The existence of 5-hydroxymethylcytosine and 5-formylcytosine in both DNA and RNA in mammals

Article abstract of DOI:10.1039/c5cc07354e

We developed a novel strategy by oxidation-derivatization combined mass spectrometry analysis for the determination of 5-hydroxymethylcytosine and 5-formylcytosine in both DNA and RNA. We reported the presence of 5-formylcytosine in RNA of mammals and found that ascorbic acid and hydroquinone can increase the oxidation of 5-methylcytosine to 5-hydroxymethylcytosine in DNA and RNA.

Full text of DOI:10.1039/c5cc07354e

ChemComm
View Article Online
View Journal
COMMUNICATION
The existence of 5-hydroxymethylcytosine and
-formylcytosine in both DNA and RNA in
5
Cite this:DOI: 10.1039/c5cc07354e
mammals†
Received 2nd September 2015,
Accepted 3rd November 2015
Hao-Ying Zhang,‡ Jun Xiong,‡ Bao-Ling Qi, Yu-Qi Feng* and Bi-Feng Yuan*
DOI: 10.1039/c5cc07354e
www.rsc.org/chemcomm
We developed a novel strategy by oxidation–derivatization combined time in RNA in mammalian cells and found that TET proteins
mass spectrometry analysis for the determination of 5-hydroxy- also possess the activity of catalyzing the formation of 5-hmC
methylcytosine and 5-formylcytosine in both DNA and RNA. We from 5-mC in RNA. However, the biological significance of
reported the presence of 5-formylcytosine in RNA of mammals and 5-hmC in RNA is less understood.
found that ascorbic acid and hydroquinone can increase the oxidation
of 5-methylcytosine to 5-hydroxymethylcytosine in DNA and RNA.
The 5-hmC content in genomic DNA of mammalian cells can
1
5
be as low as 0.0004% of cytosines, and 5-foC occurs at a
6
16–18
frequency of 1 to 20 per 10 cytosines.
In RNA, the content
6
14
DNA cytosine methylation (5-methylcytosine, 5-mC) is one of the of 5-hmC can be approximately 1 modification per 10 cytosines.
best-characterized epigenetic modifications that plays crucial Previous bisulfite sequencing data showed that the widespread
1
roles in a variety of cellular processes. Recent studies have presence of 5-mC in RNA and 5-mC may dynamically modulate
1
3
demonstrated that active DNA demethylation in mammals RNA–protein interactions. A recent study suggested that TET
can be achieved through a consecutive oxidation of 5-mC by enzymes can catalyze the formation of 5-hmC from 5-mC
1
4
ten–eleven translocation (TET) dioxygenases with the generation in RNA. Therefore the presence of 5-hmC in RNA and the
of three intermediates, 5-hydroxymethylcytosine (5-hmC), 5-formyl- involvement of TET family enzymes in generating this modifi-
2
–5
cytosine (5-foC), and 5-carboxylcytosine (5-caC).
In addition cation indicate that 5-hmC in RNA may also participate in the
to being an intermediate in 5-mC removal, 5-hmC itself is now epigenetic regulation of gene expression through modulating
considered to be an important epigenetic marker that has RNA–protein interactions. However, quantifications of these
6
–10
functions in various physiological processes.
5-foC is the cytosine modifications are challenging due to their low in vivo
contents as well as the interference from the highly abundant
further oxidation product of 5-hmC by TET dioxygenases.
Given the rich layers of epigenetic regulation that result from normal nucleosides. In this respect, sensitive and accurate
modifications of DNA, reversible RNA modification has been measurements of these epigenetic modifications are essential
proposed to represent another realm for biological regulation to elucidate their functional roles.
