Hypersensitive azobenzenes: Facile synthesis of clickable and cleavable azo linkers with tunable and high reducibility

Article abstract of DOI:10.1039/c9ob02515d

The aim of this work is to show that by increasing the number of donor substituents in a donor/acceptor system, the sensitivity of the azobenzene linkage towards a reductive cleavage reaction can be enhanced to unprecedented high levels. For instance, in a triple-donor system, less than a second constitutes the half-life of the azo (NN) bond. Synthetic access to such redox active scaffolds is highly practical and requires only 1-2 synthetic steps. The fundamental molecular design is also adaptable. This is demonstrated through scaffold functionalization by azide, tetraethylene glycol, and biotin groups. The availability of the azide group is shown in a copper-free 'click' reaction suitable in context with protein conjugation and proteomics application. Finally, the clean nature of the scission process is demonstrated with the help of liquid chromatography coupled with mass analysis. This work, therefore, describes development of cleavable azobenzene linkers that can be accessed with synthetic ease, can be multiply functionalized, and show a clean and rapid response to mild reducing conditions.

Full text of DOI:10.1039/c9ob02515d

Organic &
Biomolecular Chemistry
View Article Online
View Journal
COMMUNICATION
Hypersensitive azobenzenes: facile synthesis of
clickable and cleavable azo linkers with tunable
and high reducibility†
Cite this: DOI: 10.1039/c9ob02515d
Received 23rd November 2019,
Accepted 23rd December 2019
Taejun Eom
and Anzar Khan
*
DOI: 10.1039/c9ob02515d
rsc.li/obc
The aim of this work is to show that by increasing the number of an efficient and clean manner so that the protein remains in
donor substituents in a donor/acceptor system, the sensitivity of its native state, released quantitatively, and uncontaminated
the azobenzene linkage towards a reductive cleavage reaction can with byproducts to aid the final analysis.
be enhanced to unprecedented high levels. For instance, in a
In 2007, Bogyo and coworkers established the utility of an
cleavable linker in chemical
triple-donor system, less than a second constitutes the half-life of azobenzene group as
a
the azo (NvN) bond. Synthetic access to such redox active proteomics.6,7 They used sodium dithionite (25 mM) as the
scaffolds is highly practical and requires only 1–2 synthetic steps. reducing reagent to cleave the azo bond through 3 wash cycles
The fundamental molecular design is also adaptable. This is of 15 minutes each. The cleavage efficiency was reported to be
demonstrated through scaffold functionalization by azide, tetra- 90%. This pioneering study was critical in establishing the
ethylene glycol, and biotin groups. The availability of the azide design concept of a cleavable linker applicable for chemical
group is shown in a copper-free ‘click’ reaction suitable in context proteomics based on the reductive cleavage of the azobenzene
with protein conjugation and proteomics application. Finally, the scaffold. However, the need of high concentrations of the redu-
clean nature of the scission process is demonstrated with the help cing agent, long reaction times, and non-quantitative release
of liquid chromatography coupled with mass analysis. This work, efficiency represent drawbacks of the system. Such attributes
therefore, describes development of cleavable azobenzene linkers
that can be accessed with synthetic ease, can be multiply functio-
nalized, and show a clean and rapid response to mild reducing
conditions.
In chemical proteomics, one of the major goals is to develop
strategies to capture and enrich a specific protein from a
complex mixture of a cell lysate.1,2 One approach to achieve
this goal is through selective conjugation of the protein onto a
solid support, separation of the solid support from rest of the
cell lysate, and release of the captured protein for analytical
purposes. The conjugation and release processes often involve
the formation and disruption of biotin–streptavidin inter-
actions. These interactions are, however, one of the strongest
non-covalent interactions found in nature. Therefore, while
protein-capture benefits from this trait, harsh conditions are
required for the release step. To avoid this, a linker is installed
between the biotin and the protein-capture sites (Fig. 1).3–5
This linker should ideally be cleaved under mild conditions in
Department of Chemical and Biological Engineering, Korea University, 145 Anam-Ro,
Seongbuk-Gu, Seoul 02841, Korea. E-mail: anzar@korea.ac.kr
†Electronic supplementary information (ESI) available: Synthesis and character-
Fig. 1 Schematic representation of the protein capture and release
ization details. See DOI: 10.1039/c9ob02515d
through use of an azobenzene linker.
