Synthesis of a DNA promoter segment containing all four epigenetic nucleosides: 5-methyl-, 5-hydroxymethyl-, 5-formyl-, and 5-carboxy-2′- deoxycytidine

Article abstract of DOI:10.1002/anie.201308469

A 5-formyl-2′-deoxycytidine (fdC) phosphoramidite building block that enables the synthesis of fdC-containing DNA with excellent purity and yield has been developed. In combination with phosphoramidites for 5-methyl-dC, 5-hydroxymethyl-dC, and carboxy-dC, it was possible to prepare a segment of the OCT-4 promoter that contains all four epigenetic bases. Because of the enormous interest in these new epigenetic bases, the ability to insert all four of them into DNA should be of great value for the scientific community. Biologically relevant promoter segments with all four epigenetic nucleosides are finally accessible by solid-phase synthesis. With the fdC-phosphoramidite building block introduced herein, no side reactions of 5-formylcytosine were observed. Furthermore, mild deprotection conditions prevented oxidative or reductive lesions, which was proven by detailed mass spectrometric analysis. Copyright

Full text of DOI:10.1002/anie.201308469

Angewandte
Chemie
DOI: 10.1002/anie.201308469
Epigenetics
Hot Paper
Synthesis of a DNA Promoter Segment Containing All Four Epigenetic
Nucleosides: 5-Methyl-, 5-Hydroxymethyl-, 5-Formyl-, and 5-Carboxy-
2’-Deoxycytidine**
Arne S. Schrçder, Jessica Steinbacher, Barbara Steigenberger, Felix A. Gnerlich, Stefan Schiesser,
Toni Pfaffeneder, and Thomas Carell*
Abstract: A 5-formyl-2’-deoxycytidine (fdC) phosphoramidite
building block that enables the synthesis of fdC-containing
DNA with excellent purity and yield has been developed. In
combination with phosphoramidites for 5-methyl-dC,
5-hydroxymethyl-dC, and carboxy-dC, it was possible to
prepare a segment of the OCT-4 promoter that contains all
four epigenetic bases. Because of the enormous interest in these
new epigenetic bases, the ability to insert all four of them into
DNA should be of great value for the scientific community.
A
side from the four canonical nucleosides, the genome
contains several DNA modifications that are of epigenetic
importance. The nucleoside 2’-deoxycytidine is often methy-
lated at the C5 position to give 5-methyl-2’-deoxycytidine (1,
mdC), which is involved in epigenetic gene silencing.[1] In
recent years, the oxidation products of mdC were discov-
ered.[2–5] Ten-eleven-translocation proteins (TET1–3) are now
known to oxidize mdC to 5-hydroxymethyl- (2, hmdC),
5-formyl- (3, fdC), and 5-carboxy-2’-deoxycytidine (4,
cadC).[6,7] These “new” modifications (Figure 1A) are pri-
marily found in the CpG context in DNA promoter sequen-
ces, where they seem to be involved in regulating gene
expression.[8] Lately, the synthesis and analysis of a tRNA
segment of human mitochondrial tRNAMet that carries 5-
formylcytidine in the anticodon loop was reported.[9]
Figure 1. A) The recently found epigenetically relevant nucleosides
5-methyl- (1, mdC), 5-hydroxymethyl- (2, hmdC), 5-formyl- (3, fdC),
and 5-carboxy-2’-deoxycytidine (4, cadC). B) Sequence of an OCT-4
promoter segment, in which cytidines in CpG contexts were substi-
tuted with the epigenetic nucleosides 1–4.
The “Achillesꢀ heel” of the synthesis of DNA strands
containing all new epigenetic nucleosides is the incorporation
of fdC because of the high reactivity of the formyl group.[11–13]
The only currently available fdC building block is inserted as
the 1,2-diol precursor, and periodate treatment is required to
obtain the formyl moiety;[14] this reagent is, for instance,
incompatible with hmdC.[15] Moreover, the periodate-based
deprotection promotes side reactions that prevent the syn-
thesis of DNA strands that contain multiple fdCs. Conse-
quently, DNA strands that contain the four epigenetic
nucleosides mdC, hmdC, fdC, and cadC at defined sites are
synthetically not accessible.[16]
Herein, we report the development of a new fdC
phosphoramidite building block that can be inserted multiple
times, or even consecutively, into DNA strands by using
standard phosphoramidite chemistry. High coupling yields
during solid-phase synthesis as well as mild deprotection
conditions even enabled the preparation of a OCT-4 pro-
moter sequence that contains all four epigenetic nucleosides
(Figure 1B, ODN1).
To study the function of these new nucleosides, it is of
utmost importance to develop methods that enable the
synthesis of DNA strands that contain hmdC, fdC, and
cadC. Recently, we reported the enzymatic incorporation of
these nucleosides into DNA by means of their triphos-
phates.[10] However, a sequence-specific incorporation of the
nucleosides, particularly into CpG islands, which would be
required for proteomics studies, is not possible with this
method.
[*] A. S. Schrçder, J. Steinbacher, B. Steigenberger, F. A. Gnerlich,
S. Schiesser, T. Pfaffeneder, Prof. Dr. T. Carell
Center for Integrated Protein Science
Department fꢀr Chemie, Ludwig-Maximilians-Universitꢁt
Butenandtstrasse 5–13, 81377 Mꢀnchen (Germany)
E-mail: Thomas.Carell@lmu.de
Homepage: http://www.carellgroup.de
[**] We thank the Deutsche Forschungsgemeinschaft (SFB749 and
SFB646) for financial support. Additional support from the Fonds
der Chemischen Industrie (predoctoral fellowships for A.S.S, S.S.,
T.P.) and two excellence clusters (CiPSM, EXC114) is acknowledged.
The electron-poor heterocycle of 3 is a major chemical
obstacle for fdC incorporation, as this moiety promotes
glycosidic bond cleavage even under the mildly acidic
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201308469.
Angew. Chem. Int. Ed. 2014, 53, 315 –318
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
315