1
1
in the form of ‘‘RNA epigenetics’’. Naturally occurring RNA
In the past decade, considerable advances have been made in
molecules contain various modified nucleosides derived from the development of methods for the quantification of these cyto-
four standard nucleosides. To date, more than 100 modified sine modifications, including radioactive labelling-based thin layer
1
2
3
19
nucleosides have been identified in all three kingdoms of life,
among which, 5-mC is one of the most important modifications cence labelling-based detection,
chromatography detection, immunohistochemistry, fluores-
20,21
and liquid chromatography-
Due to its inherent selectivity and
sensitivity, LC-MS has been frequently used in the analysis of low
22,23
in the RNA molecules with potential functions in the control mass spectrometry (LC-MS).
1
3
and regulation of gene transcription and protein translation.
1
4
16,24
Recently, Fu et al. reported the existence of 5-hmC for the first abundance modified nucleosides,
whereas, in these methods,
offline HPLC purification was typically required prior to MS
analysis to remove the interference from the bulky normal
Key Laboratory of Analytical Chemistry for Biology and Medicine (Ministry of Education), nucleosides, which is laborious and time-consuming. In addi-
Department of Chemistry, Wuhan University, Wuhan 430072, P. R. China.
tion, the ionization efficiencies in electrospray ionization (ESI)
of these cytosine modifications are usually low. Thus, it is
possible to increase the detection sensitivities during LC-ESI-
MS/MS analysis by introducing an easily ionizable moiety into
these cytosine modifications.
E-mail: bfyuan@whu.edu.cn, yqfeng@whu.edu.cn; Fax: +86-27-68755595;
Tel: +86-27-68755595
†
Electronic supplementary information (ESI) available. See DOI: 10.1039/
c5cc07354e
These authors contributed equally to this work.
‡
This journal is ©The Royal Society of Chemistry 2015
Chem. Commun.

View Article Online
Communication
ChemComm
Fig. 1 Oxidation–derivatization reactions of (A) 5-hmdC and (B) 5-hmrC.
In this study, we developed a strategy using oxidation–
derivatization combined mass spectrometry (ODMS) analysis
0
for the highly sensitive detection of 5-hydroxymethyl-2 -deoxy-
cytidine (5-hmdC), 5-hydroxymethylcytidine (5-hmrC), 5-formyl-
0
2
-deoxycytidine (5-fodC) and 5-formylcytidine (5-forC). In this
was utilized to oxide 5-hmdC and 5-hmrC to
-fodC and 5-forC, respectively. The oxidization products were
strategy, MnO
5
2
then further labeled by dansylhydrazine (DNSH) with the
aldehyde group in 5-fodC and 5-forC (Fig. 1). DNSH harbors
the hydrazide moiety that can readily react with aldehyde to give
hydrazone derivatives with an easily chargeable moiety (tertiary
ammonium), which therefore can increase the ionization effi-
ciency of target analytes during LC-ESI-MS/MS analysis. In
Fig. 2 Determination of endogenous 5-hmdC and 5-hmrC in DNA and
RNA. Product ion spectra of oxidation–derivatization product of 5-hmdC
standard (A), 5-hmrC standard (B), 5-hmdC detected in HeLa cells (C) and
5-hmrC detected in HeLa cells (D) by high resolution mass spectrometry.
The numbers marked in red are the theoretical molecular weights. (E) MRM
addition, the hydrophobic phenyl group introduced by DNSH chromatograms of 5-hmdC standard (i) and 5-hmdC detected in HeLa
cells (ii) by ODMS strategy, and endogenous 5-fodC detected in HeLa cells
by DNSH derivatization (iii). (F) MRM chromatograms of 5-hmrC standard
derivatization can dramatically increase the retention of reversed-
phase LC, which, therefore can finally contribute to the increased
detection sensitivities.
(
i) and 5-hmrC detected in HeLa cells (ii) by ODMS strategy, and endo-
genous 5-forC detected in HeLa cells by DNSH derivatization (iii).
