5-Formylcytosine Could Be a Semipermanent Base in Specific Genome Sites

Article abstract of DOI:10.1002/anie.201605994

5-Formyl-2′-deoxycytosine (fdC) is a recently discovered epigenetic base in the genome of stem cells, with yet unknown functions. Sequencing data show that the base is enriched in CpG islands of promoters and hence likely involved in the regulation of tra

Full text of DOI:10.1002/anie.201605994

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201605994
German Edition: DOI: 10.1002/ange.201605994
Epigenetics
Hot Paper
5-Formylcytosine Could Be a Semipermanent Base in Specific Genome
Sites
Meng Su, Angie Kirchner, Samuele Stazzoni, Markus Mꢀller, Mirko Wagner, Arne Schrçder, and
Thomas Carell*
Abstract: 5-Formyl-2’-deoxycytosine (fdC) is a recently dis-
covered epigenetic base in the genome of stem cells, with yet
unknown functions. Sequencing data show that the base is
enriched in CpG islands of promoters and hence likely
involved in the regulation of transcription during cellular
differentiation. fdC is known to be recognized and excised by
the enzyme thymine-DNA-glycosylase (Tdg). As such, fdC is
believed to function as an intermediate during active deme-
thylation. In order to understand the function of the new
epigenetic base fdC, it is important to analyze its formation and
removal at defined genomic sites. Here, we report a new
method that combines sequence-specific chemical derivatiza-
tion of fdC with droplet digital PCR that enables such analysis.
We show initial data, indicating that the repair protein Tdg
removes only 50% of the fdCs at a given genomic site, arguing
that fdC is a semipermanent base.
Figure 1. a) Structures of cytosine epigenetic modifications; b) sche-
matic representation of the fdC detection strategy, and building blocks
used including the click-chemistry-based assembly of the rU* probe
molecule.
D
NA contains besides the sequence information a second,
epigenetic information level, which encodes how actively the
controlled gene is transcribed.[1] Today, next to the four
canonical bases, four additional epigenetic bases are known.[2]
These are 5-methylcytosine (5mC), 5-hydroxymethylcytosine
(5hmC),[3] 5-formylcytosine (5fC),[4] and 5-carboxycytosine
(5caC)[5] (Figure 1a). Over the last years, sensitive mass
spectrometry based methods have helped to reveal the global
levels of these epigenetic bases in stem cells[4,6] and tissues
including the brain.[7] In order to learn about the levels and
the distribution of the epigenetic bases at specific sites in the
genome, different sequencing methods have been devel-
oped[8] in which selective chemical derivatization of the bases
is performed,[9] sometimes in combination with bisulfite
sequencing.[9c,10] Although these methods provide informa-
tion about the distribution of the bases at a given time point, it
is a hallmark of epigenetic information that it changes
dynamically. To gain deeper insight into the dynamics of the
epigenetic information layer at a single position in the
genome, it is therefore essential to develop methods that
can be used to follow the changes of, for example, fdC at
a specific location in the genome over time.[11] A perfect
method will ultimately allow parallel monitoring of fdC
dynamics at different genomic sites.
The central question addressed in this manuscript is: Are
the measured global data of the past averages from different
processes at different positions in the genome, or do they
reflect what is happening at an individual site in the genome?
To answer this question, we developed a sequence-specific
chemical derivatization method that we then used in combi-
nation with droplet digital PCR to monitor the epigenetic
base fdC at different loci directly in the genome of stem cells.
For the sequence-specific localization of fdC in the
genome, we utilize a small probe oligonucleotide (Figure 1b,
Table S1), which contains a hydroxylamine tether that is able
to form a covalent linkage with fdC so that the probe strand is
subsequently tightly bound to the target.[4] We examined
systematically different linker lengths, linker attachment
points, and distances. Best results were obtained when we
incorporated the 2’-O-propargyl uridine using its phosphor-
amidite 1 into the probe oligonucleotide and attached the
azido-C4-hydroxylamine 2 using the CuI-catalyzed version of
the Huisgen reaction (click reaction).[12] We protected the
hydroxylamine unit for the click reaction with a monome-
thoxytrityl group (MMTr), which was cleaved afterwards with
acetic acid at 258C. This brief exposure of the probe
oligonucleotide to acidic conditions did not cause significant
depurination. After solid-phase synthesis, click modification
of the oligonucleotide, and a final purification step (Fig-
ure S1), we obtained oligonucleotides with different sequen-
[*] M. Sc. M. Su, M. Sc. A. Kirchner, M. Sc. S. Stazzoni, Dr. M. Mꢀller,
Dr. M. Wagner, M. Sc. A. Schrçder, Prof. Dr. T. Carell
Center for Integrated Protein Science at the
Department of Chemistry, Ludwig-Maximilians-Universitꢁt Mꢀnchen
Butenandtstrasse 5–13, 81377 Mꢀnchen (Germany)
E-mail: thomas.carell@lmu.de
Homepage: http://www.carellgroup.de
Supporting information and the ORCID identification number(s) for
the author(s) of this article can be found under http://dx.doi.org/10.
