Synthesis of (R)-Configured 2′-Fluorinated mC, hmC, fC, and caC Phosphoramidites and Oligonucleotides

Article abstract of DOI:10.1021/acs.orglett.6b02110

Investigation of the function of the new epigenetic bases requires the development of stabilized analogues that are stable during base excision repair (BER). Here we report the synthesis of 2′-(R)-fluorinated versions of the phosphoramidites of 5-methylcy

Full text of DOI:10.1021/acs.orglett.6b02110

Letter
pubs.acs.org/OrgLett
Synthesis of (R)‑Configured 2′-Fluorinated mC, hmC, fC, and caC
Phosphoramidites and Oligonucleotides
Arne S. Schroder, Olga Kotljarova, Edris Parsa, Katharina Iwan, Nada Raddaoui, and Thomas Carell*
̈
Center for Integrated Protein Science, Department of Chemistry, Ludwig-Maximilians-Universitat Munchen, Butenandtstraße 5-13,
̈
̈
81377 Munich, Germany
S
* Supporting Information
ABSTRACT: Investigation of the function of the new epigenetic bases requires the development of stabilized analogues that are
stable during base excision repair (BER). Here we report the synthesis of 2′-(R)-fluorinated versions of the phosphoramidites of
5-methylcytosine (mC), 5-hydroxymethylcytosine (hmC), 5-formylcytosine (fC), and 5-carboxycytosine (caC). For
oligonucleotides containing 2′-(R)-F-fdC, we show that these compounds cannot be cleaved by the main BER enzyme
thymine-DNA glycosylase (TDG).
luorine is an element that is used in medicinal chemistry to
Freplace H atoms in pharmaceutically active molecules with
astonishing effects. Fluorine substitution stabilizes molecules to
extend their lifetimes in the bloodstream, and often it increases
the affinities of molecules for their biological targets by
increasing their lipophilicities.1 In nucleoside chemistry, for
example, fluorination of dC at the 2′ position creates molecules
like gemcitabine (1), which are used as antimetabolites in
modern cancer therapy.2 The 2′-F substitution has several
effects. Most importantly, a 2′-(R)-configuration as in 2′-(R)-F-
dC (2) stabilizes the C3′-endo conformation of the ribose sugar
so that the base becomes RNA-like.3 A fluorine at C2′ also
blocks the activity of glycosylases, thereby stabilizing the base
during base excision repair (BER).4 We are currently
investigating the chemistry that occurs at the nucleoside 2′-
deoxycytidine (dC, 3) that leads to the formation and removal
of the methylated and subsequently oxidized epigenetic dC
derivatives 5-methyl- (mdC, 4), 5-hydroxymethyl- (hmdC, 5),
Figure 1. Overview of epigenetically relevant nucleosides and 2′-fluoro
5-formyl- (fdC, 6), and 5-carboxy-2′-deoxycytidine (cadC, 7)
(Figure 1).5 Nucleosides 5−7 are products of consecutive
enzymatic oxidation of 4 by the action of ten-eleven-
nucleosides that are important in this context.
the 2′-(R)-fluorinated versions of mdC (8), hmdC (9), fdC
(10), and cadC (11). We have developed phosphoramidite
building blocks for the incorporation of these bases into DNA
strands, and we show that these nucleosides are indeed stable
during BER. With the plan in mind to investigate epigenetic
processes directly in the genome of stem cells, we realized that
the 2′-arabino-configured compound 2′-(S)-F-dC (12) might
be too toxic. Indeed, when we evaluated the toxicity of the ribo-
translocation enzymes (Tet enzymes), which use molecular
oxygen and α-ketoglutarate to perform the oxidation
chemistry.6 Current data suggest that fdC and cadC are
removed from the genome by BER via the enzyme thymine-
DNA glycosylase (TDG).5d,7 Other data predict that the bases
may undergo some kind of deformylation/decarboxylation
reaction, which would convert fdC and cadC directly back into
the canonical base dC.8 In order to distinguish these processes,
it is important to have tool molecules that cannot be repaired
by BER. This would allow one to decipher chemical processes
at fdC and cadC beyond BER. Here we report the synthesis of
Received: July 20, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.6b02110
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
configured compound 2′-(R)-F-dC against 12 in stem cells (see
the Supporting Information), we noted a strongly reduced
toxicity for 2′-(R)-F-dC. This is already interesting because it is
believed that the 2′-(S)-F configuration has a much smaller
impact on the overall DNA structure.3,9 Our stem cell data are,
however, in full agreement with toxicity studies in rats and
woodchucks showing that feeding of 2′-(R)-F-dC at up to 500
mg kg−1 day−1 is possible without considerable toxicity
effects.10
14, which was converted into the 2′-(R)-F-mdC phosphor-
amidite building block 15 using standard procedures.13
For the synthesis of the 2′-(R)-F-hmdC phosphoramidite 18,
we started from intermediate 13. Carbonylative Stille coupling
with tributyltin hydride and reduction of the formyl group
under Luche conditions yielded 2′-(R)-F-hmdC derivative 16.14
The exocyclic amine together with the hydroxyl group was
protected as a carbamate using p-nitrophenyl chloroformate.15
Efficient conversion required full deprotonation of both
functional groups with NaH prior to addition of the protecting
reagent. Final silyl deprotection, DMT protection, and
synthesis of the hmdC phosphoramidite building block 18
with Bannwarth’s reagent furnished the 2′-(R)-F-hmC
phosphoramidite in high yield (34% over six steps from 13).