This journal is © The Royal Society of Chemistry 2020
Org. Biomol. Chem.

View Article Online
Communication
Organic & Biomolecular Chemistry
may compromise the native protein structure, result into which serve as the control DA system. The synthesis in this case
strong background signals from excess reagent, and limit the is a single-step diazotization reaction between commercially
overall efficacy of the system.
available ethylanilinoethanol and 4-amino methylbenzoate.17
2
To overcome these issues, Wagner and coworkers designed and 3 carry one and two methoxy substituents and represent
an azobenzene linker with much higher sensitivity by substi- two-donor/one acceptor (DDA) and three-donor/one acceptor
tuting the aromatic nucleus with oxygen-based donors and an (DDDA) systems, respectively. In 2 and 3, the copper-catalyzed
acid-based acceptor.8,9 The synthesis of the active scaffold was amination provided the alkyl aniline segments with donor sub-
accomplished in 15 steps. With this new molecular design, the stituents that could be joined by the acceptor fragment through
dithionite concentration can be decreased to 6 mM and less the diazotization reaction. Typically the isolated yields in these
than 10 seconds is required for full cleavage. This design rep- synthetic transformations ranged from 78–91%.
resents the current state-of-the-art in terms of sensitivity in
cleavable azobenzene linkers.
Having efficient access to the three compounds, the NvN
bond scission reaction was studied with the help of UV-Vis
Besides proteomics, the azobenzene cleavage reaction finds spectroscopy. As shown in Fig. 2 and Fig. S19–S21,† the π–π*
applicability in various other disciplines10 ranging from absorption band of the azobenzene chromophore located at
hypoxia sensing11 and drug delivery12 to self-assembly13 and 400–550 nm gradually decreases in intensity as a function of
disassembly of synthetic constructs.14 In most of these appli- exposure time to sodium dithionite. Based on this data
cations, highly sensitive azobenzene scaffolds are required. For (Fig. S22†), it became clear that an increase in the number of
instance, in hypoxia sensing, high sensitivity translates to the donor groups enhances the reducibility of the azo bond
reliable detection under mild conditions.11 In colon-specific (Table 1). 2 and 3 were found to be more sensitive than 1. The
drug delivery, a rapid response under reducing conditions most sensitive system, compound 3, cleaves quantitatively even
means higher efficacy of the delivery vehicle.15
at a 0.5 mM dithionite concentration. In comparison to the
The design and synthesis of highly sensitive azobenzene state-of-the-art,8,9 compound 3 has similar half-life (<1 s) and
linkers, therefore, is a valuable goal with implications in total cleavage time (<5 s). However, compound 3 achieves this
various research fields. Toward this end, with the help of three with one-twelfth of the concentration of the reducing agent. A
molecular designs, we show that an increase in the number of direct comparison would require operating at a 6 mM dithio-
donor substituents in an amine/ester-based donor/acceptor nite concentration. However, 2 and 3 fully cleave instantly over
system can systematically enhance the sensitivity of the azo 2 mM concentration of the reducing agent without allowing
(NvN) bond towards a reductive cleavage reaction. To our for accumulation of reaction kinetics data to be compared.
knowledge, the sensitivities described here are the highest
Having confirmed superior sensitivities of 2 and 3, adap-
reported thus far. In addition, the synthesis is concise and tation of the most potent molecular design was considered for
efficient and allows for tailored molecular-level adaptations in applications. For instance, if the fundamental design concept
order to meet structural requirements of a certain application.
