Thiazolidin-5-imine Formation as a Catalyst-Free Bioorthogonal Reaction for Protein and Live Cell Labeling

Article abstract of DOI:10.1021/acs.orglett.8b03195

A previously undescribed reaction involving the formation of a thiazolidin-5-imine linkage was developed for bioconjugation. Being highly specific and operating in aqueous media, this simple condensation reaction is used to chemoselectively label peptides, proteins, and living cells under physiological conditions without the need to use toxic catalysts or reducing reagents.

Full text of DOI:10.1021/acs.orglett.8b03195

Letter
pubs.acs.org/OrgLett
Cite This: Org. Lett. XXXX, XXX, XXX−XXX
Thiazolidin-5-imine Formation as a Catalyst-Free Bioorthogonal
Reaction for Protein and Live Cell Labeling
Xiaobao Bi,† Juan Yin,‡ Chang Rao,† Seetharamsing Balamkundu,§ Biplab Banerjee,† Dingpeng Zhang,†
Dawei Zhang,⊥ Peter C. Dedon,§ and Chuan-Fa Liu*,†
†School of Biological Sciences, Nanyang Technological University, 60 Nanyang Drive, Singapore 637551, Singapore
‡Program in Neuroscience and Behavioural Disorders, Duke-NUS Medical School, 8 College Road, Singapore 169857, Singapore
§Singapore-MIT Alliance for Research and Technology, 1 Create Way, Singapore 138602, Singapore
⊥Institute of Bioinformatics and Medical Engineering, School of Electrical and Information Engineering, Jiangsu University of
Technology, Changzhou 213001, China
S
* Supporting Information
ABSTRACT: A previously undescribed reaction involving the
formation of a thiazolidin-5-imine linkage was developed for
bioconjugation. Being highly specific and operating in aqueous
media, this simple condensation reaction is used to chemo-
selectively label peptides, proteins, and living cells under
physiological conditions without the need to use toxic catalysts or reducing reagents.
hemical modification of biomolecules such as proteins is
aldehyde group in a highly selective manner to form a
Cextensively used to improve their physicochemical and thiazolidin-5-imine linkage. We show that this simple, yet
previously undescribed reaction system is a useful bioconjuga-
tion method for the modification and labeling of peptides,
proteins, and live cells.
We first demonstrated this reaction scheme in a model study
using small-molecule compounds 1 and 2 (Scheme 1). Because
pharmacological properties and to facilitate structure−function
studies. Direct modification of proteins can be achieved by
targeting the naturally existing functional groups of the amino
acid residues such as the commonly used amino or thiol groups.1
Although this strategy represents a straightforward way to
modify proteins, it lacks site specificity due to the presence of
multiple lysine and cysteine residues in a protein. To overcome
this problem, uniquely reactive unnatural functionalities that are
not present in the canonical amino acids, such as aldehyde,
ketone, tetrazine, alkene, alkyne, and azide, are synthetically,
biosynthetically, or enzymatically introduced into proteins. This
makes it possible to precisely modify the protein via a
chemoselective reaction between the unnatural functionality
and a mutually reactive functional group on a modifying
reagent.2 To date, a variety of such mutually reactive pairs of
functional groups have been developed for site-specific
bioconjugation including the condensation between an
aldehyde/ketone and an alkoxy-amine/hydrazine, the Stau-
dinger ligation, the Cu (I)-catalyzed and strain promoted
cycloaddition between alkynes and azides, the inverse electron
demand Diels−Alder reactions between tetrazines and strained
alkenes.3 However, many of the available chemistries have their
shortcomings such as synthetic difficulty to obtain the chemical
probes containing the required orthogonal functionality, acidic
reaction conditions that are incompatible for sensitive proteins,
or the need for toxic catalysts or reducing reagents. Therefore,
there is extensive interest in developing new biorthogonal and
