Fluorescent Wittig reagent as a novel ratiometric probe for the quantification of 5-formyluracil and its application in cell imaging

Article abstract of DOI:10.1039/c8cc07541g

The chemically selective detection of natural nucleobase modifications has been regarded as the key step in understanding their important roles in epigenetics. Herein, for the first time, we introduce a Wittig reaction into the design of reaction-based fluorescent probes for ratiometrically detecting 5fU, selectively labelling 5fU-modified DNA and imaging intracellular 5fU produced by γ-irradiation.

Full text of DOI:10.1039/c8cc07541g

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Fluorescent Wittig reagent as a novel ratiometric
probe for the quantification of 5-formyluracil and
its application in cell imaging†
Cite this: Chem. Commun., 2018,
54, 13722
Received 18th September 2018,
Accepted 5th November 2018
Qian Zhou, Kun Li, * Yan-Hong Liu, Ling-Ling Li, Kang-Kang Yu, Hong Zhang
and Xiao-Qi Yu
*
DOI: 10.1039/c8cc07541g
rsc.li/chemcomm
The chemically selective detection of natural nucleobase modifications indole,12 benzothiazole13 and 1,2-phenylendiamine14 derivatives
has been regarded as the key step in understanding their important roles for applications as potential fluorescent labels or small molecule
in epigenetics. Herein, for the first time, we introduce a Wittig reaction tags to attach to 5fU or/and 5fC-modified DNA for detection,
into the design of reaction-based fluorescent probes for ratiometrically sequencing and enrichment.15 Taking advantage of widely used
detecting 5fU, selectively labelling 5fU-modified DNA and imaging fluorescent probe technology and chemoselective reactions, these
intracellular 5fU produced by c-irradiation.
original studies have provided effective platforms for biological
events analysis. However, it should be pointed out that their
As natural nucleobase modifications produced by oxidation of quantitation mechanism primarily involves a Schiff base reaction
thymine and cytosine, 5-formyluracil (5fU) and 5-formylcytosine or Aldol-type (Knoevenagel-like) condensation, although their
(5fC) play important roles in epigenetics.1 Recent discoveries probes vary in structure. Therefore, the development of new
have indicated that 5fU and 5fC are highly involved in DNA aldehyde-reagents based on other organic chemistry principles
methylation and demethylation processes.2 They are also regarded to expand the library of fluorescent probes toward 5fU is of great
as oxidative damage3 which results in gene regulation by causing significance and necessity.
genotoxic lesions, introducing base mispairing and inhibiting
In the past few years, we have gained a lot of experience
sequence-specific DNA–protein interactions.4 Compared with in designing reaction-based probes for reactive oxygen/sulfur
5fC, 5fU is more prone to exist in the ionized form, which is species as well as others.16 Herein, our intention is to construct
likely responsible for the high mutagenesis frequency of 5fU conjugated CQC bonds by introducing new chemical reactions.
during DNA replication under physiological conditions both The Wittig reaction,17 discovered by Georg Wittig in 1953 and
in vitro and in vivo.5 Recently, Balasubramanian et al. revealed winner of the Nobel prize in 1979, is a general methodology for
that it is 5fU in synthetic oligodeoxynucleotides that alters DNA synthesizing alkenes from aldehydes or ketones. Given the
structures.6 Therefore, developing methods to sensitively and outstanding contributions of Wittig olefination in organic
selectively detect formyl pyrimidines can not only help to under- synthesis, a few scientists have used Wittig reagents to modify
stand their epigenetic roles, but can also provide a detection proteins bearing an ‘‘aldehyde tag’’ in recent years,18 and we
technique for studying the relationship between formyl pyrimidines will take on the challenge of its application in nucleic acid
and diseases. In particular, quantitative analysis methods for 5fU in labelling, which has never been reported so far.
DNA with high accuracy are urgently required.