Regarding 2′-(R)-F-fdC phosphoramidite building block 21,
we performed a carbonylative Stille coupling reaction of 13
with tributyltin hydride (see Scheme 2). Subsequent masking of
For the synthesis of the 2′-(R)-F-xdC nucleosides and
phosphoramidites 15, 18, 21, and 24 (see Scheme 1), we
Scheme 1. Synthesis of 2′-(R)-F-mdC and 2′-(R)-F-hmdC
Phosphoramidite Building Blocks 15 and 18
Scheme 2. Synthesis of 2′-(R)-F-fdC and 2′-(R)-F-cadC
Phosphoramidite Building Blocks 21 and 24
started with 2′-(R)-F-dC (2), which was iodinated at C5 with
elemental iodine and m-CPBA.11 Subsequent silylation yielded
TBS-protected 5-iodo-2′-(R)-F-dC 13. The needed methyl-
ation was best carried out under Kumada conditions with
trimethylaluminum.12 This furnished the 2′-(R)-F-mdC com-
pound in 79% yield. Notably, the use of other methyl-
transferring agents such as MeMgCl resulted in a 1:1 mixture of
methylated and dehalogenated products. We believe that the
exocyclic amine requires complete deprotonation to avoid a
1,3-proton shift from the exocyclic amine to the Pd-activated
C5-position. Further protection with BzCl and silyl depro-
tection with Olah’s reagent furnished 2′-(R)-F-mdC derivative
the formyl group as a 1,3-dioxane unit with 1,3-propanediol and
TiCl4 as the activating Lewis acid provided compound 19. For
the protection of the exocyclic amine, we chose p-
MeOC6H4COCl as recently reported.16 The electron-pushing
methoxy unit strongly enhances the stability of the amine
protecting group during solid-phase DNA synthesis, and this is
B
DOI: 10.1021/acs.orglett.6b02110
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
strictly required in order to obtain oligonucleotides in high
yields. Again, satisfactory yields were obtained only when the
exocyclic amine was deprotonated with NaH prior to addition
of p-MeOC6H4COCl. Final silyl deprotection yielded 20, which
was converted into 2′-(R)-F-fdC phosphoramidite building
block 21 using standard procedures.
Starting from intermediate 13, we next developed the
synthesis of the 2′-(R)-F-cadC phosphoramidite building
block. The synthesis of the methyl ester was achieved using
Pd0-mediated CO insertion in methanol.17 Because of the
electron-withdrawing nature of the ester moiety, we decided to
use p-MeOC6H4COCl for stable protection of the exocyclic
amine. Conversion of 23 using standard procedures delivered
2′-(R)-F-cadC phosphoramidite building block 24 in just five
steps in an overall yield of 36% starting from 13.
phase HPLC directly after deprotection showed in all cases just
one major product. After purification, the corresponding
oligonucleotides were obtained in 20−52% yield and high
purity (>95%). MALDI-TOF/MS spectra showed the expected
masses, confirming the presence of the 2′-(R)-F-xdC bases in
the ODNs. In summary, the synthesized 2′-(R)-F-xdC
phosphoramidite building blocks enabled the synthesis of
oligonucleotides containing the corresponding fluorinated
nucleosides.