catalyst-free conjugation reactions that can be performed in
aqueous solution under physiological pH. Herein, we describe a
new condensation reaction that fulfills such requirements. We
find that a 2-aminoethanethioamide moiety can react with an
Scheme 1. Model Reaction between 1 and α-Oxo aldehyde 2
To Yield Condensation Product 3 and the Proposed
Mechanism
of its poor solubility in water, 1 was allowed to react with 2 in
PBS/acetonitrile/tBuOH mixture (2:2:5) at 37 °C. After 4.5 h,
the product 3 was formed in ∼85% yield. Then 3 was isolated
and characterized by NMR (1H and 13C) and mass spectrometry
1
(Figures S1 and S2 in the Supporting Information (SI)). H
NMR analysis clearly assigned the peaks to the protons on C2
and C4 of the five-membered ring thiazolidin-5-imine structure
of 3 (Figure S1). As a plausible reaction mechanism, the amine
of 1 first reacts with the aldehyde to form the imine Schiff base,
which was attacked by the sulfur of the thioamide tautomer to
form the thiazolidin-5-imine linkage (Scheme 1). Clearly, the
driving force here is ring formation, a mechanism that is similar
Received: October 7, 2018
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.8b03195
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
to the condensation reaction between an aldehyde and a 1,2-
aminothiol moiety.4 To test the stability of 3 in the aqueous
media, we incubated 3 in the aqueous buffer at pH 4.5, 7.2, and
8.5 at 37 °C. At different time points (0, 2, 4, 8, 24, 72, 168 h), an
equal aliquot of the sample was taken for HPLC analysis. Results
show that 3 was quite stable in a neutral and slightly basic
aqueous buffer (pH 7.2 and 8.5) in which only a small portion of
3 was hydrolyzed after 168 h at 37 °C (Figure S3 in SI).
However, it was susceptible to hydrolysis in the acidic buffer at
pH 4.5 as a significant degree of hydrolysis was observable after
incubation for 2 h (Figure S3).
Next, we tested whether we could modify α-oxo aldehyde-
containing proteins with the thioamide functionalized labeling
reagents (Figure 2). Using the same NaIO4 oxidation method,
Encouraged by these results, we proceeded to test this
reaction for chemoselective conjugation using a peptide.
Therefore, we synthesized an aldehyde-containing peptide, H-
IETSAK(X)T-NH2, where X = α-oxo aldehyde, which was
introduced via NaIO4 oxidation of the 1,2-aminoethanol moiety
of a Ser residue (Scheme 2 and SI). The condensation reaction
Figure 2. Structures of the thioamide-containing small compounds and
peptide probes used in this study. Fluorescent probe 9 was used as the
control.
Scheme 2. Generation of α-Oxo aldehyde in Model Peptide
and Its Subsequent Bioconjugation with 1
the aldehyde functionality was introduced onto ubiquitin at its
N-terminus via the oxidation of the first Ser residue (SI). Then
ubiquitin-aldehyde 4 at 50 μM was incubated with 5 or 10 equiv
of 1 in PBS (pH 7.2) at 4 and 24 °C, respectively. As seen by
HPLC and MS analysis of the reaction mixture (Figure 3A,B),
was conducted at intentionally high dilutions of the reactants
(Scheme 2). Thus, the α-oxo aldehyde-containing peptide (0.1
mM) was incubated with 1 (0.2 mM) in PBS (pH 7.0) and the
reaction was monitored by analytical HPLC and mass
spectrometry. To our delight, after 1 h, the conjugation product
could be observed on HPLC (peak c in Figure 1A). After 12 h,
about 65−70% of the starting peptide was converted to the
desired conjugate (observed mass = 952.47) (Figure 1A,B).
These initial data indicate that this reaction is potentially
compatible with large biomolecules.
Figure 3. Characterization of the conjugation reaction between 1, 5,
and ubiquitin-aldehyde 4. (A) HPLC analysis of the bioconjugation of
1 with 4 for overnight at 4 and 24 °C, respectively. (B) Mass
spectrometry analysis showing the mass of the conjugated product
(calcd mass: 9531.9, found mass: 9534.0). (C) HPLC analysis of the
bioconjugation of 5 with 4 for overnight at 4 and 24 °C, respectively.
