- Email: [email protected]
3′-Modification stabilizes mRNA and increases translation in cells
Bioorganic & Medicinal Chemistry Letters 28 (2018) 2451–2453
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters journal homepage: www.elsevier.com/locate/bmcl
3′-Modification stabilizes mRNA and increases translation in cells ⁎
T
Christian Gampe , Amy C. Seila White, Swetha Siva, Frédéric Zécri, John Diener Novartis Institutes for BioMedical Research, 181 Massachusetts Ave., Cambridge, MA 02139, USA
A R T I C LE I N FO
A B S T R A C T
Keywords: mRNA 3′-Terminus Translation Gene therapy Bioconjugation Nucleotide
Successful implementation of mRNA gene therapy is facing many hurdles, for example poor expression levels of the exogenously delivered mRNA transcripts. Herein we describe the synthesis of various 3′-modified RNA oligonucleotides, and we show that 3′-modification drastically stabilizes these oligonucleotides in cell extracts. Modification of the 3′-terminus of gaussia luciferase mRNA results in 3-fold increased and extended (> 48 h) translation of the mRNA. Our findings suggest 3′-modification of RNA-transcripts as a valid approach to increase expression levels for application in mRNA gene therapy.
Introduction The prospect of mRNA gene therapy has attracted much attention since the first report in 1990 of cellular translation of exogenously delivered mRNA.1 However, to date implementation of mRNA transcript therapy remains an ambitious goal facing multiple challenges including delivery, immunogenicity, translation efficiency, and stability of the therapeutic mRNA.2 One promising approach to address some of these challenges is chemical modification of the RNA strand. For example, nucleobase modification has resulted in RNA strands with reduced immunogenicity.3 Herein, we describe a chemical method to selectively modify the RNA 3′-end and show that these modifications lead to improved stability in cell lysates and increased expression in cells. Deadenylation by 3′-exonucleases is the rate limiting step in eukaryotic mRNA decay.4 One of the dominant exo-nucleases is the Poly (A)-specific Ribonuclease (PARN), which selectively deadenylates single-stranded mRNA via recognition of the 2′,3′-adenosine diol.5 We hypothesized that 2′,3′-cis diol moiety alteration could reduce mRNA susceptibility to PARN and lead to increased mRNA translation. To test our hypothesis we established a robust method for high-yielding, chemical modification of the RNA 3′-end. We developed an analytical method based on ion pairing chromatography – mass spectrometry (IPC-MS) that allowed us to follow the reaction progress and to optimize reaction conditions to obtain synthetically useful yields. For analytical purposes, selective 3′-diol cleavage using sodium periodate, followed by condensation with hydrazines or hydroxylamines, or reductive amination, is well established on RNAs of various lengths.6 Recently, Gillingham and coworkers highlighted the extraordinary utility of the intermediate bis-aldehyde
⁎
Corresponding author. E-mail address: [email protected] (C. Gampe).
https://doi.org/10.1016/j.bmcl.2018.06.008 Received 15 May 2018; Received in revised form 1 June 2018; Accepted 2 June 2018
Available online 04 June 2018 0960-894X/ © 2018 Elsevier Ltd. All rights reserved.
for bioconjugation reactions. In these reactions, the formation of a cyclic intermediate increases the reaction rate significantly, resulting in useful conversions even at low substrate concentration and neutral pH.7 Using a short oligonucleotide as a model substrate, we established conditions that allow for synthetically useful, complete starting material conversion to a single 3′-capped product as observed by IPC-MS (Fig. 1, see SI for details). Besides various hydrazine and hydroxylamine adducts, and their morpholine derivatives resulting from reductive amination, we could also obtain adducts with Meldrum’s acid and cysteine-conjugates. We next assessed the stability of 3′-modified oligonucleotides in cell lysates using the method developed by Walter and coworkers.8 These researchers had reported that an oligonucleotide containing a FRETpair can be used to determine the oligonucleotide decay rates in cell extracts. We monitored the stability of the 3′-modified oligonucleotides obtained in Fig. 1 as a function of their 3′-modification in the presence of 1.2% S100 HeLa cell extract. 3′-Modification drastically extended the oligonucleotide’s half-life. The unmodified oligonucleotide had a halflife of 7 min while 3′-modified oligonucleotides had half-lifes of up to > 48 h (see SI for details). We treated the modified oligonucleotides with S100 HeLa cell extract at different concentrations and plotted the observed rate constants against % S100 extract by fitting to the Michaelis-Menten equation (Fig. 2). We found that even simple modification with methoxyamine and methylhydrazine decreased the decay rate drastically, even at 5% S100 HeLa extract (See Fig. 2). We tested if 3′-end modification would result in increased mRNA translation in cells. We treated uncapped gaussia luciferase (gLuc) mRNA under the same conditions we used for the oligonucleotide modification. Following 3′-modification, the RNA was enzymatically capped and purified by LiCl-precipitation. The mRNA (30 ng) was