.
Angewandte
Communications
conditions that are commonly used for oligonucleotide syn-
thesis. Furthermore, the formyl group of 3 is sensitive to
oxidation, and, most importantly, 5-formylcytosine is readily
attacked by nucleophiles, particularly at the C6 position
(Figure 1A).[16] To mask the reactivity of the aldehyde
functionality, we examined various protecting groups (for
details, see the Supporting Information). Finally, the best
results were obtained when the aldehyde is converted into
a 1,3-dioxane unit with propane-1,3-diol. This acetal turned
out to be sensitive enough for later cleavage under mild
conditions (80% aqueous acetic acid, 208C); nevertheless, it
is sufficiently stable to remain intact during solid-phase DNA
synthesis. However, the temperature at which the deprotec-
tion is carried out at the end of the synthesis is a crucial
parameter. Cleavage at 158C is too slow, whereas higher
temperatures (ꢀ 258C) lead to a significant amount of
glycosidic bond cleavage, mostly at fdC (for details, see the
Supporting Information). Therefore, deprotection of the fdC
acetal has to occur at approximately 208C. The amino group
at the C4 position was protected with 4-methoxybenzoyl
chloride, which is compatible with the acetal moiety. For fdC,
we discovered that the commonly used benzoyl group is
already efficiently cleaved during the solid-phase DNA
synthesis, which gives rise to branched DNA products.
The synthesis of the fdC building block was achieved
starting from tert-butyldimethylsilyl (TBS)-protected fdC 5,
which was synthesized from 5-iodo-dC according to literature
procedures (Scheme 1).[17] Notably, standard procedures for
acetal protection did not provide the desired product 6.
Activation of the formyl group is strictly required, and TiCl4
turned out to be the only Lewis acid that gave acceptable
yields. Subsequent protection of the amino group at the
C4 position with 4-methoxybenzoyl chloride and silyl depro-
tection furnished the fdC derivative 7, which was converted
into the fdC phosphoramidite building block 8 by standard
procedures.[18]
Scheme 1. Synthesis of acetal-protected fdC phosphoramidite 8.
To examine the quality of the new fdC building block 8,
we prepared oligonucleotide ODN2, which contains five fdCs
(Figure 2B). The solid-phase synthesis was performed using
standard phosphoramidite conditions.[19,20] For fdC, the cou-
pling times were increased from 30 to 60 seconds to ensure
maximum coupling yields. For deprotection of the standard
nucleosides and cleavage of the oligonucleotide from the solid
support, we first treated the solid-phase material with
saturated aqueous NH4OH solution (24 h, 258C). The HPL
chromatogram depicted in Figure 2B shows the result for the
crude product of the synthesis of the dimethoxytrityl (DMT)-
protected oligonucleotide. In a second deprotection step, the
oligonucleotide was treated with acetic acid for six hours at
208C, which cleaved the acetal and DMT protecting groups.
The HPL chromatogram that was obtained for the unpurified
DNA strand ODN2, which contains five fdCs, is shown in
Figure 2C. As only one peak was observed, the synthesis had
occurred in high yield.
Figure 2. A) Sequence of the synthesized ODN2 with fivefold incorpo-
ration of fdC using 8. B) Reversed-phase HPL chromatogram of crude
ODN2 directly after basic cleavage from the resin (NH4OH, 23 h,
258C, 0–80% buffer B in 45 min). C) Reversed-phase HPL chromato-
gram of crude ODN2 directly after acidic cleavage of the DMT and
acetal groups (80% aq. acetic acid, 6 h, 208C, 0–40% buffer B in
45 min). D) MALDI-TOF spectrum of fully deprotected, but crude
ODN2. E) UHPL chromatogram and quantification data of unpurified
digested ODN2.
To verify that five fdC nucleosides are indeed present in
the DNA strand, we measured a MALDI-TOF spectrum
(Figure 2D), which showed the expected exact mass. Further
proof for correct incorporation was obtained by digestion of
the synthesized, but still crude oligonucleotide with nucle-
316
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 315 –318