(D) Mass spectrometry analysis showing the conjugated product (calcd
mass: 9555.9, found mass: 9558.0).
ubiquitin-aldehyde 4 was quantitatively conjugated with 1 after
overnight reaction even at 4 °C. This result points to the
potential of this method to label proteins that are sensitive to
thermal stress. We also prepared a 2-aminoethanethioamide
derivative 5, which contains an alkyne group and performed the
conjugation reaction with 4 under the same conditions. Again, it
was shown that 4 was also quantitatively modified with 5 after
overnight reaction (Figure 3C,D). Next, to label the protein with
Figure 1. Characterization of the conjugation reaction of 1 with α-oxo
aldehyde-containing peptide H-IETSAK(aldehyde)T-NH2. (A) Ana-
lytical HPLC analysis of the reaction at different time slots. Peak a is H-
IETSAK(aldehyde)T-NH2, peak b is 1, and peak c is the conjugated
product. (B) Mass spectrometry analysis showing the mass of H-
IETSAK(aldehyde)T-NH2 (calcd mass: 803.40, found m/z: 804.45 [M
+ H]+ and 822.42 [M+H2O+H]+) and the conjugated product (calcd
mass: 951.45, found m/z: 952.47 [M + H]+).
B
DOI: 10.1021/acs.orglett.8b03195
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
a fluorescent dye, we generated the dansyl-based fluorescent
compound 6. Initially, incubation with 6 resulted in very poor
labeling of 4 in the aqueous PBS buffer (pH 7.2), likely due to
the poor solubility of 6 in PBS. Therefore, we performed the
reaction in the PBS/tBuOH (5:1), which has been used as a
biocompatible solvent system for protein modification.5 The
reaction was kept at different temperatures (4, 24, and 37 °C) for
overnight. Analytical HPLC and MS analysis show that 4 was
more efficiently converted to the desired product at 37 °C with
an estimated yield of 90% than at 4 °C (15%) and 24 °C (65%)
(Figures 4A and 3B), indicating that increasing the solubility of 6
can help improve the conjugation efficiency.
mide condensation system is a novel bioorthogonal conjugation
reaction for site-specific modification of proteins in their native
states.
Specific labeling of live cell surfaces has proven valuable to
manipulate cell fate and functions for basic and translational
research with numerous applications in biotechnology and
medicine.8 Thus, we decided to demonstrate such a utility of our
method for cell surface labeling and test how efficient it is. The
aldehyde group was easily generated on the cell surface (Hela
cell line) via the widely used NaIO4 oxidation of the sialic acid
residues in cell-surface glycans.9 Two fluorescent probes 8 and 9
were prepared (Figure 2). Fluorescent probe 8 harbors the 2-
aminoethanethioamide at the N-terminus of the peptide, while
the control probe 9 does not (Figure S12−S13). In the labeling
experiment, 20 μM 8 or 9 was added to the NaIO4-treated and
nontreated cells, respectively, and the cells were then incubated
at 37 °C for 30 min. The probes were removed by washing the
cells with PBS three times, and the treated cells were analyzed
using confocal microscopy. Results show that only in the NaIO4-
oxidized cells treated with 8 was significant fluorescence
observed on the rim of their membranes (Figure 5A). NaIO4-
Figure 4. Characterization of the conjugation of 6 to ubiquitin-
aldehyde 4 and myoglobin-aldehyde 7. (A) HPLC analysis of the
reaction between 6 (0.25 mM) and 4 (50 μM) at 4 °C, 24 and 37 °C for
overnight, respectively. Peaks a′ and a′′ are 6, peak b is 4, peak c is the
conjugated product. (B) Mass spectrometry analysis showing the
conjugated product (calcd mass: 9836.3, found mass: 9837.0). (C)
Structure of the myoglobin-aldehyde 7. (D) Coomassie blue stained
SDS-PAGE gel and fluorescent imaging of 7 (15 μM) before and after
reacting with 6 (0.75 mM) for overnight at 37 °C. Fluorescent imaging
was taken under UV 365 nm.
Figure 5. Labeling of cell surface using 2-aminoethanethioamide and
aldehyde condensation reaction. (A) Confocal microscope images of
cells labeled by probe 8 and 9 at 37 °C for 30 min. Nontreated cell by
NaIO4 was used as the control. (B) Confocal microscope images of cells
labeled by probe 8 and 9 at 37 °C during 30 s to 10 min. Nucleus was
stained with DAPI.