Bioorganic & Medicinal Chemistry Letters 28 (2018) 2451–2453
C. Gampe et al.
Figure 1. Diol cleavage - condensation sequence allows for chemical 3′-diversification of oligonucleotides. Reagents and conditions: (a) oligo-nucleotide (10 µM in PBS buffer), 50 eq. NaIO4 (500 µM), 0 °C, 1 h; (b) condensation: 100 eq. Na2SO3 (1 mM), 100 eq. nucleophile, RT, 16 h; reductive amination: 200 eq. Na2SO3 (2 mM), 100 eq. nucleophile, 1000 eq. NaBH3CN, 37 °C, 16 h. oligonucleotide sequences: 5-/5Phos/r(GA AAA AAA AAA)-3 or 5-/5Phos/r(GU/iFluorT/UCG CCA UU/ i6-TAMN/AAA AAA AAA A-3), iFluorT: fluoresceinmodified T; i6-TAMN: 6-carboxytetramethylrhodamine-modified T. The fluorophores were chosen to engage in Förster resonance energy transfer (FRET), to enable monitoring of integrity of the strand (vide infra). Reactions were purified by desalting columns. Reaction progress was monitored via ion pairing chromatography (IPC). Final concentrations in the reaction solution are given.
Figure 3. 3′-Modification of gaussia luciferase mRNA results in increased expression in HEK293-cells. 30 ng of capped, 3′-modified gLuc mRNA was transfected into HEK293 cells using DharmaFECT. Luminescence was read after 24 and 48 h and plotted normalized to the luminescence of the unmodified gLuc at the 24 h time point. Legend indicates nature of the nucleophile used in the modification of the oligonucleotide.
Figure 2. 3′-Selective modification increases RNA half-life in HeLa cell extract. An oligonucleotide containing a FRET-pair [sequence: 5-/5Phos/r(GU/iFluorT/ UCG CCA UU/i6-TAMN/ AAA AAA AAA A)-3]8 was incubated with varying concentrations of S100 HeLa cell extract, and degradation was monitored using fluorescence. Legend indicates nature of the nucleophile used in the oligonucleotide modification.
almost 3-fold increased expression levels after 24 h, and translation was sustained for longer than 48 h. These findings suggest that 3′-mRNA modification is a valid approach to stabilize mRNA transcripts and boost expression levels. Thus, 3′-RNA modification may extend the halflife of a therapeutic mRNA transcript and could be important for mRNA gene therapy development.
transfected into HEK293 cells and mRNA expression was assessed using luminescence readout. After 24 h, modified mRNAs showed 2.5 to 3times higher expression levels than unmodified control Fig. 3. Notably, 3′-modification resulted in sustained expression past the 48 h time point, at which translation of the parent mRNA had dropped below 10%. These results support our hypothesis that increased mRNA stability results in increased and prolonged expression in cells. In conclusion, we showed that oligonucleotides can quantitatively and selectively be 3′-modified using an optimized periodate cleavage/ condensation protocol. This modification results in drastically increased RNA stability in HeLa cell extracts. 3′-Modified gLuc mRNA showed
Acknowledgments We thank Jared Ek, Carmelina Rakiec and Jennifer Porier for their help in setting up the LC-IPC methods. Dr. Shari Caplan is acknowledged for the providing the GLuc constructs. 2452
Bioorganic & Medicinal Chemistry Letters 28 (2018) 2451–2453
C. Gampe et al.
A. Supplementary data
b) Wu M, Reuter M, Lilie H, Liu Y, Wahle E, Song H. EMBO J. 2005;23:4082. 6. a) Imai J, Johnston MI, Torrence PF. J. Biol. Chem. 1982;257:12739;
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.bmcl.2018.06.008.