We next started to evaluate the extent to which the 2′-(R)-F
substitution would affect typical epigenetic processes. First, we
wanted to know whether the H-to-F chemical mutation
influences the activity of methyltransferases (see Figure 3).
To examine the ability to prepare oligonucleotides
containing 2′-(R)-F-xdC, we prepared the corresponding
ODN1a−d (see Figure 2). The modified nucleotides were
Figure 3. (A) Sequences of the synthesized ODN2 and ODN3 with
incorporation of dC or 2′-(R)-F-dC nucleoside. (B) The methylation
assay of ODN2 and ODN3 with methyltransferase M.SssI showed that
the fluoro label in 2′-(R)-F-dC has no influence on the level of
methylation.
To study this, we synthesized ODN2 having either dC or 2′-
(R)-F-dC in a CpG context. After hybridization of ODN2 with
ODN3, they were incubated with methyltransferase M.SssI. To
determine the level of mdC or 2′-(R)-F-mdC, we digested the
DNA strands to the nucleoside level and performed UHPLC-
MS/MS (QQQ) analysis. As the verification of our hypothesis,
we observed methylation of dC (48%) and 2′-(R)-F-dC (50%).
This demonstrates that the 2′-(R)-F substitution does not affect
the native behavior of the DNA and that 2′-(R)-F-xdC
nucleosides are suitable tools for the investigation of the active
demethylation beyond base excision repair.
Figure 2. (A) Sequence of the synthesized ODN1a−d with
incorporation of the corresponding 2′-(R)-F-xdC phosphoramidite
building blocks. (B−D) Reversed-phase HPL chromatograms and
MALDI-TOF data for the corresponding purified ODN1a−d after
basic and, in the case of 2′-(R)-F-fdC, acidic cleavage from the resin
and deprotection.
In 2011 and 2012, the groups of Drohat7a and Cheng19
showed that fdC and cadC are excised by human TDG
(hTDG). Previously, glycosylase activity was blocked with
fluorinated DNA bases (2′-F-(S)-cadC, 2′-F-(S/R)-dU).4a,c In
order to determine whether the 2′-(R)-F-fdC compounds
would block hTDG activity, we synthesized oligonucleotides
ODN4 with either the fdC or F-fdC nucleoside at a central
position and hybridized the strands to the complementary
oligonucleotide ODN5. After hybridization and incubation with
hTDG, the DNA strand was treated with piperidine.7a,19,20
Subsequently, we analyzed the products by HPLC (see Figure
4). As expected, we detected complete strand cleavage for the
fdC-containing ODN4. However, in the case of the ODN4
containing 2′-(R)-F-fdC, we did not observe any strand
cleavage products. Thus, we proved that the 2′-(R)-F label
indeed inhibits the hTDG activity, blocking BER of fdC.
In summary, we have synthesized 2′-(R)-F phosphoramidite
building blocks of the epigenetically relevant nucleosides. These
building blocks enabled the synthesis of oligonucleotides
placed in a CpG context. The solid-phase syntheses were
performed using standard phosphoramidite conditions.18 For
the 2′-(R)-F nucleosides, the coupling times were increased
from 30 to 180 s to ensure good coupling yields. For
deprotection of the oligonucleotides containing 2′-(R)-F-mdC
and 2′-(R)-F-fdC, including cleavage from the solid support, we
first treated the solid-phase material with saturated aqueous
ammonia solution (18 h, 25−28 °C). Subsequently, the
oligonucleotide containing 2′-(R)-F-fdC was exposed to
aqueous acetic acid (80%) at 20 °C until MALDI-TOF/MS
analysis indicated complete hydrolysis of the 1,3-dioxane unit
(∼6 h). Because of the carbamate and ester units, the
oligonucleotides containing 2′-(R)-F-hmdC and 2′-(R)-F-
cadC were deprotected with NaOH (0.4 M in 4:1 methanol/
water) for 18 h. This procedure avoided the formation of
aminomethyl and amide moieties.11,17a Analytical reversed-
C
DOI: 10.1021/acs.orglett.6b02110
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Tereshko, V.; Ikeda, H.; Marquez, V. E.; Egli, M. Nucleic Acids Res.
1998, 26, 2473−2480.