In this work, we have proved the feasibility of the Wittig
To date, HPLC-MS together with complete enzymatic digestion reaction in detecting nucleoside modifications. As outlined in
of target DNA has been well established as a powerful tool for Scheme 1, a coumarin scaffold was selected as the fluorophore
5fU/5fC detection, although it is time-consuming, expensive owing to its good photostability, high quantum yield and large
and requires an isotope-labelled internal standard.7 In the last Stokes shift. Furthermore, the lactone structure coupled with a
ten years, focus has been on the highly reactive 5-formyl group, 7-carbonyl group can stabilize the adjacent carbanion, making
and O-amino(thio)phenol,8 amine,9 hydroxylamine,10 hydrazide,11 it easy to preserve and avoiding anhydrous/anaerobic environ-
ments when reacting with aldehydes. Our elaborately designed
Key Laboratory of Green Chemistry and Technology (Ministry of Education),
fluorescent Wittig reagent Ylide U (abbr. as YU), could selectively
tag 5fU in DNA via a Wittig reaction, while 5fC and the abasic sites
(AP) weren’t disturbed. After the reaction, YU–5fU (Scheme S3 and
College of Chemistry, Sichuan University, Chengdu 610064, P. R. China.
E-mail: kli@scu.edu.cn, xqyu@scu.edu.cn
† Electronic supplementary information (ESI) available: Experimental proce-
Fig. S3, ESI†) with an emission peak at 555 nm was obtained in
good yield. In addition, we also designed YU-A, YU-B and YU-C
dures, characterization of all compounds, supplementary UV absorption spectra
and fluorescence emission spectra. See DOI: 10.1039/c8cc07541g
13722 | Chem. Commun., 2018, 54, 13722--13725
This journal is ©The Royal Society of Chemistry 2018

View Article Online
Communication
ChemComm
Scheme 1 The reported strategy for fluorogenic labelling of 5-formyluracil and the novel strategy for fluorogenic labelling of 5-formyluracil in DNA
based on the Wittig reaction.
(Schemes S6 and S7, ESI†), in which naphthalene, coumarin and exhibited an absorption maximum at 411 nm and an emission
tetraphenylethylene (TPE) served as fluorophores, respectively. maximum at 498 nm (F = 0.5% in DMSO), respectively. Upon
However, we failed to synthesise YU-B and YU-C due to the addition of 5fU, the absorption intensity at 411 nm decreased and
instability of the phenol ester groups in their structures. Although a redshift of 57 nm, as well as a deeper colour, was observed.
YU-A was successfully obtained, it partially decomposed during Concomitantly, the characteristic emission band belonging to YU
alkalization before reacting with 5fU. YU was synthesized readily vanished completely, while an intense fluorescence emission band
in four steps using 4-(diethylamino)salicylaldehyde and ethyl centred at 555 nm appeared, accompanied with a green-to-yellow
acetoacetate as the staring materials (Scheme S1, ESI†). The solution fluorescence colour change. Notably, the spectral
structure of YU was fully characterized by 1H NMR, 13C NMR, performance of the reaction mixture is in good agreement with
and HRMS (see ESI†).
To explore the feasibility of our design, we first studied the that the yield of this Wittig detection was really high.
Wittig reactivity between YU and the 5fU nucleoside in CH3OH We then evaluated the specificity of YU for 5fU compared to
pure YU–5fU (F = 9.5% in DMSO, see Fig. S3, ESI†), indicating
(Scheme S3, ESI†). As expected, after incubation at 37 1C for 24 h, an various potential interfering species, including five canonical
orange product which proved to be a YU–5fU adduct was obtained in deoxynucleosides and their natural modifications, named
more than 85% yield. Then, we measured the UV absorbance and 5hmC, 5hmU, and 5fC under the optimized conditions. The
fluorescence emission properties of YU and YU–5fU in various data (Fig. 2a and b) show that other deoxynucleosides induced
solvents. As shown in Fig. S1, and Tables S1 and S2 (ESI†), different no obvious spectral changes for YU, and only 5fU triggered a
degrees of red shift were observed in both absorption and emission remarkable enhancement of the emission ratio I555/I485. At the
spectra, which is consistent with our theoretical prediction and also
indicates that YU could be applied as a ratiometric fluorescent probe
for the recognition of 5fU. The detection reaction toward 5fU could
proceed smoothly in pure water or under weak acid/base conditions
(Fig. S2, ESI†). All of these results prompted us to optimize our
experimental conditions such as solvent, temperature, reaction time,
etc. Finally, the detection was carried out in EtOH/H2O (1/1, v/v) at
37 1C for 48 h. As can be seen from Fig. 1, before the reaction, YU
Fig. 2 (a) Fluorescence spectra and (b) the emission intensity ratio
(I555/I485) of YU after being incubated with 5fU and other interfering
Fig. 1 (a) Normalized UV/vis absorption spectra and (b) fluorescence
emission spectra (lex: 460 nm) of YU before (blue) and after (red) reaction
with 5fU in EtOH/H2O medium (1/1, v/v) at 37 1C for 48 h. Inset: Images of
YU before and after reaction with 5fU under daylight or UV lamp (365 nm).