To evaluate the detection sensitivity of the ODMS method,
we performed the experiment under optimized MnO oxidation
2
conditions (Fig. S1, ESI†) and derivatization conditions (Fig. S2, spectrometry to analyze the oxidation–derivatization products
ESI†). The MRM chromatogram shows that the retentions of of 5-hmdC and 5-hmrC. The results showed that the product
5-hmdC and 5-hmrC were relatively weak (B3.2 min) and they ion spectra of the oxidation–derivatization products of 5-hmdC
coeluted on the C18 reversed-phase column even under optimized and 5-hmrC standards (Fig. 2A and B) were identical to that of
separation conditions (Fig. S3, ESI†). However, the retentions of oxidation–derivatization products of 5-hmdC and 5-hmrC from
the oxidation–derivatization products of 5-hmdC and 5-hmrC human HeLa cells (Fig. 2C and D), supporting the existence of
increased dramatically, and the separation resolution also 5-hmdC and 5-hmrC in human HeLa cells. And we further
improved (36.4 min for oxidation–derivatization products of explored the lowest amounts of DNA and RNA for the analysis
5
5
-hmrC and 38.2 min for oxidation–derivatization products of of epigenetic modifications. The results showed that under
-hmdC) (Fig. S3, ESI†). optimized conditions, 200 ng of DNA and RNA were sufficient
The limits of detection (LODs) of the oxidation–derivatization for the analysis of endogenous 5-fodC and 5-forC (Fig. S4A and
products of 5-hmdC and 5-hmrC were 0.04 and 0.03 fmol, S4B, ESI†). And 20 ng and 100 ng of DNA and RNA were
respectively (Table S1, ESI†), which are, to the best of our sufficient for the analysis of 5-hmdC and 5-hmrC, respectively
knowledge, the lowest LODs of 5-hmdC and 5-hmrC ever (Fig. S4C and S4D, ESI†).
achieved (the summary of the detection sensitivities of 5-hmdC
Fig. 2E and F show the typical MRM chromatograms of the
and 5-hmrC by different methods is shown in Table S1, ESI†). oxidation–derivatization products of 5-hmdC and 5-hmrC,
Compared with the native 5-hmdC and 5-hmrC, the detection respectively. We first quantified the endogenous 5-fodC and
sensitivities increased by 363-fold and 380-fold for 5-hmdC and 5-forC with DNSH derivatization. The results showed that the
6
5
-hmrC, respectively (Table S2, ESI†).
measured contents of 5-fodC were 4.5 ꢀ 0.6/10 dG in HeLa
6
Using the established ODMS strategy, we quantified 5-hmdC, cells and 19.1 ꢀ 3.3/10 dG in 293T cells (Table S3, ESI†), which
1
6
5
-hmrC, 5-fodC and 5-forC contents in DNA and RNA from are comparable to previously reported results. In addition, we
HeLa and 293T cells. We first employed high-resolution mass quantified 5-hmdC contents in HeLa cells and 293T cells using
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2015

View Article Online
ChemComm
Communication
the developed ODMS strategy. In this respect, the endogenous
5-hmdC contents can be obtained by subtracting the endogenous
contents of 5-fodC from the total measured contents of 5-fodC.
The results showed that the calculated contents of 5-hmdC in
6
HeLa and 293T cells were 68.4 ꢀ 10.1/10 dG and 109.1 ꢀ
6
1
4.4/10 dG, respectively (Table S3, ESI†), which are consistent
1
5
with previously measured contents.
Similarly, the measured contents of 5-hmrC in RNA were
6
6
3
(
.1 ꢀ 0.4/10 rG in HeLa cells and 6.9 ꢀ 0.7/10 rG in 293T cells
Table S2, ESI†), which are comparable with previously reported
1
4
results. And the measured contents of 5-forC were 9.0 ꢀ
6
6
1
(
.2/10 rG in HeLa cells and 8.5 ꢀ 1.4/10 rG in 293T cells
Table S3, ESI†). It is worth noting that 5-forC, for the first time,
reported the presence in RNA in mammalian cells. And the
biological functions of 5-forC in RNA need further investigation.
Ascorbic acid is one of the most important nutrients for
mammals. Recently, it has been demonstrated that ascorbic
acid could enhance the catalytic activity of TET dioxygenases
2
5,26
for the oxidation of 5-mdC to 5-hmdC.