(4) (a) Dai, Q.; Lu, X.; Zhang, L.; He, C. Tetrahedron 2012, 68,
5145−5151. (b) Schaerer, O. D.; Verdine, G. L. J. Am. Chem. Soc.
1995, 117, 10781−10782. (c) Scharer, O. D.; Kawate, T.; Gallinari, P.;
̈
Jiricny, J.; Verdine, G. L. Proc. Natl. Acad. Sci. U. S. A. 1997, 94, 4878−
4883.
(5) (a) Kriaucionis, S.; Heintz, N. Science 2009, 324, 929−930.
(b) Tahiliani, M.; Koh, K. P.; Shen, Y.; Pastor, W. A.; Bandukwala, H.;
Brudno, Y.; Agarwal, S.; Iyer, L. M.; Liu, D. R.; Aravind, L.; Rao, A.
Science 2009, 324, 930−935. (c) Pfaffeneder, T.; Hackner, B.; Truß,
M.; Munzel, M.; Muller, M.; Deiml, C. A.; Hagemeier, C.; Carell, T.
̈
̈
Angew. Chem., Int. Ed. 2011, 50, 7008−7012. (d) He, Y.-F.; Li, B.-Z.;
Li, Z.; Liu, P.; Wang, Y.; Tang, Q.; Ding, J.; Jia, Y.; Chen, Z.; Li, L.;
Sun, Y.; Li, X.; Dai, Q.; Song, C.-X.; Zhang, K.; He, C.; Xu, G.-L.
Science 2011, 333, 1303−1307.
(6) (a) Loenarz, C.; Schofield, C. J. Chem. Biol. 2009, 16, 580−583.
(b) Ito, S.; Shen, L.; Dai, Q.; Wu, S. C.; Collins, L. B.; Swenberg, J. A.;
He, C.; Zhang, Y. Science 2011, 333, 1300−1303.
(7) (a) Maiti, A.; Drohat, A. C. J. Biol. Chem. 2011, 286, 35334−
35338. (b) Raiber, E.-A.; Beraldi, D.; Ficz, G.; Burgess, H. E.; Branco,
M. R.; Murat, P.; Oxley, D.; Booth, M. J.; Reik, W.; Balasubramanian,
S. Genome Biol. 2012, 13, R69. (c) Zhang, L.; Lu, X.; Lu, J.; Liang, H.;
Dai, Q.; Xu, G.-L.; Luo, C.; Jiang, H.; He, C. Nat. Chem. Biol. 2012, 8,
328−330.
Figure 4. (A) Sequence of synthesized ODN4 and ODN5 for the
hTDG glycosylation assay. (B) Reversed-phase HPL chromatogram of
the hTDG glycosylation assay.
containing 2′-(R)-F-mdC, 2′-(R)-F-hmdC, 2′-(R)-F-fdC, and
2′-(R)-F-cadC in high yields and quality. Furthermore, we
showed that a 2′-(R)-F label on fdC blocks the activity of the
critical TDG enzyme, thus inhibiting base excision repair of this
base. The 2′-(R)-F label is consequently the ideal tool for
analysis of the epigenetic metabolism beyond base excision.
(8) (a) Schiesser, S.; Hackner, B.; Pfaffeneder, T.; Muller, M.;
̈
Hagemeier, C.; Truss, M.; Carell, T. Angew. Chem., Int. Ed. 2012, 51,
̌ ̌
6516−6520. (b) Liutkeviciute, Z.; Kriukiene, E.; Licyte, J.; Rudyte, M.;
̇ ̇ ̇ ̇
̅
Urbanavici
̌
ute, G.; Klimasa
̌
uskas, S. J. Am. Chem. Soc. 2014, 136,
5884−5887. (c) Schiesser, S.; Pfaffeneder, T.; Sadeghian, K.; Hackner,
B.; Steigenberger, B.; Schroder, A. S.; Steinbacher, J.; Kashiwazaki, G.;
̇
̅
̈
Hofner, G.; Wanner, K. T.; Ochsenfeld, C.; Carell, T. J. Am. Chem. Soc.
̈
ASSOCIATED CONTENT
* Supporting Information
2013, 135, 14593−14599.
■
(9) Lee, S.; Bowman, B. R.; Ueno, Y.; Wang, S.; Verdine, G. L. J. Am.
Chem. Soc. 2008, 130, 11570−11571.
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.6b02110.