1002/anie.201605994.
Angew. Chem. Int. Ed. 2016, 55, 1 – 5
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

Angewandte
Communications
Chemie
ces and lengths containing an rU-hydroxylamine base (rU*)
at different positions for reaction with the fdC base on the
target strand. For the following experiments, we prepared 13-
mer long oligonucleotides.
To investigate at which position the linker in the probe
strand would react best with fdC in the target strand, we
varied the position of rU* relative to fdC and explored
different reaction conditions. Excellent results were finally
obtained when probe strand P1, containing rU* exactly four
base pairs in the 5’-direction relative to fdC, was hybridized to
the fdC target strand T1 in the presence of catalytic amounts
of 4-methoxyaniline (Figure 2a). With this catalyst, the
crosslinking reaction is complete after 24 h with yields
exceeding 95%. Without the catalyst, only about 50% yield
could be obtained (Figure 2b).
Figure 3. Denaturing PAGE gel showing duplex formation of T3–8 and
P1 at 258C after 24 h.
whether the reaction of rU* is possible with other pyrimidines
(Figure 3). To our delight, hybridization of the rU*-contain-
ing probe strand with target strands containing dT, dC, mdC,
hmdC, and cadC (T3–8) gave no reaction. Reaction is,
however, observed with abasic sites. This is important because
fdC and cadC are substrates for base excision repair and
hence could in principle be precursors for abasic sites.[13] In
this sense, rU* always reports the presence of fdC and also
potentially of fdC- and cadC-derived abasic sites.
We finally turned our procedure for sequence-specific fdC
detection into a method for detecting single fdC bases at
a defined position in whole genomes. To this end, we coupled
the chemistry to droplet digital PCR[14]-based amplification
and readout.
Figure 2. Denaturing PAGE gel showing the duplex formation between
T1 and P1 at 258C: a) with the catalyst 4-methoxyaniline; b) without
a catalyst; c) with the catalyst 1,4-diaminobenzene. d) Quantification
of the DNA duplex formation during the reaction. Black: catalyst 4-
methoxyaniline, blue: no catalyst, red: catalyst 1,4-diaminobenzene.
Error bars represent the standard error of the mean calculated from
three replicates. Conditions: 2 mm oligonucleotides, 100 mm NaCl,
10 mm NaOAc buffer pH 6.0, 10 mm 4-methoxyaniline. M=single-
strand marker. The time point 0 is after re-annealing.
In the first step genomic target DNA (Tg) was isolated
from mouse embryonic stem cells (mESCs) at different time
points during priming from naive cells. We also isolated
genomic DNA from mESCs with a knockout of the Tdg repair
enzyme (TdgÀ/À) to block excision and repair of fdC and cadC.
We finally also isolated genomic DNA from mESCs lacking
any of the three methyltransferases (Dnmt1, 3a, and 3b).
These stem cells lack mdC and are hence unable to produce
the oxidized xdC (x = hm, f, and ca) epigenetic bases. This
genomic DNA served in our studies consequently as a neg-
ative control. For analysis, we selected two different fdC sites
that were reported to have high fdC contents.[10c] We focused
initially on the 30020539th site of chromosome 16 Mus
musculus (MM9) located on the exon 3 of 632428C04Rik. It
was found to contain 23% of fdC based on redBS-sequencing.
The second site we studied was the 8846677th site of
chromosome 15, which is located in noncoding DNA. This
site was reported to contain 32% of fdC.