b) Wu TP, Ruan KC, Liu WY. Nucleic Acids Res. 1996;24:3472; c) Proudnikov D, Mirzabekov A. Nucleic Acids Res. 1996;24:4535;
References
d) Bellon L, Workman C, Scherrer J, Usman N, Wincott F. J Am Chem Soc. 1996;118:3771;
1. Wolff JA, Malone RW, Williams P, et al. Science. 1990;247:1465. 2. Weissman D. Expert Rev Vaccines. 2015;14:265. 3. a) Anderson BR, Muramatsu H, Nallagatla SR, et al. Nucleic Acid Res. 2010;38:5884;
e) Gite S, RajBhandary UL. J Biol Chem. 1997;272:5305; f) Wang Z, Chen L, Bayly SF, Torrence PF. Bioorg Med Chem Lett. 2000;10:1357;
b) Andries O, Mc Cafferty S, De Smedt SC, Weiss R, Sanders NN, Kitada T. J Controlled Release. 2015;217:337;
g) Nandanan E, Jang SY, Moro S, et al. J Med Chem. 2000;43:829; h) Kurata S, Ohtsuki T, Suzuki T, Wantanabe K. Nucleic Acid Res. 2003;31:e145;
c) Svitkin YV, Cheng YM, Chakraborty T, Presnyak V, John M, Sonenberg N. Nucleic Acids Res. 2017;45:6023. 4. a) Moore M. Science. 2005;309:1514;
i) Zhou L, Thakur CS, Molinaro RJ, et al. Bioorg Med Chem Lett. 2006;14:7862. 7. Schmidt P, Zhou L, Tishinov K, Zimmermann K, Gillingham D. Angew Chem Int Ed. 2014;53:10928. 8. Uhler SA, Cai D, Man Y, Figge C, Walter NG. J Am Chem Soc. 2003;125:14230.
b) Chen CY, Shyu AB. Wiley Interdiscp Rev RNA. 2011;2:167. 5. a) Mitchell P, Tollervey D. Curr Opin Genet Dev. 2000;10:193;
2453
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters journal homepage: www.elsevier.com/locate/bmcl
3′-Modification stabilizes mRNA and increases translation in cells ⁎
T
Christian Gampe , Amy C. Seila White, Swetha Siva, Frédéric Zécri, John Diener Novartis Institutes for BioMedical Research, 181 Massachusetts Ave., Cambridge, MA 02139, USA
A R T I C LE I N FO
A B S T R A C T
Keywords: mRNA 3′-Terminus Translation Gene therapy Bioconjugation Nucleotide
Successful implementation of mRNA gene therapy is facing many hurdles, for example poor expression levels of the exogenously delivered mRNA transcripts. Herein we describe the synthesis of various 3′-modified RNA oligonucleotides, and we show that 3′-modification drastically stabilizes these oligonucleotides in cell extracts. Modification of the 3′-terminus of gaussia luciferase mRNA results in 3-fold increased and extended (> 48 h) translation of the mRNA. Our findings suggest 3′-modification of RNA-transcripts as a valid approach to increase expression levels for application in mRNA gene therapy.
Introduction The prospect of mRNA gene therapy has attracted much attention since the first report in 1990 of cellular translation of exogenously delivered mRNA.1 However, to date implementation of mRNA transcript therapy remains an ambitious goal facing multiple challenges including delivery, immunogenicity, translation efficiency, and stability of the therapeutic mRNA.2 One promising approach to address some of these challenges is chemical modification of the RNA strand. For example, nucleobase modification has resulted in RNA strands with reduced immunogenicity.3 Herein, we describe a chemical method to selectively modify the RNA 3′-end and show that these modifications lead to improved stability in cell lysates and increased expression in cells. Deadenylation by 3′-exonucleases is the rate limiting step in eukaryotic mRNA decay.4 One of the dominant exo-nucleases is the Poly (A)-specific Ribonuclease (PARN), which selectively deadenylates single-stranded mRNA via recognition of the 2′,3′-adenosine diol.5 We hypothesized that 2′,3′-cis diol moiety alteration could reduce mRNA susceptibility to PARN and lead to increased mRNA translation. To test our hypothesis we established a robust method for high-yielding, chemical modification of the RNA 3′-end. We developed an analytical method based on ion pairing chromatography – mass spectrometry (IPC-MS) that allowed us to follow the reaction progress and to optimize reaction conditions to obtain synthetically useful yields. For analytical purposes, selective 3′-diol cleavage using sodium periodate, followed by condensation with hydrazines or hydroxylamines, or reductive amination, is well established on RNAs of various lengths.6 Recently, Gillingham and coworkers highlighted the extraordinary utility of the intermediate bis-aldehyde
⁎
Corresponding author. E-mail address: [email protected] (C. Gampe).