In addition, another
study revealed that hydroquinone can also promote the genera-
tion of 5-hmdC in a mechanism involving the elevated activity of
27
TET dioxygenases. Considering the similarity of the chemical
structure between 5-hmdC and 5-hmrC, we speculate that ascorbic
acid and hydroquinone may also increase the content of 5-hmrC in
Fig. 3 Contents change of (A) 5-hmdC, (B) 5-hmrC, (C) 5-fodC and
(
D) 5-forC in HeLa and 293T cells by ascorbic acid and hydroquinone
treatments. Ascorbic acid and hydroquinone concentrations, 100 mM;
RNA since the previous study demonstrated that TET dioxygenases treatment time, 24 h. Error bars represent the SEM (n = 6); *p o 0.05;
could oxidize 5-mrC to 5-hmrC in a similar way to TET **p o 0.01; ***p o 0.001.
1
4
dioxygenases converting 5-mdC to 5-hmdC. In this respect,
here we used ascorbic acid and hydroquinone treated HeLa and
293T cells and simultaneously examined the change in the p = 0.003; 2.0-fold increase for hydroquinone treatment, p = 0.012)
contents of 5-hmdC and 5-hmrC in DNA and RNA. and 293T cells (2.6-fold increase for ascorbic acid treatment,
As shown in Fig. 3A, both ascorbic acid and hydroquinone p = 0.001; 2.3-fold increase for hydroquinone treatment, p = 1.9 ꢁ
ꢂ4
treatment induced a significant increase of 5-hmdC in HeLa 10 ) (Fig. 3D). Since 5-fodC is produced from 5-hmdC by TET
cells (2.4-fold increase for ascorbic acid treatment, p = 0.002; dioxygenases, both the increased activity of TET dioxygenases and
1.7-fold increase for hydroquinone treatment, p = 0.009) and the elevated 5-hmdC contents could contribute to the increased
2
93T cells (2.7-fold increase for ascorbic acid treatment, contents of 5-fodC upon ascorbic acid and hydroquinone treat-
ꢂ4
p = 6.5 ꢁ 10 ; 1.3-fold increase for hydroquinone treatment, ments. And similar phenomena were observed for 5-forC, further
2
5,27
p = 0.017), which is consistent with previous studies.
indicating that the endogenous 5-forC may be generated through
Similarly, 5-hmrC in RNA also significantly increased in HeLa the oxidation of 5-hmrC by TET dioxygenases.
ꢂ4
cells (4.5-fold increase for ascorbic acid treatment, p = 2.7 ꢁ 10
;
The discovery of 5-forC in cellular RNA in the current study
ꢂ4
2
.3-fold increase for hydroquinone treatment, p = 4.0 ꢁ 10 ) and together with the recently identified presence of 5-hmrC in RNA
2
93T cells (3.3-fold increase for ascorbic acid treatment, p = 0.002; suggest that the function of TET dioxygenases are not limited
1.6-fold increase for hydroquinone treatment, p = 0.022) by ascorbic to the epigenetic regulation at the DNA level, but can also be
acid and hydroquinone treatment (Fig. 3B). The results suggested extended to RNA. The present work sets the stage for future
that the induced increase of 5-hmrC in RNA by ascorbic acid and studies in defining the site-specific localization of 5-forC in
hydroquinone treatments might go through the same mechanism RNA as well as the specific binding proteins, and the functions
as that of 5-hmdC, i.e., the elevated catalytic activity of TET of 5-mrC oxidation in RNA biology. To fully understand the
dioxygenases caused by ascorbic acid and hydroquinone can mechanism of these modifications on gene expression regulations,
enhance both 5-hmdC and 5-hmrC contents in DNA and RNA. it is essential to develop a high resolution sequencing method to
As for the content of 5-fodC upon ascorbic acid and hydro- map the distribution of these modifications.