species, including A, G, C, T, U, 5hmC, 5hmU and 5fC. 10 mM YU without
any nucleoside serving as the control. (c) Fluorescence titration spectra
and (d) linear relationship of the emission intensity ratio (I555/I485) of YU
(10 mM) toward 5fU (0–300 mM). lex: 460 nm.
This journal is ©The Royal Society of Chemistry 2018
Chem. Commun., 2018, 54, 13722--13725 | 13723

View Article Online
ChemComm
Communication
same time, only the reaction liquid corresponding to 5fU the control lanes. The gel was then stained with 3Â SuperRed to
undergoes a colour change (Fig. S4, ESI†), which might be image the remaining DNA bands. The PAGE experiment con-
ascribed to its active aldehyde group and the highly selective firmed the ability of YU to distinguish ODN-5fU from other DNA
Wittig reaction. In 2011, Thomas Carell et al. first reported the modifications once again.
crystal structure of 5fC, and revealed a strong intramolecular
Given that YU shows an excellent performance in detecting
H-bond between the exocyclic amino group NH2(4) and 5fU in buffer, we proceed to investigate whether it is capable of
the carbonyl–oxygen at C5,19 which was suggested to be the imaging 5fU in HeLa cells. Pouget20 and Wang et al.7a have reported
molecular reason why 5fC is less reactive than 5fU and why our that g-rays would increase 5fU mutations in some mammalian cells.
Wittig reagent can react with 5fU rather than 5fC.
We hence exposed HeLa cells to a 60Co g-source at a dose rate of
Subsequently, fluorescence titration experiments of YU 18 Gy minÀ1 (60 min, r.t.).14b Immediately after irradiation, the cells
toward 5fU were conducted. As depicted in Fig. 2c and d, with were fixed and incubated with YU at 37 1C. As can be seen from
an increase of 5fU (0–300 mM), the fluorescence emission band Fig. 4a–i, upon excitation at 488 nm, g-irradiated HeLa cells,
centred at 485 nm drops gradually and a distinct rise at 555 nm generating much more 5fU, displayed bright fluorescence in the
occurs simultaneously, indicating a 5fU-mediated transforma- emission range of 500–600 nm, whereas those without g-irradiation
tion from YU to YU–5fU. The detection limit of YU toward 5fU maintained weak fluorescence. Fig. 4j shows that there was almost
was calculated to be 0.33 mM. These results verify that YU can 1-fold enhancement in the average fluorescence intensities caused
sensitively detect 5fU under mild conditions.