Our previous work on the development of polymeric drug
delivery and imaging vehicles indicated that amine- and ester-
based donor/acceptor (DA) azobenzenes were more sensitive to
the reductive cleavage reaction than are donor/donor (DD),
acceptor alone (A), and non-substituted systems.16 Encouraged
by these results, we explored easily accessible synthetic
avenues that would allow further increase in the electron
density of the azobenzene nucleus. This is achieved by placing
electron-donating methoxy-substituent(s) in ortho-position to
the azo bond (Scheme 1). Initially, three molecules were pre-
pared (Fig. S1–S18†). 1 having dialkyl amine and ester groups
of compound 3 is to be translated to chemical proteomics
Fig. 2 UV-Vis spectra of compounds 1–3 (50 μm) upon exposure to
1 mM sodium dithionite in 2 : 8 DMSO : Tris-HCl buffer solution. The
time lapse between each trace is 1 second. The digital picture shows
change in color of the azo compound 3 (left) upon reduction (right). The
azobenzene scission reaction is shown with the help of chemical
structures.
Scheme 1 Synthesis of 1, 2, and 3.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Organic & Biomolecular Chemistry
Communication
Table 1 Details of the azo cleavage reaction in 1–3
(7). These functional group decorations do not alter the funda-
mental electronic structure of the scaffold 3 and therefore the
molecule retains its sensitivity characteristics. The isolated
yields in these synthetic transformations range from 78–95%.
Overall, compound 7 is decorated with biotin, ethylene glycol,
and azide functionalities to meet immobilization and capture
requirements under aqueous conditions, has a DDD/A type
electronic structure to meet release requirements, and requires
only five simple and high-yielding synthetic steps.
Liquid chromatography coupled with mass spectrometry
(LC-MS) was then utilized to examine the cleanliness of the
cleavage process. Initially, compound 7 was subjected to the
analysis (Fig. 3 and Fig. S23†) and the elution chromatograms
indicated a clean cleavage to two expected compounds and
absence of byproducts. The corresponding mass analyses
showed molecular ions of the cleaved aniline segments. No
signals were observed at a molecular weight range higher than
that of the expected molecular ion. This indicates a clean clea-
vage to expected products and absence of byproducts.
Encouraged by these results, compound 7 was subjected to
a copper-free ‘click’ reaction with a cyanine dye having a cyclic
acetylene to probe the availability and reactivity of the azide
group (Scheme 3).18 The azide–alkyne reaction in this case,
therefore, requires only mixing of the two reactants. IR spec-
troscopy indicated disappearance of the azide signal at
2099 cm−1 upon reaction (Fig. S24†). In UV-Vis spectroscopy,
the dye-azo conjugate displays signals for three chromophores
(Fig. 4). The stilbene-like triazole that results from ‘click’ con-
jugation can be observed at 350 nm. The azo and cyanine seg-
ments can be seen at 400–550 and 600–750 nm, respectively.
As expected, after cleavage, the azo absorption band dis-
appears whereas the stilbene and cyanine bands remain.
Compound 8 was also subjected to the LC-MS analysis
(Fig. S25†), which again demonstrated a clean cleavage process
due to a high sensitivity of the azobenzene scaffold and the
need to use only a very dilute solution of the reducing agent.
In summary, facile synthetic access to adaptable azo-
benzene scaffolds is developed that shows a quantitative and
[Na2S2O4]
Compound
Half-life (s)
2 mM
2 mM
2 mM
1 mM
1 mM
1 mM
0.75 mM
0.75 mM
0.75 mM
0.5 mM
0.5 mM
0.5 mM
1
2
3
1
2
3
1
2
3
1
2
3
15
0.7
0.1
35
1.4
0.1
—
2.2
0.3
—
—
0.6
—: The cleavage reaction does not proceed to completion.
applications, it must carry protein capture and immobilization
sites. Typically, to aid water solubility, a hydrophilic segment
is added to the scaffold as well. In keeping with these estab-
lished design traits, the hydroxyl group present at the alkyl
segment of the amine was transformed into an azide (4) so
that the known ‘click’ chemistry protocols can be employed for
the capture process (Scheme 2).18 Separately, the acid group in
aminobenzoic acid was used to incorporate a tetraethylene
glycol unit. A diazotization reaction between electron-rich 4
and electron-poor 5 afforded the azobenzene 6. Finally, the
terminal hydroxyl group of the ethylene glycol segment could
be engaged in an esterification reaction to attach biotin ligand
Scheme 2 Total synthesis of azide, biotin, and tetraethylene glycol-
functionalized azo 7.
Fig. 3 Liquid chromatography (top) and corresponding mass data
(bottom) for compound 7 before and after azo cleavage reaction.