Myoglobin was also used as another model protein to further
test this conjugation reaction. PLP (pyridoxal 5′-phosphate)-
mediated transamination was used to generate the unique α-oxo
aldehyde at the N-terminus of the protein.6 Consistent with the
previous reports,5,6 the aldehyde can be quantitatively
introduced into the myoglobin (Figures 4C and S4). Incubation
of myoglobin-aldehyde 7 (15 μM) with 1 (0.75 mM, 50 equiv)
for overnight afforded the conjugate in more than 95% yield at
all tested temperatures (4, 24, or 37 °C) by LC-MS analysis
(Figure S5). Compound 5 was also efficiently conjugated with 7
in the same conditions with an estimated yield of 60% at 4 °C,
70% at 24 °C, and 90% at 37 °C, respectively (Figure S6). While
6 was also conjugated with 7, the overall yield was 40−45% even
at 37 °C after overnight reaction. No significant product was
observed at 4 and 24 °C, again due to poor solubility of 6
(Figures S7). Furthermore, we also demonstrated specific
labeling of the phage-surface protein p8 after PLP-mediated
transamination,7 which introduced a ketone group, as confirmed
by analytical HPLC, ESI-MS, and fluorescent gel analysis
(Figures S8−S11). However, the labeling was significantly less
efficient, likely because the ketone group in the α-oxopropana-
mide is much less reactive than the aldehyde in α-oxoacetamide.
All these data indicate that the aldehyde−aminoethanethioa-
oxidized cells treated with 9 or nonoxidized cells did not
generate any significant fluorescent signal (Figure 5A). We also
performed the time course labeling of the cells from 30 s to 10
min. As seen from Figure 5B, the labeling of cells could be
observed at as early as 5 min. These results further confirm the
efficient nature of the aldehyde−aminoethane thioamide
condensation reaction, which can be used to label living cell
surfaces under mild conditions. Efficient cell labeling was also
confirmed by the fluorescence-activated cell sorting (FACS)
analysis (Figure S14). Fast labeling is usually required to study
certain cellular events such as dynamic membrane trafficking or
for specific receptor tracking. In our study, after 30 min of
labeling, the cells were returned to culture medium at 37 °C for 1
and 2 h. Then the cells were fixed and subjected to confocal
microscopy. It was shown that after 1 h, some portions of the
labeled membrane were found inside the cells, presumably via
internalization, which revealed a dynamic process of membrane
C
DOI: 10.1021/acs.orglett.8b03195
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
component trafficking and turnover (Figure S15). Previously, an
oxime ligation-based method used 10 mM aniline as the catalyst
and a higher concentration (100 μM) of the labeling probe and
required an incubation time of 90 min to achieve efficient cell
labeling.9 Our method does not need a catalyst and requires a
lower concentration (20 μM) of the probes and shorter labeling
time (30 min). These are desirable features of a biocompatible
method to label, monitor, and track the molecules on the cell
membranes.
REFERENCES
■
(1) (a) Boutureira, O.; Bernardes, G. J. L. Chem. Rev. 2015, 115, 2174.
(b) Tung, C. L.; Wong, C. T. T.; Fung, E. Y. M.; Li, X. C. Org. Lett.
2016, 18, 2600. (c) Yu, J.; Yang, X.; Sun, Y.; Yin, Z. Angew. Chem., Int.
Ed. 2018, 57, 11598.
(2) (a) Zhang, G.; Zheng, S. Q.; Liu, H. P.; Chen, P. R. Chem. Soc. Rev.
2015, 44, 3405. (b) Lang, K.; Chin, J. W. Chem. Rev. 2014, 114, 4764.
(c) Spicer, C. D.; Davis, B. G. Nat. Commun. 2014, 5, 4740.
(d) Stephanopoulos, N.; Francis, M. B. Nat. Chem. Biol. 2011, 7, 876.
(e) Addy, P. S.; Erickson, S. B.; Italia, J. S.; Chatterjee, A. J. Am. Chem.
Soc. 2017, 139, 11670. (f) Yu, J.; Shen, D.; Zhang, H.; Yin, Z.
Bioconjugate Chem. 2018, 29, 1016. (g) Wang, P.; Na, Z.; Fu, J.; Tan, C.