by g-exposure or increased 5fU. In fact, the bulk of DNA damage
Encouraged by the above-mentioned findings, we further induced by the g-source is mediated by hydroxyl radicals (ꢀOH),
estimated the ability of YU to label 5fU-modified DNA. Generally, which act as an indirect effect of water radiolysis.21 In view of this, we
we chose an oligodeoxynucleotide containing one 5fU (ODN-5fU) pre-treated HeLa cells with Fenton’s reagent (FeSO4 :H2O2 = 1:5)
as a model and incubated it with a large excess of YU (50 eq.) to and then incubated them with YU for different times. Confocal
ensure complete DNA labelling under optimized conditions. microscopy analysis reveals that the fluorescence became stronger as
Considering that ODN-5fC (where the 5fU site is replaced by the incubation time increased (Fig. S6a, ESI†). At the same time, the
5fC), ODN-AP (contains one abasic site bearing an aldehyde cells pre-treated with 5fU before incubation with YU exhibit the same
source), ODN-C, and ODN-T (where the 5fU site is C or T) might performance (Fig. S6b, ESI†). Furthermore, we also treated HeLa
ꢀ
affect the Wittig reaction between YU and ODN-5fU, the fluores- cells with different doses of OH to stimulate the cells to produce
cence responses of YU toward these interfering species were different doses of 5fU mutations. As shown in Fig. S7a (ESI†), a
tested at the same time. The data (Fig. 3a and Fig. S5, ESI†) progressively brighter fluorescence was observed. However, com-
ꢀ
ꢀ
demonstrate that only 5fU showed an 8-fold fluorescence pared to 25 mM OH-stimulated cells, 50 mM OH-stimulated cells
enhancement at 555 nm, other ODNs, including ODN-5fC and did not exhibit significant fluorescence enhancement, probably
ODN-AP, showed negligible spectral fluctuation on YU. The because 25 mM ꢀOH caused saturation to be reached, or the cell
superior selectivity was attributed to the higher reactivity of began to repair the 5fU lesion more extensively over such a long
ODN-5fU and the specificity of the Wittig reagent.
incubation period. This may also be the reason why the fluorescence
We further attempted a denaturing polyacrylamide gel electro- intensity is consistent even if the cells are pre-treated with different
phoresis (PAGE) experiment. Individually, ODN-5fC, ODN-AP, ODN-T, concentrations of 5fU (Fig. S7b, ESI†). All of the above results imply
and ODN-C were incubated with YU in EtOH/H2O (1/1, v/v) at 37 1C. that YU could respond to intracellular 5fU.
48 h later, the reaction mixtures were directly mixed with a
loading buffer and then underwent electrophoresis without
further purification. Once a fluorophore was introduced into
DNA, a fluorescent band was visualized on the gel imaging
system. As shown in Fig. 3b, we indeed observed a distinct
fluorescent band in lane 5, corresponding to the reaction
product derived from ODN-5fU, while no signal was found in
Fig. 4 (a–i) Confocal fluorescence images of g-irradiated HeLa cells with
YU (10 mM). (j) Statistical analysis (average fluorescent intensity calculated
from d and g) was performed using Student’s t-test (n = 10 fields of cells).
***P o 0.001, error bars are ÆSEM. Scale bar: 10 mm. The images were
acquired upon excitation at 488 nm and fluorescence emission was
collected at 500–600 nm.
Fig. 3 (a) Fluorescence intensity of YU at 555 nm after reaction with ODN-5fU,
ODN-5fC, ODN-AP, ODN-T and ODN-C. 10 mM YU without any ODN served as
the control. lex: 495 nm. (b) PAGE analysis of ODN-5fU, ODN-5fC, ODN-AP,
ODN-T and ODN-C after incubation with YU. The images before and after
SuperRed staining are above and below the dashed line, respectively.
13724 | Chem. Commun., 2018, 54, 13722--13725
This journal is ©The Royal Society of Chemistry 2018

View Article Online
Communication
ChemComm
8 (a) W. Hirose, K. Sato and A. Matsuda, Angew. Chem., Int. Ed., 2010,
122, 8570–8572; (b) W. Hirose, K. Sato and A. Matsuda, Eur. J. Org.
Chem., 2011, 6206–6217; (c) D. B. Gophane and S. T. Sigurdsson,
Chem. Commun., 2013, 49, 999–1001.
9 J.-L. Hu, X.-W. Xing, X.-W. Xu, F. Wu, P. Guo, S.-Y. Yan, Z.-H. Xu,
J.-H. Xu, X.-C. Weng and X. Zhou, Chem. – Eur. J., 2013, 19, 5836–5840.