https://doi.org/10.1016/j.bmcl.2018.06.008 Received 15 May 2018; Received in revised form 1 June 2018; Accepted 2 June 2018
Available online 04 June 2018 0960-894X/ © 2018 Elsevier Ltd. All rights reserved.
for bioconjugation reactions. In these reactions, the formation of a cyclic intermediate increases the reaction rate significantly, resulting in useful conversions even at low substrate concentration and neutral pH.7 Using a short oligonucleotide as a model substrate, we established conditions that allow for synthetically useful, complete starting material conversion to a single 3′-capped product as observed by IPC-MS (Fig. 1, see SI for details). Besides various hydrazine and hydroxylamine adducts, and their morpholine derivatives resulting from reductive amination, we could also obtain adducts with Meldrum’s acid and cysteine-conjugates. We next assessed the stability of 3′-modified oligonucleotides in cell lysates using the method developed by Walter and coworkers.8 These researchers had reported that an oligonucleotide containing a FRETpair can be used to determine the oligonucleotide decay rates in cell extracts. We monitored the stability of the 3′-modified oligonucleotides obtained in Fig. 1 as a function of their 3′-modification in the presence of 1.2% S100 HeLa cell extract. 3′-Modification drastically extended the oligonucleotide’s half-life. The unmodified oligonucleotide had a halflife of 7 min while 3′-modified oligonucleotides had half-lifes of up to > 48 h (see SI for details). We treated the modified oligonucleotides with S100 HeLa cell extract at different concentrations and plotted the observed rate constants against % S100 extract by fitting to the Michaelis-Menten equation (Fig. 2). We found that even simple modification with methoxyamine and methylhydrazine decreased the decay rate drastically, even at 5% S100 HeLa extract (See Fig. 2). We tested if 3′-end modification would result in increased mRNA translation in cells. We treated uncapped gaussia luciferase (gLuc) mRNA under the same conditions we used for the oligonucleotide modification. Following 3′-modification, the RNA was enzymatically capped and purified by LiCl-precipitation. The mRNA (30 ng) was
Bioorganic & Medicinal Chemistry Letters 28 (2018) 2451–2453
C. Gampe et al.
Figure 1. Diol cleavage - condensation sequence allows for chemical 3′-diversification of oligonucleotides. Reagents and conditions: (a) oligo-nucleotide (10 µM in PBS buffer), 50 eq. NaIO4 (500 µM), 0 °C, 1 h; (b) condensation: 100 eq. Na2SO3 (1 mM), 100 eq. nucleophile, RT, 16 h; reductive amination: 200 eq. Na2SO3 (2 mM), 100 eq. nucleophile, 1000 eq. NaBH3CN, 37 °C, 16 h. oligonucleotide sequences: 5-/5Phos/r(GA AAA AAA AAA)-3 or 5-/5Phos/r(GU/iFluorT/UCG CCA UU/ i6-TAMN/AAA AAA AAA A-3), iFluorT: fluoresceinmodified T; i6-TAMN: 6-carboxytetramethylrhodamine-modified T. The fluorophores were chosen to engage in Förster resonance energy transfer (FRET), to enable monitoring of integrity of the strand (vide infra). Reactions were purified by desalting columns. Reaction progress was monitored via ion pairing chromatography (IPC). Final concentrations in the reaction solution are given.
Figure 3. 3′-Modification of gaussia luciferase mRNA results in increased expression in HEK293-cells. 30 ng of capped, 3′-modified gLuc mRNA was transfected into HEK293 cells using DharmaFECT. Luminescence was read after 24 and 48 h and plotted normalized to the luminescence of the unmodified gLuc at the 24 h time point. Legend indicates nature of the nucleophile used in the modification of the oligonucleotide.