This journal is © The Royal Society of Chemistry 2020
Org. Biomol. Chem.

View Article Online
Communication
Organic & Biomolecular Chemistry
Notes and references
1 G. C. Rudolf, W. Heydenreuter and S. A. Sieber, Curr. Opin.
Chem. Biol., 2013, 17, 110–117.
2 A. M. Sadaghiani, S. H. L. Verhelst and M. Bogyo, Curr.
Opin. Chem. Biol., 2007, 11, 20–28.
3 G. Leriche, L. Chisholm and A. Wagner, Bioorg. Med.
Chem., 2012, 20, 571–582.
4 R. Bielski and Z. Witczak, Chem. Rev., 2013, 113, 2205–2243.
5 Y. Yang, M. Fonović and S. H. Verhelst, in Activity-Based
Proteomics, Springer, 2017, pp. 185–203.
6 S. H. L. Verhelst, M. Fonovic and M. Bogyo, Angew. Chem.,
Int. Ed., 2007, 46, 1284–1286.
7 M. Fonovic, S. H. L. Verhelst, M. T. Sorum and M. Bogyo,
Mol. Cell. Proteomics, 2007, 6, 1761–1770.
Scheme 3 Strain-promoted copper-free ‘click’ reaction to examine
availability of the azide group in 7 towards conjugation. 8a and 8b are
positional isomers.
8 G. Leriche, G. Budin, L. Brino and A. Wagner, Eur. J. Org.
Chem., 2010, 4360–4364.
9 G. Budin, M. Moune-Dimala, G. Leriche, J. M. Saliou,
J. Papillon, S. Sanglier-Cianferani, A. Van Dorsselaer,
V. Lamour, L. Brino and A. Wagner, ChemBioChem, 2010,
11, 2359–2361.
10 For cleavable linkers based on azobenzene motif with
chemical biology applications, see: (a) J. Szychowski,
A. Mahdavi, J. J. L. Hodas, J. D. Bagert, J. T. Ngo,
P. Landgraf, D. C. Dieterich, E. M. Schuman and
D. A. Tirrell, J. Am. Chem. Soc., 2010, 132, 18351–18360;
(b) Y. Y. Yang, J. M. Ascano and H. C. Hang, J. Am. Chem.
Soc., 2010, 132, 3640–3641; (c) Y. Y. Yang, M. Grammel,
A. S. Raghavan, G. Charron and H. C. Hang, Chem. Biol.,
2010, 17, 1212–1222; (d) F. Landi, C. M. Johansson,
D. J. Campopiano and A. N. Hulme, Org. Biomol. Chem.,
2010, 8, 56–59; (e) K. S. Yang, G. Budin, C. Tassa, O. Kister
and R. Weissleder, Angew. Chem., Int. Ed., 2013, 52, 10593–
10597; (f) Y. L. Yang and S. H. L. Verhelst, Chem. Commun.,
2013, 49, 5366–5368; (g) J. H. L. Claessen, M. D. Witte,
N. C. Yoder, A. Y. Zhu, E. Spooner and H. L. Ploegh,
ChemBioChem, 2013, 14, 343–352; (h) Y. Qian, J. Martell,
N. J. Pace, T. E. Ballard, D. S. Johnson and E. Weerapana,
ChemBioChem, 2013, 14, 1410–1414; (i) J. A. L. Choo,
S. Y. Thong, J. Yap, W. J. E. van Esch, M. Raida, R. Meijers,
J. Lescar, S. H. L. Verhelst and G. M. Grotenbreg, Angew.
Chem., Int. Ed., 2014, 53, 13390–13394; ( j) C. Sibbersen,
L. Lykke, N. Gregersen, K. A. Jorgensen and M. Johannsen,
Chem. Commun., 2014, 50, 12098–12100; (k) R. L. Kwant,
J. Jaffe, P. J. Palmere and M. B. Francis, Chem. Sci., 2015, 6,
2596–2601; (l) L. J. Tan, Y. Z. Liu, Q. L. Yang, X. W. Li,
X. Y. Wu, B. Gong, Y. M. Shen and Z. F. Shao, Chem.
Commun., 2016, 52, 954–957; (m) S. S. Liew, S. B. Du,
J. Y. Ge, S. J. Pan, S. Y. Jang, J. S. Lee and S. Q. Yao, Chem.
Commun., 2017, 53, 13332–13335; (n) P. S. Addy,
S. B. Erickson, J. S. Italia and A. Chatterjee, J. Am. Chem.
Soc., 2017, 139, 11670–11673.