10 (a) P. Guo, S.-Y. Yan, J.-L. Hu, X.-W. Xing, C.-C. Wang, X.-W. Xu,
X.-Y. Qiu, W. Ma, C.-J. Lu, X.-C. Weng and X. Zhou, Org. Lett., 2013,
15, 3266–3269; (b) S.-R. Wang, Y.-Y. Song, L. Wei, C. X. Liu, B.-S. Fu,
J.-Q. Wang, X.-R. Yang, Y.-N. Liu, S.-M. Liu, T. Tian and X. Zhou,
J. Am. Chem. Soc., 2017, 139, 16903–16912.
11 (a) C.-X. Liu, Y.-Q. Chen, Y.-F. Wang, F. Wu, X. Zhang, W. Yang,
J.-Q. Wang, Y. Chen, Z.-Y. He, G.-R. Zou, S.-R. Wang and X. Zhou,
Nano Res., 2017, 10, 2449–2458; (b) Y.-F. Wang, C.-X. Liu, W. Yang,
G.-R. Zou, X. Zhang, F. Wu, S.-Y. Yu, X.-M. Luo and X. Zhou, Chem.
Commun., 2018, 54, 1497–1500; (c) L. Xu, Y.-C. Chen, J. Chong,
A. Fin, L. S. McCoy, J. Xu, C. Zhang and D. Wang, Angew. Chem., Int.
Ed., 2014, 53, 11223–11227; (d) L. Xu, Y.-C. Chen, S. Nakajima,
J. Chong, L. Wang, L. Lan, C. Zhang and D. Wang, Chem. Sci., 2014,
5, 567–574.
In summary, we have established a novel strategy to design
and synthesize a small molecular fluorescent probe based on
the Wittig reaction, for ratiometrically detecting 5fU, labelling
5fU-modified DNA and imaging intracellular 5fU. The sensitivity
and selectivity of a coumarin-derived phosphorus ylide toward
an active aldehyde were confirmed by our studies. These results
broaden the application of the Wittig reaction and offer an
attractive candidate for marking aldehyde-containing biologically
related species.
This work was financially supported by the National Natural
Science Foundation of China (No: 21572147 and 21877082).
Conflicts of interest
There are no conflicts to declare.
12 B. Samanta, J. Seikowski and C. Hobartner, Angew. Chem., Int. Ed.,
2016, 55, 1912–1916.
13 C.-X. Liu, Y.-F. Wang, W. Yang, F. Wu, W.-W. Zeng, Z.-G. Chen,
J.-G. Huang, G.-R. Zou, X. Zhang, S.-R. Wang, X.-C. Weng, Z.-G. Wu,
Y. Zhou and X. Zhou, Chem. Sci., 2017, 8, 7443–7447.
Notes and references
1 E.-A. Raiber, R. Hardisty, P. van Delft and S. Balasubramanian, Nat.
Rev. Chem., 2017, 1, 2–13.
14 (a) J. Krim, C. Grunewald, M. Taourirte and J. W. Engels, Bioorg.
Med. Chem., 2012, 20, 480–486; (b) C.-X. Liu, Y.-F. Wang, X. Zhang,
F. Wu, W. Yang, G.-R. Zou, Q. Yao, J.-Q. Wang, Y.-Q. Chen, S.-R.
Wang and X. Zhou, Chem. Sci., 2017, 8, 4505–4510.
15 R. E. Hardisty, F. Kawasaki, A. B. Sahakyan and S. Balasubramanian,
J. Am. Chem. Soc., 2015, 137, 9270–9272.
16 (a) K. Li, J.-T. Hou, J. Yang and X.-Q. Yu, Chem. Commun., 2017, 53,
5539–5541; (b) J. Yang, K. Li, J.-T. Hou, L.-L. Li, C.-Y. Lu, Y.-M. Xie,
X. Wang and X.-Q. Yu, ACS Sens., 2016, 1, 166–172; (c) Y. Liu, K. Li,
M.-Y. Wu, Y.-H. Liu, Y.-M. Xie and X.-Q. Yu, Chem. Commun., 2015,
51, 6781–6784; (d) M.-Y. Wu, K. Li, C.-Y. Li, J.-T. Hou and X.-Q. Yu,
Chem. Commun., 2014, 50, 183–185; (e) J.-T. Hou, M.-Y. Wu, K. Li,
J. Yang, K.-K. Yu, Y.-M. Xie and X.-Q. Yu, Chem. Commun., 2014, 50,
8640–8643.