(10) (a) Richardson, F. C.; Tennant, B. C.; Meyer, D. J.; Richardson,
K. A.; Mann, P. C.; Mcginty, G. R.; Wolf, J. L.; Zack, P. M.; Bendele, R.
A. Toxicol. Pathol. 1999, 27, 607−617. (b) Kierdaszuk, B.; Krawiec, K.;
Kazimierczuk, Z.; Jacobsson, U.; Johansson, N. G.; Munch-petersen,
B.; Eriksson, S.; Shugar, D. Nucleosides Nucleotides 1999, 18, 1883−
1903.
Synthesis of all compounds and oligonucleotides and
details of the assays (PDF)
(11) Munzel, M.; Globisch, D.; Trindler, C.; Carell, T. Org. Lett.
̈
AUTHOR INFORMATION
Corresponding Author
2010, 12, 5671−5673.
■
(12) Hirota, K.; Kitade, Y.; Kanbe, Y.; Isobe, Y.; Maki, Y. Synthesis
1993, 1993, 213−215.
*E-mail: Thomas.Carell@lmu.de. Website: http://www.
carellgroup.de.
Notes
(13) Bannwarth, W.; Trzeciak, A. Helv. Chim. Acta 1987, 70, 175−
186.
(14) Luche, J. L. J. Am. Chem. Soc. 1978, 100, 2226−2227.
(15) Sammet, B. Synlett 2009, 2009, 3050−3051.
The authors declare no competing financial interest.
(16) Schroder, A. S.; Steinbacher, J.; Steigenberger, B.; Gnerlich, F.
̈
A.; Schiesser, S.; Pfaffeneder, T.; Carell, T. Angew. Chem., Int. Ed. 2014,
53, 315−318.
ACKNOWLEDGMENTS
We thank Nadine Hinkel, Korbinian Krieger, Stefan Marchner,
and Robert Rampmaier (all Ludwig-Maximilians-Universitat
Munchen) for practical assistance. Furthermore, we thank the
Deutsche Forschungsgemeinschaft (SFB749), the Fonds der
Chemischen Industrie (predoctoral fellowship for A.S.S.), and
the Excellence Cluster (CiPSM, EXC114 and GRK 2062) for
support.
■
(17) (a) Munzel, M.; Lischke, U.; Stathis, D.; Pfaffeneder, T.;
̈
̈
Gnerlich, F. A.; Deiml, C. A.; Koch, S. C.; Karaghiosoff, K.; Carell, T.
Chem. - Eur. J. 2011, 17, 13782−13788. (b) Nomura, Y.; Haginoya, N.;
Ueno, Y.; Matsuda, A. Bioorg. Med. Chem. Lett. 1996, 6, 2811−2816.
(18) (a) Beaucage, S. L.; Caruthers, M. H. Tetrahedron Lett. 1981, 22,
̈
1859−1862. (b) Sinha, N. D.; Biernat, J.; McManus, J.; Koster, H.
̈
Nucleic Acids Res. 1984, 12, 4539−4557.
(19) Hashimoto, H.; Hong, S.; Bhagwat, A. S.; Zhang, X.; Cheng, X.
Nucleic Acids Res. 2012, 40, 10203−10214.
(20) Bennett, M. T.; Rodgers, M. T.; Hebert, A. S.; Ruslander, L. E.;
Eisele, L.; Drohat, A. C. J. Am. Chem. Soc. 2006, 128, 12510−12519.
REFERENCES
■
(1) (a) Gillis, E. P.; Eastman, K. J.; Hill, M. D.; Donnelly, D. J.;
Meanwell, N. A. J. Med. Chem. 2015, 58, 8315−8359. (b) Purser, S.;
Moore, P. R.; Swallow, S.; Gouverneur, V. Chem. Soc. Rev. 2008, 37,
320−330.
(2) Parker, W. B. Chem. Rev. 2009, 109, 2880−2893.
(3) (a) Sinnott, M. L. Chem. Rev. 1990, 90, 1171−1202. (b) Williams,
A. A.; Darwanto, A.; Theruvathu, J. A.; Burdzy, A.; Neidigh, J. W.;
Sowers, L. C. Biochemistry 2009, 48, 11994−12004. (c) Berger, I.;
D
DOI: 10.1021/acs.orglett.6b02110
Org. Lett. XXXX, XXX, XXX−XXX