quinone treatments, significant increases were observed in HeLa
In the current study, we developed a highly sensitive method
cells (3.1-fold increase for ascorbic acid treatment, p = 0.005; to simultaneously determine 5-hmC and 5-foC in both DNA and
ꢂ5
4
.1-fold increase for hydroquinone treatment, p = 3.5 ꢁ 10
)
RNA by MnO oxidation and DNSH derivatization coupled with
2
and 293T cells (2.9-fold increase for ascorbic acid treatment, mass spectrometry (ODMS) analysis. Upon MnO
p = 0.004; 2.2-fold increase for hydroquinone treatment, p = 0.007) 5-hmdC and 5-hmrC were converted to 5-fodC and 5-forC,
Fig. 3C). Similarly, significant increases were also observed for respectively, which were then further labeled by DNSH. DNSH
-forC in HeLa cells (2.7-fold increase for ascorbic acid treatment, derivatization notably improved the retentions of these cytosine
2
oxidation,
(
5
This journal is ©The Royal Society of Chemistry 2015
Chem. Commun.

View Article Online
Communication
ChemComm
modifications on reversed-phase chromatography and dramati- 10 J. Xiong, H. P. Jiang, C. Y. Peng, Q. Y. Deng, M. D. Lan, H. Zeng,
F. Zheng, Y. Q. Feng and B. F. Yuan, Clin. Epigenet., 2015, 7, 72.
1 C. He, Nat. Chem. Biol., 2010, 6, 863–865.
cally increased the detection sensitivities of 5-hmdC and 5-hmrC
1
1
by 363- and 380-fold, respectively. Using this ODMS method, we
successfully quantified 5-hmdC, 5-hmrC, 5-fodC and 5-forC in
DNA and RNA of mammalian cells. We reported for the first time
the presence of endogenous 5-forC in RNA of mammalian cells. In
addition, we found that both ascorbic acid and hydroquinone
treatment of the mammalian cells can stimulate the oxidation of
2 M. A. Machnicka, K. Milanowska, O. Osman Oglou, E. Purta,
M. Kurkowska, A. Olchowik, W. Januszewski, S. Kalinowski,
S. Dunin-Horkawicz, K. M. Rother, M. Helm, J. M. Bujnicki and
H. Grosjean, Nucleic Acids Res., 2013, 41, D262–D267.
3 J. E. Squires, H. R. Patel, M. Nousch, T. Sibbritt, D. T. Humphreys,
B. J. Parker, C. M. Suter and T. Preiss, Nucleic Acids Res., 2012, 40,
1
1
5023–5033.
4 L. Fu, C. R. Guerrero, N. Zhong, N. J. Amato, Y. Liu, S. Liu, Q. Cai,
D. Ji, S. G. Jin, L. J. Niedernhofer, G. P. Pfeifer, G. L. Xu and Y. Wang,
J. Am. Chem. Soc., 2014, 136, 11582–11585.
5 Y. Tang, J. M. Chu, W. Huang, J. Xiong, X. W. Xing, X. Zhou,
Y. Q. Feng and B. F. Yuan, Anal. Chem., 2013, 85, 6129–6135.
6 S. Ito, L. Shen, Q. Dai, S. C. Wu, L. B. Collins, J. A. Swenberg, C. He
and Y. Zhang, Science, 2011, 333, 1300–1303.
7 B. F. Yuan and Y. Q. Feng, TrAC, Trends Anal. Chem., 2014, 54, 24–35.
8 B. F. Yuan, Adv. Clin. Chem., 2014, 67, 151–187.
9 M. C. Haffner, A. Chaux, A. K. Meeker, D. M. Esopi, J. Gerber,
L. G. Pellakuru, A. Toubaji, P. Argani, C. Iacobuzio-Donahue,
W. G. Nelson, G. J. Netto, A. M. De Marzo and S. Yegnasubramanian,
Oncotarget, 2011, 2, 627–637.