17 G. Wittig and G. Geissler, Justus Liebigs Ann. Chem., 1953, 580,
44–57.
18 (a) M.-J. Han, D.-C. Xiong and X.-S. Ye, Chem. Commun., 2012, 48,
11079–11081; (b) K. M. Lum, V. J. Xavier, M. J. H. Ong,
C. W. Johannes and K.-P. Chan, Chem. Commun., 2013, 49,
11188–11190.
19 M. Munzel, U. Lischke, D. Stathis, T. Pfaffeneder, F. A. Gnerlich,
C. A. Deiml, S. C. Koch, K. Karaghiosoff and T. Carell, Chem. – Eur. J.,
2011, 17, 13782–13788.
20 J. P. Pouget, S. Frelon, J. L. Ravanat, I. Testard, F. Odin and J. Cadet,
Radiat. Res., 2002, 157, 589–595.
2 J. S. Hardwick, A. N. Lane and T. Brown, BioEssays, 2018, 40,
1700199.
3 N. C. Bauer, A. H. Corbett and P. W. Doetsch, Nucleic Acids Res.,
2015, 43, 10083–10101.
4 (a) A. Klungland, R. Paulsem, V. Rolseth, Y. Yamada, Y. Ueno,
P. WiiK, A. Matsuda, E. Seeberg and S. Bjelland, Toxicol. Lett.,
2001, 119, 71–78; (b) D. K. Rogstad, P. F. Liu, A. Burdzy, S. S. Lin
and L. C. Sowers, Biochemistry, 2002, 41, 8093–8102; (c) C. Dohno,
A. Okamoto and I. Saito, J. Am. Chem. Soc., 2005, 127, 16681–16684;
(d) F. Li, Y. Zhang, J. Bai, M. M. Greenberg, Z. Xi and C. Zhou, J. Am.
Chem. Soc., 2017, 139, 10617–10620.
5 (a) E. J. Privat and L. C. Sowers, Mutat. Res., 1996, 354, 151–156;
(b) M. Yoshida, K. Makino, H. Morita, H. Terato, Y. Ohyama and
H. Ide, Nucleic Acids Res., 1997, 25, 1570–1577; (c) K. Fujikawa,
H. Kamiya and H. Kasai, Nucleic Acids Res., 1998, 26, 4582–4587;
(d) H. Anensen, F. Provan, A. T. Lian, S. Reinertsen, Y. Ueno,
A. Matsuda, E. Seeberg and S. Bjelland, Mutat. Res., 2001, 476,
99–107.
6 F. Kawasaki, P. Murat, Z. Li, T. Santner and S. Balasubramanian,
Chem. Commun., 2017, 53, 1389–1392.
7 (a) H.-Z. Hong and Y.-S. Wang, Anal. Chem., 2007, 79, 322–326;
(b) J. Wang, B.-F. Yuan, C. Guerrero, R. Bahde, S. Gupta and Y.-S.
Wang, Anal. Chem., 2011, 83, 2201–2209; (c) H.-P. Jiang, T. Liu,
N. Guo, L. Yu, B.-F. Yuan and Y.-Q. Feng, Anal. Chim. Acta, 2017, 981,
1–10; (d) Y. Tang, J. Xiong, H.-P. Jiang, S.-J. Zheng, Y.-Q. Feng and
B.-F. Yuan, Anal. Chem., 2014, 86, 7764–7772; (e) Y. Tang, S.-J. Zheng,
C.-B. Qi, Y.-Q. Feng and B.-F. Yuan, Anal. Chem., 2015, 87,
3445–3452.
21 (a) H. Kasai, A. Iida, Z. Yamaizumi, S. Nishimura and H. Tanooka,
Mutat. Res., 1990, 243, 249–253; (b) J. Cadet and J. R. Wagner, Cold
Spring Harbor Perspect. Biol., 2013, 5, a012559.
This journal is ©The Royal Society of Chemistry 2018
Chem. Commun., 2018, 54, 13722--13725 | 13725