Angewandte
Chemie
ase S1 and snake-venom phosphodiesterase I to give the 5’-
monophosphates. These were further hydrolyzed to the
nucleosides by antarctic phosphatase. The obtained nucleo-
side mixture was analyzed by UHPLC-MS/MS (QQQ; Fig-
ure 2E). The obtained UV chromatogram was clean and
showed signals only for the expected nucleosides. The fdC
nucleoside was identified based on the retention time and its
characteristic MS/MS fragmentation (m/z = 256.1!140.1).
Next, we quantified the amount of dC, fdC, and dT (Fig-
ure 2E, inset) by using synthetic isotopologues of the nucleo-
sides as internal standards, which is a method that was
recently described.[16] We obtained the expected dC/fdC ratio
of approximately 4:5, which is in full agreement with the
sequence. The fact that these values are obtained with the
crude synthesis product underpins the superior properties of
the new fdC phosphoramidite. Most importantly, we did not
detect any a-fdC nucleosides, which indicates that b- to a-
anomerization was fully suppressed with the new building
block. We also investigated whether hmdC and cadC are
present during DNA synthesis; these are the reduction and
oxidation products of fdC, which might be formed during the
iodine oxidation step. As expected, none of these compounds
were detected (Figure 2E, inset). In summary, the fdC
phosphoramidite 8 enabled the synthesis of fdC-containing
oligonucleotides. With this building block in hand, we have
established the basis for the synthesis of oligonucleotides that
contain all four epigenetic nucleosides.
Figure 3. A) Used phophoramidites. B) Sequence of the synthesized
OCT-4 promoter segment, in which cytidines in CpG contexts were
substituted with the epigenetic nucleosides 1–4. C) Reversed-phase
HPL chromatogram of crude and purified (inset) ODN1 after complete
deprotection (0–40% buffer B in 45 min). D) UHPL chromatogram
and quantification data of digested ODN1. Bz=benzoyl.
Therefore, we then synthesized oligonucleotide ODN1
with the sequence of a segment of the OCT-4 promoter (mus
musculus, chromosome 17, 35505895-35505943; Figure 3B).
All cytidines in the respective CpG contexts were substituted
with the epigenetic nucleosides 1–4. The best results were
obtained with the cyclic-carbamate-protected hmdC phos-
phoramidite (for detailed conditions of the oligonucleotide
synthesis, see the Supporting Information).[17] This carbamate
building block and the ester-protected cadC unit forced us to
use NaOH (0.4m in methanol/water (4:1)) instead of NH4OH
in the first deprotection step to avoid the formation of amide
moieties (Figure 3A).[12,17] After treatment of the synthesized
ODN1 with NaOH, the oligonucleotide was precipitated with
ethanol. We next performed the second deprotection step
with acetic acid. Figure 3C shows the HPL chromatogram of
the crude material after the second deprotection step. It is
clearly visible that the process yielded ODN1 as the major
product, despite the fact that ODN1 is a 49 mer oligonucleo-
tide that contains four modified nucleosides (for detailed data
of crude ODN1, see the Supporting Information). Purifica-
tion by HPLC after the first deprotection step and subsequent
acidic treatment as described above gave pure ODN1 in 31%
(Figure 3C, inset; see also the Supporting Information). The
oligonucleotide was fully digested, and the product of
hydrolysis was analyzed by UHPLC-MS/MS, which corrobo-
rated the presence of all incorporated nucleosides. Isotope-
based quantification confirmed the amounts of the epigenetic
nucleosides relative to dC to be 1.2:1.2:0.9:1.0:15.0, which is
in full agreement with the sequence (Figure 3B,D).
fdC. In combination with phosphoramidite building blocks for
mdC, hmdC, and cadC, it is now possible to synthesize
oligonucleotides that contain all four epigenetically important
cytidine nucleosides 1–4. This will pave the way for a more
detailed analysis of how these nucleosides influence biolog-
ical processes and stem-cell development.[8]
Received: September 28, 2013
Published online: November 26, 2013
Keywords: 5-formylcytosine · DNA methylation · epigenetics ·
.
oligonucleotides · solid-phase synthesis
[1] J. A. Law, S. E. Jacobsen, Nat. Rev. Genet. 2010, 11, 204 – 220.
[2] S. Kriaucionis, N. Heintz, Science 2009, 324, 929 – 930.
[3] M. Tahiliani, K. P. Koh, Y. Shen, W. A. Pastor, H. Bandukwala,
Y. Brudno, S. Agarwal, L. M. Iyer, D. R. Liu, L. Aravind, A.
Rao, Science 2009, 324, 930 – 935.
[4] T. Pfaffeneder, B. Hackner, M. Truss, M. Mꢁnzel, M. Mꢁller,
C. A. Deiml, C. Hagemeier, T. Carell, Angew. Chem. 2011, 123,
7146 – 7150; Angew. Chem. Int. Ed. 2011, 50, 7008 – 7012.
[5] Y. F. He, B. Z. Li, Z. Li, P. Liu, Y. Wang, Q. Tang, J. Ding, Y. Jia,
Z. Chen, L. Li, Y. Sun, X. Li, Q. Dai, C. X. Song, K. Zhang, C.
He, G. L. Xu, Science 2011, 333, 1303 – 1307.
[6] C. Loenarz, C. J. Schofield, Chem. Biol. 2009, 16, 580 – 583.
[7] S. Ito, L. Shen, Q. Dai, S. C. Wu, L. B. Collins, J. A. Swenberg, C.
He, Y. Zhang, Science 2011, 333, 1300 – 1303.
[8] C. X. Song, C. He, Trends Biochem. Sci. 2013, 38, 480 – 484.
[9] E. M. Gustilo, F. A. P. Vendeix, R. Kaiser, M. O. Delaney, W. D.
Graham, V. A. Moye, W. A. Cantara, P. F. Agris, A. Deiters,
Nucleic Acids Res. 2008, 36, 6548 – 6557.
In summary, the new fdC building block 8 enabled the
synthesis of fdC-containing DNA strands in unprecedented
yield and quality, which will facilitate biological studies of
Angew. Chem. Int. Ed. 2014, 53, 315 –318
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
317