5-mC to 5-hmC in both DNA and RNA. The developed ODMS
strategy is an efficient method for the sensitive and simultaneous
determination of 5-hmC and 5-foC in both DNA and RNA, which
can promote the functional studies of epigenetic modifications.
The authors thank the financial support from the National
Basic Research Program of China (973 Program) (2013CB910702,
1
1
1
1
1
2012CB720601), and the National Natural Science Foundation of
China (21522507, 21475098).
References
2
0 J. Hu, X. Xing, X. Xu, F. Wu, P. Guo, S. Yan, Z. Xu, J. Xu, X. Weng and
1
G. Egger, G. Liang, A. Aparicio and P. A. Jones, Nature, 2004, 429,
57–463.
L. Shen, C. X. Song, C. He and Y. Zhang, Annu. Rev. Biochem., 2014,
3, 585–614.
X. Zhou, Chem. – Eur. J., 2013, 19, 5836–5840.
4
21 P. Guo, S. Yan, J. Hu, X. Xing, C. Wang, X. Xu, X. Qiu, W. Ma, C. Lu,
X. Weng and X. Zhou, Org. Lett., 2013, 15, 3266–3269.
22 Y. Tang, J. Xiong, H. P. Jiang, S. J. Zheng, Y. Q. Feng and B. F. Yuan,
Anal. Chem., 2014, 86, 7764–7772.
2
8
3
4
S. Kriaucionis and N. Heintz, Science, 2009, 324, 929–930.
M. Tahiliani, K. P. Koh, Y. Shen, W. A. Pastor, H. Bandukwala, 23 R. Yin, J. Mo, M. Lu and H. Wang, Anal. Chem., 2015, 87, 1846–1852.
Y. Brudno, S. Agarwal, L. M. Iyer, D. R. Liu, L. Aravind and A. Rao, 24 S. Liu, J. Wang, Y. Su, C. Guerrero, Y. Zeng, D. Mitra, P. J. Brooks,
Science, 2009, 324, 930–935.
Y. F. He, B. Z. Li, Z. Li, P. Liu, Y. Wang, Q. Tang, J. Ding, Y. Jia,
D. E. Fisher, H. Song and Y. Wang, Nucleic Acids Res., 2013, 41,
5
6421–6429.
Z. Chen, L. Li, Y. Sun, X. Li, Q. Dai, C. X. Song, K. Zhang, C. He and 25 R. Yin, S. Q. Mao, B. Zhao, Z. Chong, Y. Yang, C. Zhao, D. Zhang,
G. L. Xu, Science, 2011, 333, 1303–1307.
M. R. Branco, G. Ficz and W. Reik, Nat. Rev. Genet., 2012, 13, 7–13.
Y. Tang, S. J. Zheng, C. B. Qi, Y. Q. Feng and B. F. Yuan, Anal. Chem.,
H. Huang, J. Gao, Z. Li, Y. Jiao, C. Li, S. Liu, D. Wu, W. Gu,
Y. G. Yang, G. L. Xu and H. Wang, J. Am. Chem. Soc., 2013, 135,
10396–10403.
6
7
2
015, 87, 3445–3452.
26 K. Blaschke, K. T. Ebata, M. M. Karimi, J. A. Zepeda-Martinez,
P. Goyal, S. Mahapatra, A. Tam, D. J. Laird, M. Hirst, A. Rao,
M. C. Lorincz and M. Ramalho-Santos, Nature, 2013, 500, 222–226.
8
9
M. L. Chen, F. Shen, W. Huang, J. H. Qi, Y. Wang, Y. Q. Feng,
S. M. Liu and B. F. Yuan, Clin. Chem., 2013, 59, 824–832.
S. G. Jin, Y. Jiang, R. Qiu, T. A. Rauch, Y. Wang, G. Schackert, 27 J. B. Coulter, C. M. O’Driscoll and J. P. Bressler, J. Biol. Chem., 2013,
D. Krex, Q. Lu and G. P. Pfeifer, Cancer Res., 2011, 71, 7360–7365. 288, 28792–28800.
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2015