offers the convenience of phosphoramidite access in one pot from commercially available starting materials.

Experimental

5′-O-(4,4’-Dimethoxytrityl)-2′-O-(tert-butyldimethylsilyl)-N6-phenoxyacetyladenosine 3′-N,N-dimethyl-S-(2,4-dichlorobenzyl) phosphorothioamidite (2a)

To a solution of 5′-O-(dimethoxytrityl)-2′-O-(tert-butyldimethylsilyl)-N6-phenoxyacetyladenosine (1a) (0.408 g, 0.50 mmol) in dry acetonitrile (2.5 mL) were simultaneously added i-Pr2NEt (0.14 mL, 0.80 mmol) and ClP(NCH3)2 (93 mg, 0.60 mmol). After the reaction mixture was stirred at room temperature for 30 min, 2,4-dichlorobenzyl mercaptan (0.10 mL, 0.70 mmol) was added, and stirring continued at room temperature for an additional 30 min. The reaction was quenched with aqueous saturated NaHCO3, and the product was extracted with ethyl acetate and dried over anhydrous magnesium sulfate. The solvent was removed, and the residue was precipitated in degassed hexane to give a white solid. The white solid was rinsed with hexane (2×5 mL) and dried over vacuum to give 2a in 86% yield (468 mg, >90% purity as assessed by 31P NMR). 31P NMR (CD3CN): δ 179.9, 173.5; HRMS calcd for C54H61N6O8NaSiPSCl2 (MNa+) 1105.3053, found 1105.3056.

5′-O-(4,4’-Dimethoxytrityl)-2′-O-(tert-butyldimethylsilyl)-N4-phenoxyacetylcytidine 3′-N,N-dimethyl-S-(2,4-dichlorobenzyl) phosphorothioamidite (2b)

Phosphoramidite 2b was obtained from 5′-O-(dimethoxytrityl)-2′-O-(tert-butyldimethylsilyl)-N4-phenoxyacetylcytidine (1b) (397 mg, 0.50 mmol) in 94% yield (499 mg, >90% purity as assessed by 31P NMR) according to the procedure described for the synthesis of 2a. 31P NMR (CD3CN): δ 180.9, 176.5; HRMS calcd for C53H61N4O9NaSiPSCl2 (MNa+) 1081.2935, found 1081.2966.

5′-O-(4,4-Dimethoxytrityl)-2′-O-(tert-butyldimethylsilyl)-N2-phenoxyacetylguanosine 3′-N,N-dimethyl-S-(2,4-dichlorobenzyl) phosphorothioamidite (2c)

Phosphoramidite 2c was obtained from 5′-O-(dimethoxytrityl)-2′-O-(tert-butyldimethylsilyl)-N2-phenoxyacetylguanosine (1c) (417 mg, 0.50 mmol) in 95% yield (523 mg, >90% purity as assessed by 31P NMR) according to the procedure described for the synthesis of 2a. 31P NMR (CD3CN): δ 180.3, 173.4; HRMS calcd for C54H62N6O9SiPSCl2 (MH+) 1099.3177, found 1099.3166.

5′-O-(4,4-Dimethoxytrityl)-2′-O-(tert-butyldimethylsilyl)uridine 3′-N,N-dimethyl-S-(2,4-dichlorobenzyl) phosphorothioamidite (2d):32

Phosphoramidite 2d was obtained from 5′-O-(dimethoxytrityl)-2′-O-(tert-butyldimthylsilyl)uridine (1d) (660 mg, 1.0 mmol) in 99% yield (917 mg, >90% purity as assessed by 31P NMR) according to the procedure described for the synthesis of 2a. 31P NMR (CD3CN): δ 180.5, 174.2.

Incorporation of phosphorothioamidites 2a–2d into oligoribonucleotides by solid-phase synthesis