Figure 2. 3′-Selective modification increases RNA half-life in HeLa cell extract. An oligonucleotide containing a FRET-pair [sequence: 5-/5Phos/r(GU/iFluorT/ UCG CCA UU/i6-TAMN/ AAA AAA AAA A)-3]8 was incubated with varying concentrations of S100 HeLa cell extract, and degradation was monitored using fluorescence. Legend indicates nature of the nucleophile used in the oligonucleotide modification.
almost 3-fold increased expression levels after 24 h, and translation was sustained for longer than 48 h. These findings suggest that 3′-mRNA modification is a valid approach to stabilize mRNA transcripts and boost expression levels. Thus, 3′-RNA modification may extend the halflife of a therapeutic mRNA transcript and could be important for mRNA gene therapy development.
transfected into HEK293 cells and mRNA expression was assessed using luminescence readout. After 24 h, modified mRNAs showed 2.5 to 3times higher expression levels than unmodified control Fig. 3. Notably, 3′-modification resulted in sustained expression past the 48 h time point, at which translation of the parent mRNA had dropped below 10%. These results support our hypothesis that increased mRNA stability results in increased and prolonged expression in cells. In conclusion, we showed that oligonucleotides can quantitatively and selectively be 3′-modified using an optimized periodate cleavage/ condensation protocol. This modification results in drastically increased RNA stability in HeLa cell extracts. 3′-Modified gLuc mRNA showed
Acknowledgments We thank Jared Ek, Carmelina Rakiec and Jennifer Porier for their help in setting up the LC-IPC methods. Dr. Shari Caplan is acknowledged for the providing the GLuc constructs. 2452
Bioorganic & Medicinal Chemistry Letters 28 (2018) 2451–2453
C. Gampe et al.
A. Supplementary data
b) Wu M, Reuter M, Lilie H, Liu Y, Wahle E, Song H. EMBO J. 2005;23:4082. 6. a) Imai J, Johnston MI, Torrence PF. J. Biol. Chem. 1982;257:12739;
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.bmcl.2018.06.008.
b) Wu TP, Ruan KC, Liu WY. Nucleic Acids Res. 1996;24:3472; c) Proudnikov D, Mirzabekov A. Nucleic Acids Res. 1996;24:4535;
References
d) Bellon L, Workman C, Scherrer J, Usman N, Wincott F. J Am Chem Soc. 1996;118:3771;
1. Wolff JA, Malone RW, Williams P, et al. Science. 1990;247:1465. 2. Weissman D. Expert Rev Vaccines. 2015;14:265. 3. a) Anderson BR, Muramatsu H, Nallagatla SR, et al. Nucleic Acid Res. 2010;38:5884;
e) Gite S, RajBhandary UL. J Biol Chem. 1997;272:5305; f) Wang Z, Chen L, Bayly SF, Torrence PF. Bioorg Med Chem Lett. 2000;10:1357;
b) Andries O, Mc Cafferty S, De Smedt SC, Weiss R, Sanders NN, Kitada T. J Controlled Release. 2015;217:337;
g) Nandanan E, Jang SY, Moro S, et al. J Med Chem. 2000;43:829; h) Kurata S, Ohtsuki T, Suzuki T, Wantanabe K. Nucleic Acid Res. 2003;31:e145;
c) Svitkin YV, Cheng YM, Chakraborty T, Presnyak V, John M, Sonenberg N. Nucleic Acids Res. 2017;45:6023. 4. a) Moore M. Science. 2005;309:1514;
i) Zhou L, Thakur CS, Molinaro RJ, et al. Bioorg Med Chem Lett. 2006;14:7862. 7. Schmidt P, Zhou L, Tishinov K, Zimmermann K, Gillingham D. Angew Chem Int Ed. 2014;53:10928. 8. Uhler SA, Cai D, Man Y, Figge C, Walter NG. J Am Chem Soc. 2003;125:14230.
b) Chen CY, Shyu AB. Wiley Interdiscp Rev RNA. 2011;2:167. 5. a) Mitchell P, Tollervey D. Curr Opin Genet Dev. 2000;10:193;
2453