For the first site, we reacted a 25-mer probe (P2, SI)
containing the rU* base with Tg using 1,4-diaminobenzene as
the catalyst. In the absence of fdC, a covalent bond between
P2 and Tg cannot form. To remove the excess of probe, we
loaded the Tg:P2 complex onto an NEB Monarch DNA
cleanup column and rinsed the column with wash buffer to
elute oligonucleotides shorter than the 50-mer, which is the
unbound P2. After this washing, we eluted the Tg:P2 complex
with TE buffer. UV/Vis analysis of the eluted material
showed a typical gDNA spectrum. We next added a 70-mer 5’-
phosphorylated reporter strand (R1, SI) which hybridizes
with an 18-nt stretch directly adjacent to the probe strand and
ligated both probe and the reporter at 608C upon addition of
Ampligase to form R1–P2 as depicted schematically in
Figure 4a.
In order to increase the rate of the reaction, we tested
other catalysts. We observed the best results when we used
1,4-diaminobenzene as a catalyst, in which case the cross-
linking reaction between T1 and P1 is completed already after
3 h (Figure 2c). Duplex formation (T1:P1) was analyzed using
denaturing PAGE and quantified by fluorescence (Fig-
ure 2d).
When fdC is located one base pair further away from rU*
without changes in the probe strand, we observe slower
reaction (Figure S2). These results show that rU* placed four
or five bases away from fdC in the 5’-direction to fdC allows
the tether to reach the formyl group of fdC via the major
groove of the duplex (Figure S3).
MALDI-TOF data confirmed that the crosslinks form as
expected (Figure S4). For the crosslinked duplex T1:P1, we
obtained the correct molecular weight with m/zfound = 8081.9
(m/zcalc = 8084.7). As expected, the oxime formation reaction
between T1 and P1 leads to a higher melting temperature of
the hybridized and crosslinked duplex (Figure S5). Typically,
we observed that the un-crosslinked 13-mer duplex melts at
around 448C. The duplex after crosslink formation shows
a melting temperature of above 808C.
Because pyrimidine bases are able to react with nucleo-
philes also at the C6 position in a Michael-type reaction,
which is the basis for bisulfite sequencing, we next tested
2
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 5
These are not the final page numbers!

Angewandte
Communications
Chemie
performed at 608C for 10 h. The red signals are finally
generated from droplets that contain both PCR products. For
the calculation, please see the Supporting Information.
Using the method, we first studied mES cells lacking the
Tdg enzyme (TdgÀ/À). A rather high level of 28.5% fdC was
measured at the first locus (Figure 4c) in agreement with the
results from redBS-seq.[10c] When we performed the study,
however, with mESCs having an active Tdg repair enzyme
(Tdg+/À) we measured that the fdC level drops at this
particular position to 15.7% (Figure 4c). This is very impor-
tant because it shows that Tdg removes only half of the fdCs
at a given site and also unusual since repair glycosylases are
known to find basically all possible substrates. The result
underpins the high dynamics of fdC formation and repair at
a given site. When we studied the fdC content at this location
in mESCs lacking any methyltransferase (Dnmt TKO) the
fdC level drops as expected to a little more than 5%, showing
that the reported levels of fdC in the Tdg+/À cells are real and
not an artifact. In order to elucidate whether single-site fdC
levels (Figure 4c) follow global genomic fdC levels, we
quantified the total levels of fdC in these cells (Figure 4d).
These global data are in good agreement with the data
obtained from single-site fdC quantifications. Thus, our new
data make a scenario where fdC is fully removed at one site
and shielded from repair at another place unlikely. Instead,
fdC is even at a given position only partially removed in a cell
population. Alternatively, it may be that Tdg removes fdC
differently on the two chromosomes, which however needs
further investigation.
In order to verify the data, we repeated the Tdg study at
a second genomic site (8846677th nucleoside of chromosome
15). For this site, we designed a new probe strand P3 and
a new reporter strand R2 and performed again ddPCR with
two sets of primers (Figure S5). When we compare the data
obtained from TdgÀ/À cells with the data from Tdg+/À cells,
again only a 50% reduction of the fdC level is shown at this
position, in full agreement with the data above obtained from
the first position (Figure 4e).
We finally performed a kinetic study in which we
monitored the fdC development at the first position during
priming of stem cells (Figure 4 f). We see that the fdC levels
rise at the given position with a strong increase in the early
phase of priming, followed by a small decline phase and
finally stable values (Figure 4 f) again in agreement with the
global data that we measured using our reported method
(Figure 4g).