Oligonucleotides were synthesized on a 1-µmol scale with standard phosphoramidites (Pac-A-CE, Pac-C-CE, Pac-G-CE, U-CE, ChemGenes) using an Expedite Nucleic Acid Synthesis System (8900) and a modified RNA protocol. Each phosphorothioamidite (2a–2d) (~0.1 M, 100 mg in 1 mL CH3CN) was double-coupled, followed by sulfurization with DDTT (0.05 M solution in 40% pyridine/CH3CN) for 400 s to form the phosphorodithioate linkage. Capping, detritylation and various washes are the same as those described in the standard RNA protocol. The oligonucleotides were deprotected according to the following steps: (1) the solid support inside the column was treated with a mixture of thiophenol (0.4 mL), triethylamine (0.8 mL) and 1,4-dioxane (0.8 mL) at room temperature for 2 h; (2) the thiophenolate solution was removed and the solid support was rinsed with methanol and subsequently with ether; (3) the support was treated with 2 mL of ammonium hydroxide/ethanol (3:1 v/v) at 55 °C for 2 h; (4) after cooling to room temperature, the supernatant was removed and the support was rinsed with 2×1 mL of an ethanol:acetonitrile:water mixture (3:1:1 v/v/v); (5) the rinses were combined with the supernatant, and the resulting solution was evaporated to dryness; (6) the oligo was desilylated with triethylamine trihydrofluoride/N-methylpyrrolidinone solution (~300 µL) [made from N-methylpyrrolidinone (NMP) (135 µL), triethylamine (70 µL) and TEA-3HF (95 µL)] at 65 °C for 1.5 h. The crude oligonucleotides were precipitated from n-butanol (1 mL), redissolved into water (500 µL), and purified via ion exchange HPLC (Dionex DNAPac PA-100, 9×250 mm; buffer A, 0.25 M tris, pH 8.93; B, water; C, 1.0 M NaCl; flow rate: 2.0 mL/min).

Supplementary Material

1_si_001

Acknowledgements

Dedicated to the 100th birthday memory of Prof. Yao-Zeng Huang. This work was supported by an N.I.H. grant to J.A.P. (1R01AI081987). J.K.F. was also supported by the University of Chicago Medical Scientist Training Program (5 T32 GM07281).

Footnotes

Supporting Information Available

1H NMR, 13C NMR and 31P NMR spectra of phosphorothioamidites 2a–2d; HRMS spectra of new phosphorothioamidites 2a–2c; MALDI-TOF mass spectra of oligonucleotides 3a–3f; HPLC profiles of the crude and purified oligonucleotides (3a, 3c, 3e and 3f); Trityl yields of 3e and 3f from the solid-phase synthesis. These materials are available free of charge via the Internet at http://pubs.acs.org.