Y. J.; Zhang, H.; Yao, S. Q.; Sun, H. Y. Chem. Commun. 2014, 50, 11818.
(h) Yang, R. L.; Bi, X. B.; Li, F. P.; Cao, Y.; Liu, C. F. Chem. Commun.
2014, 50, 7971. (i) Bi, X. B.; Yang, R. L.; Feng, X. Y.; Rhodes, D.; Liu, C.
F. Org. Biomol. Chem. 2016, 14, 835. (j) Bi, X. B.; Yin, J.; Hemu, X. Y.;
Rao, C.; Tam, J. P.; Liu, C. F. Bioconjugate Chem. 2018, 29, 2170.
(3) (a) Dirksen, A.; Dawson, P. E. Bioconjugate Chem. 2008, 19, 2543.
(b) Kiick, K. L.; Saxon, E.; Tirrell, D. A.; Bertozzi, C. R. Proc. Natl. Acad.
Sci. U. S. A. 2002, 99, 19. (c) Rostovtsev, V. V.; Green, L. G.; Fokin, V.
V.; Sharpless, K. B. Angew. Chem., Int. Ed. 2002, 41, 2596. (d) Meldal,
M.; Tornøe, C. W. Chem. Rev. 2008, 108, 2952. (e) Agarwal, P.; van der
Weijden, J.; Sletten, E. M.; Rabuka, D.; Bertozzi, C. R. Proc. Natl. Acad.
Sci. U. S. A. 2013, 110, 46. (f) Jewett, J. C.; Bertozzi, C. R. Chem. Soc.
Rev. 2010, 39, 1272. (g) Oliveira, B. L.; Guo, Z.; Bernardes, G. J. L.
Chem. Soc. Rev. 2017, 46, 4895. (h) Lang, K.; Davis, L.; Torres-Kolbus,
J.; Chou, C.; Deiters, A.; Chin, J. W. Nat. Chem. 2012, 4, 298. (i) Wang,
P.; Zhang, C. J.; Chen, G. C.; Na, Z.; Yao, S. Q.; Sun, H. Y. Chem.
Commun. 2013, 49, 8644.
(4) (a) Schubert, M. P. J. Biol. Chem. 1936, 114, 341. (b) Ratner, S.;
Clarke, T. J. Am. Chem. Soc. 1937, 59, 200. (c) Liu, C.-F.; Tam, J. P. J.
Am. Chem. Soc. 1994, 116, 4150.
(5) (a) Alam, J.; Keller, T. H.; Loh, T. P. J. Am. Chem. Soc. 2010, 132,
9546. (b) Wang, P.; Zhang, S.; Meng, Q.; Liu, Y.; Shang, L.; Yin, Z. Org.
Lett. 2015, 17, 1361.
(6) Gilmore, J. M.; Scheck, R. A.; Esser-Kahn, A. P.; Joshi, N. S.;
Francis, M. B. Angew. Chem., Int. Ed. 2006, 45, 5307.
(7) Carrico, Z. M.; Farkas, M. E.; Zhou, Y.; Hsiao, S. C.; Marks, J. D.;
Chokhawala, H.; Clark, D. S.; Francis, M. B. ACS Nano 2012, 6, 6675.
In summary, we have shown that the condensation between 2-
aminoethanethioamide and an aldehyde is highly chemose-
lective. It was successfully utilized to site-selectively label
peptides, proteins, and living cells. The reaction proceeds
smoothly in an aqueous buffer under mild conditions, uses very
low reactant concentrations, and does not require any catalysts
or reducing agents that might be toxic to living systems. The
thiazolidin-5-imine linkage so-formed is stable at physiological
and weakly basic pH, while hydrolyzable in acidic media (e.g.,
pH 4.5). Such a property of pH-dependent stability and
reversibility would be useful for drug delivery purpose.10 It is
noteworthy that a related thiazolidine ring formed between the
1,2-aminothiol moiety of an N-terminal Cys residue and an
aldehyde is not very stable under physiological pH but rather
stable in slightly acidic pH.11 As another favorable characteristic
of this new method, the 2-aminoethanethioamide moiety can be
easily installed in any compounds and is very stable in air. On the
contrary, the 1,2-aminothiol moiety required for thiazolidine
formation is very prone to oxidation in which the thiol is
oxidized to disulfide.12 Aldehyde is one of the most extensively
utilized bioconjugation functionalities, and many chemical and
enzymatic methods are available to introduce an aldehyde group
into proteins at either a terminal or internal site.6,13 Therefore,
we anticipate that this new condensation reaction will be an
attractive alternative to the currently used bioconjugation
techniques for a wide range of applications in the future.