Figure 4. The fdC sequencing method: a) schematic representation of
the method, black line: gDNA; red segments: fdC probe; blue seg-
ments: reporter strands; arrows: PCR primer pairs. b) Typical 2D plot
of droplet fluorescence. c) Cluster ratios for position 1 in TdgÀ/À, Tdg+/
À, and Dnmt TKO mES cells. d) Global fdC quantification in TdgÀ/À
,
Tdg+/À, and Dnmt TKO mES cells using our LC-MS method. e) Cluster
ratios for position 2 in TdgÀ/À and Tdg+/À mES cells. f) Cluster ratios
for position 1 in wild-type mES cells at different days after priming.
g) Global quantification data for the wild-type mES cells measured by
LC-MS.
We next added two sets of primers to the assay (blue and
green arrows, Figure 4a) to amplify the ligation product
relative to the target duplex. Importantly, the blue primers
recognize only the hybrid R1–P2 probe generated in the
ligation step, while the green primers indicate the presence of
gDNA. The amplification was monitored with two different
TaqMan probes which showed fluorescence at 520 and
556 nm. This relative detection is needed to normalize on
the amount of input gDNA. Because conventional real-time
PCR is known to become inaccurate when copy number
differences less than the 10-fold need to be resolved, we used
droplet digital PCR. In this method, small droplets are
generated with one droplet containing a maximum of one of
the fully assembled analysis constructs shown in Figure 4a.
The PCR reaction takes place in the droplets, producing
a specific signal. Subsequent fluorescence-based counting of
each individual droplet yields numbers from which one can
accurately calculate the amount of fdC, even if the fdC values
are very low. A representative plot of the data is given in
Figure 4b. Empty drops give no PCR signal (black dots in
Figure 4b). Drops containing only Tg give only the PCR
signal from the green primers (green dots in Figure 4b). Blue
dots are obtained due to the dissociation of the ligated
product R1–P2 from Tg in the ligation process which is
The fact that our method provides the same trends as seen
in the global data at a single genomic site makes us confident
that our method is robust and reliable in reporting what
happens at an individual site. Because single-site and global
data correlate, we have now first evidence that the reported
global trends are reflecting what happens at each individual
fdC site, rather than evening out largely different dynamics at
separate sites. Another interesting result of this study is that
the repair enzyme Tdg removes only half of the fdC bases at
a given genomic site in an mESC population, which argues
that fdC is a semipermanent base at a given position in the
genome.
Angew. Chem. Int. Ed. 2016, 55, 1 – 5
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

Angewandte
Communications
Chemie
Li, Y. Sun, X. Li, Q. Dai, C.-X. Song, K. Zhang, C. He, G.-L. Xu,
Science 2011, 333, 1303 – 1307.
[6] S. Schiesser, B. Hackner, T. Pfaffeneder, M. Mꢂller, C. Hage-
meier, M. Truss, T. Carell, Angew. Chem. Int. Ed. 2012, 51, 6516 –
6520; Angew. Chem. 2012, 124, 6622 – 6626.
[7] a) M. Mꢂnzel, D. Globisch, T. Brꢂckl, M. Wagner, V. Welzmiller,
S. Michalakis, M. Mꢂller, M. Biel, T. Carell, Angew. Chem. Int.
Ed. 2010, 49, 5375 – 5377; Angew. Chem. 2010, 122, 5503 – 5505;
b) D. Globisch, M. Mꢂnzel, M. Mꢂller, S. Michalakis, M.
Wagner, S. Koch, T. Brꢂckl, M. Biel, T. Carell, PLoS ONE
2010, 5, e15367; c) M. Wagner, J. Steinbacher, T. F. J. Kraus, S.
Michalakis, B. Hackner, T. Pfaffeneder, A. Perera, M. Mꢂller, A.
Giese, H. A. Kretzschmar, T. Carell, Angew. Chem. Int. Ed.
2015, 54, 12511 – 12514; Angew. Chem. 2015, 127, 12691 – 12695.
[8] a) N. Plongthongkum, D. H. Diep, K. Zhang, Nat. Rev. Genet.