Fig. 4 UV-Vis spectra of cyanine alkyne (solid line) and 8 before (dash
dot line) and after (dashed line) azo bond scission in 2 : 8 DMSO : Tris-
HCl buffer solution.
clean cleavage of the NvN bond in a few seconds of exposure
to low concentrations of a reducing agent. We anticipate that
the results presented here would be useful not only for the
protein conjugation applications but also in the arena of
hypoxia sensing,11 drug delivery,12 self-assembly,13 and self-
immolative polymers,14 in which the triggered cleavage of the
azobenzene nucleus is applied for release/activation purposes
and a rapid and tunable response time is required.
Conflicts of interest
There are no conflicts to declare.
11 (a) R. B. P. Elmes, Chem. Commun., 2016, 52, 8935–8956;
(b) T. Thambi, J. H. Park and D. S. Lee, Chem. Commun.,
2016, 52, 8492–8500; (c) A. Chevalier, P. Y. Renard and
A. Romieu, Chem. – Asian J., 2017, 12, 2008–2028.
Acknowledgements
We thank National Research Foundation of Korea grant funded
by the Korean government (MSIP) (NRF-18R1D1A1B07048527).
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Organic & Biomolecular Chemistry
Communication
12 (a) S. H. Lee, E. Moroz, B. Castagner and J. C. Leroux, J. Am.
Chem. Soc., 2014, 136, 12868–12871; (b) S. H. Medina,
M. V. Chevliakov, G. Tiruchinapally, Y. Y. Durmaz,
S. P. Kuruvilla and M. E. H. ElSayed, Biomaterials, 2013, 34,
4655–4666; (c) S. P. Kuruvilla, G. Tiruchinapally, M. ElAzzouny
and M. E. H. ElSayed, Adv. Healthcare Mater., 2017, 6, 15;
E. R. Gillies, Polym. Chem., 2016, 7, 1871–1881; (c) B. Fan,
J. F. Trant and E. R. Gillies, Macromolecules, 2016, 49,
9309–9319; (d) J. T. Offenloch, J. Willenbacher,
P. Tzvetkova, C. Heiler, H. Mutlu and C. Barner-Kowollik,
Chem. Commun., 2017, 53, 775–778; (e) H. Mutlu and
C. Barner-Kowollik, Polym. Chem., 2016, 7, 2272–2279.
(d) T. Eom, W. Yoo, Y. D. Lee, J. H. Park, Y. Choe, J. Bang, 15 G. V. d. Mooter, B. Maris, C. Samyn, P. Augustijns and
S. Kim and A. Khan, J. Mater. Chem. B, 2017, 5, 4574–4578. R. Kinget, J. Pharm. Sci., 1997, 86, 1321.
13 (a) J. Rao and A. Khan, J. Am. Chem. Soc., 2013, 135, 14056– 16 T. Eom, W. Yoo, S. Kim and A. Khan, Biomaterials, 2018,
14059; (b) J. Rao, C. Hottinger and A. Khan, J. Am. Chem. 185, 333–347.
Soc., 2014, 136, 5872–5875; (c) J. Rao and A. Khan, Polym. 17 For an excellent review article on synthesis of azobenzenes,
Chem., 2015, 6, 686–690. see: E. Merino, Chem. Soc. Rev., 2011, 40, 3835–3853.
14 (a) A. D. Wong, T. M. Gungor and E. R. Gillies, ACS Macro 18 J. C. Jewett, E. M. Sletten and C. R. Bertozzi, J. Am. Chem.
Lett., 2014, 3, 1191–1195; (b) A. D. Wong, A. L. Prinzen and
Soc., 2010, 132, 3688–3690.
This journal is © The Royal Society of Chemistry 2020
Org. Biomol. Chem.