References

  • 1.Eckstein F. Annu. Rev. Biochem. 1985;54:367–402. doi: 10.1146/annurev.bi.54.070185.002055. [DOI] [PubMed] [Google Scholar]
  • 2.Burgers PM, Eckstein F. Proc. Natl. Acad. Sci. USA. 1978;75:4798–4800. doi: 10.1073/pnas.75.10.4798. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Frederiksen JK, Piccirilli JA. Methods. 2009;49:148–166. doi: 10.1016/j.ymeth.2009.07.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Moore MJ, Sharp PA. Nature. 1993;365:364–368. doi: 10.1038/365364a0. [DOI] [PubMed] [Google Scholar]
  • 5.Sasnauskas G, Connolly BA, Halford SE, Siksnys V. Proc. Natl. Acad. Sci. USA. 2007;104:2115–2120. doi: 10.1073/pnas.0608689104. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Shan S, Kravchuk AV, Piccirilli JA, Herschlag D. Biochemistry. 2001;40:5161–5171. doi: 10.1021/bi002887h. [DOI] [PubMed] [Google Scholar]
  • 7.Yoshida A, Sun S, Piccirilli JA. Nat. Struct. Biol. 1999;6:318–321. doi: 10.1038/7551. [DOI] [PubMed] [Google Scholar]
  • 8.Cazenave C, Stein CA, Loreau N, Thuong NT, Neckers LM, Subasinghe C, Helene C, Cohen JS, Toulme J-J. Nucleic Acids Res. 1989;17:4255–4273. doi: 10.1093/nar/17.11.4255. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Milligan JF, Matteucci M, Martin JC. J. Med. Chem. 1993;36:1923–1937. doi: 10.1021/jm00066a001. [DOI] [PubMed] [Google Scholar]
  • 10.Morvan F, Rayner B, Imbach JL. Tetrahedron Lett. 1990;31:7149–7152. [Google Scholar]
  • 11.Kala A, Friedman SH. Pharm. Res. 2011;28:3050–3057. doi: 10.1007/s11095-011-0529-z. [DOI] [PubMed] [Google Scholar]
  • 12.Koller E, Vincent TM, Chappell A, De S, Manoharan M, Bennett CF. Nucleic Acids Res. 2011;39:4795–4807. doi: 10.1093/nar/gkr089. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Kraynack BA, Baker BF. RNA. 2006;12:163–176. doi: 10.1261/rna.2150806. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Li ZY, Mao H, Kallick DA, Gorenstein DG. Biochem. Biophys. Res. Commun. 2005;329:1026–1030. doi: 10.1016/j.bbrc.2005.02.071. [DOI] [PubMed] [Google Scholar]
  • 15.Overhoff M, Sczakiel G. EMBO Rep. 2005;6:1176–1181. doi: 10.1038/sj.embor.7400535. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Frederiksen JK, Piccirilli JA. Methods Enzymol. 2009;468:289–309. doi: 10.1016/S0076-6879(09)68014-9. [DOI] [PubMed] [Google Scholar]
  • 17.Capaldi DC, Cole DL, Ravikumar VT. Nucleic Acids Res. 2000;28:e40. doi: 10.1093/nar/28.9.e40. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Dahl BH, Bjergarde K, Nielsen J, Dahl O. Tetrahedron Lett. 1990;31:3489–3492. [Google Scholar]
  • 19.Eldrup AB, Bjergarde K, Felding J, Kehler J, Dahl O. Nucleic Acids Res. 1994;22:1797–1804. doi: 10.1093/nar/22.10.1797. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Murata T, Iwai S, Ohtsuka E. Nucleic Acids Res. 1990;18:7279–7286. doi: 10.1093/nar/18.24.7279. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Porritt GM, Reese CB. Tetrahedron Lett. 1989;30:4713–4716. [Google Scholar]
  • 22.Kehler J, Puschl A, Dahl O. Nucleosides & Nucleotides. 1997;16:1633–1636. [Google Scholar]
  • 23.Yau EK, Ma Y-X, Caruthers MH. Tetrahedron Lett. 1990;31:1953–1956. [Google Scholar]
  • 24.Brill WK-D, Nielsen J, Caruthers MH. J. Am. Chem. Soc. 1991;113:3972–3980. [Google Scholar]
  • 25.Grandas A, Marshall WS, Nielsen J, Caruthers MH. Tetrahedron Lett. 1989;30:543–546. [Google Scholar]
  • 26.Nielsen J, Brill WK-D, Caruthers MH. Tetrahedron Lett. 1988;29:2911–2914. [Google Scholar]
  • 27.Brill WK-D, Nielsen J, Caruthers MH. Tetrahedron Lett. 1988;29:5517–5520. [Google Scholar]
  • 28.Bjergarde K, Dahl O. Nucleic Acids Res. 1991;19:5843–5850. doi: 10.1093/nar/19.21.5843. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Brill WK-D, Tang J-Y, Ma Y-X, Caruthers MH. J. Am. Chem. Soc. 1989;111:2321–2322. [Google Scholar]
  • 30.Wiesler WT, Caruthers MH. J. Org. Chem. 1996;61:4272–4281. doi: 10.1021/jo960274y. [DOI] [PubMed] [Google Scholar]
  • 31.Greef CH, Seeberger PH, Caruthers MH. Tetrahedron Lett. 1996;37:4451–4454. [Google Scholar]
  • 32.Petersen KH, Nielsen J. Tetrahedron Lett. 1990;31:911–914. [Google Scholar]
  • 33.Yang X, Sierant M, Janicka M, Peczek L, Martinez C, Hassell T, Li N, Li X, Wang T, Nawrot B. ACS Chem. Biol. 2012;7:1214–1220. doi: 10.1021/cb300078e. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Derrick WB, Greef CH, Caruthers MH, Uhlenbeck OC. Biochemistry. 2000;39:4947–4954. doi: 10.1021/bi000146a. [DOI] [PubMed] [Google Scholar]
  • 35.Bergot BJ, Egan W. J. Chromatography. 1992;599:35–42. [Google Scholar]
  • 36.Iyer RP, Egan W, Regan JB, Beaucage SL. J. Am. Chem. Soc. 1990;112:1253–1254. [Google Scholar]
  • 37.Frederiksen JK, Li N-S, Das R, Herschlag D, Piccirilli JA. RNA. 2012;18:1123–1141. doi: 10.1261/rna.028738.111. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.An attempt to extend the scope of our approach to prepare 5'-Ups2Ups2Ups2Ups2Ups2U-3' containing exclusively phosphorodithioate linkages was not successful. The yields dropped significantly even though double coupling was used for each linkage. The reasons are still unclear. Therefore, our method may not be suitable for synthesizing oligonucleotides containing exclusively phosphorodithioate linkages.

Associated Data

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

Supplementary Materials

1_si_001

RESOURCES