2014, 15, 647 – 661; b) M. J. Booth, E.-A. Raiber, S. Balasubra-
manian, Chem. Rev. 2015, 115, 2240 – 2254.
Experimental Section
Probe crosslinking: gDNA solution (1.2 mg), fdC probe (1 mm, 2 mL),
NaH2PO4–Na2HPO4 buffer (200 mm, pH 6.0, 2 mL), NaCl aq. (1.5m,
2 mL), and doubly distilled (dd) H2O were mixed to a final volume of
18 mL. The mixture was heated to 958C for 3 min and then cooled
down rapidly to 258C. 1,4-Diaminobenzene aq. (10 mm, 2 mL) was
added and the reaction vial was shaken for 6 h at 258C. The mixture
was neutralized with Na2HPO4 aq. (200 mm, 40 mL) before purifica-
tion with the NEB Monarch PCR DNA Cleanup Kit.
Ligation: The above-described gDNA solution (300 ng), reporter
strand (20 nm, 1 mL), Ampligase reaction buffer (10 ꢀ , 2 mL),
Ampligase from Epicentre (10 U), and ddH2O were mixed to
a final volume of 20 mL. The mixture was heated to 958C for 3 min,
and then 948C for 1 min, 608C for 1 h, and back to 948C for 10 cycles.
Then, the reaction mixture was diluted with Tris-HCl buffer (200 mm,
pH 7.6, 50 mL) before purification using the NEB Monarch PCR
DNA Cleanup Kit.
[9] a) W. A. Pastor, U. J. Pape, Y. Huang, H. R. Henderson, R.
Lister, M. Ko, E. M. McLoughlin, Y. Brudno, S. Mahapatra, P.
Kapranov, M. Tahiliani, G. Q. Daley, X. S. Liu, J. R. Ecker, P. M.
Milos, S. Agarwal, A. Rao, Nature 2011, 473, 394 – 397; b) E.-A.
Raiber, D. Beraldi, G. Ficz, H. Burgess, M. Branco, P. Murat, D.
Oxley, M. Booth, W. Reik, S. Balasubramanian, Genome Biol.
2012, 13, R69; c) C.-X. Song, K. E. Szulwach, Q. Dai, Y. Fu, S.-Q.
Mao, L. Lin, C. Street, Y. Li, M. Poidevin, H. Wu, J. Gao, P. Liu,
L. Li, G.-L. Xu, P. Jin, C. He, Cell 2013, 153, 678 – 691; d) B. Xia,
D. Han, X. Lu, Z. Sun, A. Zhou, Q. Yin, H. Zeng, M. Liu, X.
Jiang, W. Xie, C. He, C. Yi, Nat. Methods 2015, 12, 1047 – 1050.
[10] a) M. Yu, G. C. Hon, K. E. Szulwach, C.-X. Song, L. Zhang, A.
Kim, X. Li, Q. Dai, Y. Shen, B. Park, J.-H. Min, P. Jin, B. Ren, C.
He, Cell 2012, 149, 1368 – 1380; b) M. J. Booth, M. R. Branco, G.
Ficz, D. Oxley, F. Krueger, W. Reik, S. Balasubramanian, Science
2012, 336, 934 – 937; c) M. J. Booth, G. Marsico, M. Bachman, D.
Beraldi, S. Balasubramanian, Nat. Chem. 2014, 6, 435 – 440; d) X.
Lu, C.-X. Song, K. Szulwach, Z. Wang, P. Weidenbacher, P. Jin,
C. He, J. Am. Chem. Soc. 2013, 135, 9315 – 9317.
Droplet digital PCR (ddPCR) was conducted on a Bio-Rad
QX100 ddPCR System. For one reaction, gDNA (6 ng), four primers
(18 mm each, 1 mL), two TaqMan probes (5 mm each, 1 mL), digital
PCR Supermix for Probes (no dUTP, 2 ꢀ , 10 mL), and ddH2O were
mixed to a final volume of 20 mL. PCR cycle: heating at 958C for
10 min, 948C for 30 s, and 648C for 1 min for 35 cycles, then 988C for
10 min and cooling to 128C, with a temperature ramp of 28CsÀ1. For
a detailed description please see the Supporting Information.