.
Angewandte
Communications
[10] B. Steigenberger, S. Schiesser, B. Hackner, C. Brandmayr, S. K.
Laube, J. Steinbacher, T. Pfaffeneder, T. Carell, Org. Lett. 2013,
15, 366 – 369.
[11] S. Tardy-Planechaud, J. Fujimoto, S. S. Lin, L. C. Sowers, Nucleic
Acids Res. 1997, 25, 553 – 559.
[12] M. Mꢁnzel, U. Lischke, D. Stathis, T. Pfaffeneder, F. A. Gnerlich,
C. A. Deiml, S. C. Koch, K. Karaghiosoff, T. Carell, Chem. Eur. J.
2011, 17, 13782 – 13788.
[13] Q. Dai, C. He, Org. Lett. 2011, 13, 3446 – 3449.
[14] N. Karino, Y. Ueno, A. Matsuda, Nucleic Acids Res. 2001, 29,
2456 – 2463.
[16] S. Schiesser, T. Pfaffeneder, K. Sadeghian, B. Hackner, B.
Steigenberger, A. S. Schrçder, J. Steinbacher, G. Kashiwazaki,
G. Hçfner, K. T. Wanner, C. Ochsenfeld, T. Carell, J. Am. Chem.
Soc. 2013, 135, 14593 – 14599.
[17] M. Mꢁnzel, D. Globisch, C. Trindler, T. Carell, Org. Lett. 2010,
12, 5671 – 5673.
[18] W. Bannwarth, A. Trzeciak, Helv. Chim. Acta 1987, 70, 175 – 186.
[19] S. L. Beaucage, M. H. Caruthers, Tetrahedron Lett. 1981, 22,
1859 – 1862.
[20] N. D. Sinha, J. Biernat, J. McManus, H. Kçster, Nucleic Acids
Res. 1984, 12, 4539 – 4557.
[15] T. M. A. Shaikh, L. Emmanuvel, A. Sudalai, J. Org. Chem. 2006,
71, 5043 – 5046.
318
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 315 –318