́
(8) (a) Custodio, C. A.; Mano, J. F. ChemNanoMat 2016, 2, 376.
ASSOCIATED CONTENT
■
(b) Bi, X. B.; Yin, J.; Chen, G. A.; Liu, C.-F. Chem. - Eur. J. 2018, 24, 1.
(c) Bi, X. B.; Yin, J.; Nguyen, G. K. T.; Rao, C.; Halim, N. B. A.; Hemu,
X. Y.; Tam, J. P.; Liu, C.-F. Angew. Chem., Int. Ed. 2017, 56, 7822.
(9) Zeng, Y.; Ramya, T. N. C.; Dirksen, A.; Dawson, P. E.; Paulson, J.
C. Nat. Methods 2009, 6, 207.
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.orglett.8b03195.
Experimental procedures and characterization data
(PDF)
(10) Casi, G.; Huguenin-Dezot, N.; Zuberbu
Neri, D. J. Am. Chem. Soc. 2012, 134, 5887.
̈
hler, K.; Scheuermann, J.;
(11) (a) Zhang, L.; Tam, J. P. Anal. Biochem. 1996, 233, 87. (b) Bi, X.
B.; Pasunooti, K. K.; Tareq, A. H.; Takyi-Williams, J.; Liu, C.-F. Org.
Biomol. Chem. 2016, 14, 5282.
AUTHOR INFORMATION
■
(12) Bermejo-Velasco, D.; Nawale, G. N.; Oommen, O. P.; Hilborn,
J.; Varghese, O. P. Chem. Commun. 2018, 54, 12507.
Corresponding Author
*E-mail: cfliu@ntu.edu.sg
ORCID
(13) (a) Wu, P.; Shui, W.; Carlson, B. L.; Hu, N.; Rabuka, D.; Lee, J.;
Bertozzi, C. R. Proc. Natl. Acad. Sci. U. S. A. 2009, 106, 3000.
(b) Rabuka, D.; Rush, J. S.; deHart, G. W.; Wu, P.; Bertozzi, C. R. Nat.
Protoc. 2012, 7, 1052. (c) Tuley, A.; Lee, Y. J.; Wu, B.; Wang, Z. Y.; Liu,
W. S. Chem. Commun. 2014, 50, 7424. (d) Bi, X. B.; Pasunooti, K. K.;
Lescar, J.; Liu, C. F. Bioconjugate Chem. 2017, 28, 325. (e) El-Mahdi, O.;
Melnyk, O. Bioconjugate Chem. 2013, 24, 735. (f) Spears, R. J.;
Brabham, R. L.; Budhadev, D.; Keenan, T.; McKenna, S.; Walton, J.;
Brannigan, J. A.; Brzozowski, A. M.; Wilkinson, A. J.; Plevin, M.;
Fascione, M. A. Chem. Sci. 2018, 9, 5585. (g) Howard, T. S.; Cohen, R.
D.; Nwajiobi, O.; Muneeswaran, Z. P.; Sim, Y. E.; Lahankar, N. N.; Yeh,
J. T.; Raj, M. Org. Lett. 2018, 20, 5344.
Peter C. Dedon: 0000-0003-0011-3067
Chuan-Fa Liu: 0000-0001-7433-2081
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This research is supported by A*STAR (ETPL-QP-19-06) and
the Ministry of Education (NGF-2017-03-040; MOE 2016-T3-
1-003) of Singapore and by the Singapore National Research
Foundation under its Antimicrobial Resistance IRG adminis-
tered by the Singapore-MIT Alliance for Research and
Technology.
D
DOI: 10.1021/acs.orglett.8b03195
Org. Lett. XXXX, XXX, XXX−XXX