Acknowledgments
We thank K. Hufnagel for preparing the phosphoramidites of
the epigenetic bases. This project received funding from
Deutsche Forschungsgemeinschaft SFB1032 and the Excel-
lence Cluster CiPSM (EXC114). Further support was obtained
from the European Unionꢁs Horizon 2020 research and
innovation program under grant agreement no. 642023 (ITN
clickgene).
[11] a) A. Nomura, K. Sugizaki, H. Yanagisawa, A. Okamoto, Chem.
Commun. 2011, 47, 8277 – 8279; b) J. Duprey, G. A. Bullen, Z.-Y.
Zhao, D. M. Bassani, A. F. A. Peacock, J. Wilkie, J. H. R. Tucker,
ACS Chem. Biol. 2016, 11, 717 – 721.
Keywords: 5-formylcytosine · click chemistry ·
[12] a) P. M. E. Gramlich, S. Warncke, J. Gierlich, T. Carell, Angew.
Chem. Int. Ed. 2008, 47, 3442 – 3444; Angew. Chem. 2008, 120,
3491 – 3493; b) J. Willibald, J. Harder, K. Sparrer, K.-K. Con-
zelmann, T. Carell, J. Am. Chem. Soc. 2012, 134, 12330 – 12333.
[13] A. Maiti, A. C. Drohat, J. Biol. Chem. 2011, 286, 35334 – 35338.
[14] B. J. Hindson, K. D. Ness, D. A. Masquelier, P. Belgrader, N. J.
Heredia, A. J. Makarewicz, I. J. Bright, M. Y. Lucero, A. L.
Hiddessen, T. C. Legler, T. K. Kitano, M. R. Hodel, J. F.
Petersen, P. W. Wyatt, E. R. Steenblock, P. H. Shah, L. J.
Bousse, C. B. Troup, J. C. Mellen, D. K. Wittmann, N. G. Erndt,
T. H. Cauley, R. T. Koehler, A. P. So, S. Dube, K. A. Rose, L.
Montesclaros, S. Wang, D. P. Stumbo, S. P. Hodges, S. Romine,
F. P. Milanovich, H. E. White, J. F. Regan, G. A. Karlin-Neu-
mann, C. M. Hindson, S. Saxonov, B. W. Colston, Anal. Chem.
2011, 83, 8604 – 8610.
droplet digital PCR · epigenetic bases · genomic DNA
[1] P. A. Jones, Nat. Rev. Genet. 2012, 13, 484 – 492.
[2] T. Carell, C. Brandmayr, A. Hienzsch, M. Mꢂller, D. Pearson, V.
Reiter, I. Thoma, P. Thumbs, M. Wagner, Angew. Chem. Int. Ed.
2012, 51, 7110 – 7131; Angew. Chem. 2012, 124, 7220 – 7242.
[3] a) S. Kriaucionis, N. Heintz, Science 2009, 324, 929 – 930; b) 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. Mꢂnzel, M. Mꢂller, C.
Deiml, C. Hagemeier, T. Carell, Angew. Chem. Int. Ed. 2011, 50,
7008 – 7012; Angew. Chem. 2011, 123, 7146 – 7150.
[15] M. Wendeler, L. Grinberg, X. Wang, P. E. Dawson, M. Baca,
Bioconjugate Chem. 2014, 25, 93 – 101.
[5] a) S. Ito, L. Shen, Q. Dai, S. C. Wu, L. B. Collins, J. A. Swenberg,
C. He, Y. Zhang, Science 2011, 333, 1300 – 1303; b) Y.-F. He, B.-
Z. Li, Z. Li, P. Liu, Y. Wang, Q. Tang, J. Ding, Y. Jia, Z. Chen, L.
Received: June 20, 2016
Published online: && &&, &&&&
4
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 5
These are not the final page numbers!

Angewandte
Communications
Chemie
Communications
Epigenetics
Big hug for fdC: A method for the site-
specific quantification of the epigenetic
base 5-formylcytosine (fdC) was devel-
oped. Applied on the genome of an mESC
population, the assay showed that the
repair enzyme Tdg removes only half of
the fdC bases at a given genomic site.
M. Su, A. Kirchner, S. Stazzoni, M. Mꢁller,
M. Wagner, A. Schrçder,
T. Carell*
&&&& — &&&&
5-Formylcytosine Could Be
a Semipermanent Base in Specific
Genome Sites
Angew. Chem. Int. Ed. 2016, 55, 1 – 5
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!