Synthesis and Photophysical Characterization of Azoheteroarenes

Article abstract of DOI:10.1021/acs.orglett.8b02014

A set of azoheteroarenes have been synthesized with Buchwald-Hartwig coupling and microwave-assisted O₂ oxidation as the key steps. Several compounds exhibit good to excellent photoswitching properties (high switching efficiency, good fatigue r

Full text of DOI:10.1021/acs.orglett.8b02014

Letter
pubs.acs.org/OrgLett
Cite This: Org. Lett. XXXX, XXX, XXX−XXX
Synthesis and Photophysical Characterization of Azoheteroarenes
†
†
,‡
Yongjin Xu, Chunxia Gao, Joakim Andreasson,* and Morten Grøtli*,†
́
†Department of Chemistry and Molecular Biology, University of Gothenburg, SE-412 96 Gothenburg, Sweden
‡Department of Chemistry and Chemical Engineering, Physical Chemistry, Chalmers University of Technology, SE-412 96
Gothenburg, Sweden
S
* Supporting Information
ABSTRACT: A set of azoheteroarenes have been synthesized with Buchwald−Hartwig
coupling and microwave-assisted O2 oxidation as the key steps. Several compounds
exhibit good to excellent photoswitching properties (high switching efficiency, good
fatigue resistance, and thermal stability of Z-isomer) relevant for photocontrolled
applications, which pave the way for use in photopharmacology.
Nitrogen heterocycles are among the most significant
hotoswitches have long held the interest of the scientific
Pcommunity for their ability to alter molecular structure, structural components of pharmaceuticals.18 Therefore, in
and thus properties, using light as a stimulus.1−5 By
order to integrate the azobenzene-based molecular photo-
switches into the pharmacophore of small drug-like molecules,
e.g., kinase inhibitors, the development of heterocyclic
azobenzene derivatives are fundamental.
introducing photoswitches into bioactive molecules, their
biological activities can be switched on or off upon
photoisomerization between the implicated isomeric forms.6,7
In this way, photocontrolled regulators of ion channels,8
enzymes,9 receptors,10 as well as living organisms11 have been
developed. Such photoresponsive molecules are powerful tools
for studying living systems as they are noninvasive and offer a
high spatiotemporal resolution.11−14 Azobenzene-based mo-
lecular photoswitches (Scheme 1a) have been extensively used
in such developments.15 Potential future applications of this
class of photoswitches include also optically controlled (bio-)
materials and information/data processing to mention a few.16
Of particular interest for the latter are compounds displaying
very fast thermal Z → E isomerization, allowing for ultrafast
data transmission.17
Despite this fact, only a small portion of all heterocycles have
been used in photoswitchable heterocyclic azoswitches. Some
heterocyclic azo compounds have been well-studied,19−29
while others have been reported without mentioning photo-
physics/photochemistry.30,31 The classic strategies for synthe-
sizing azobenzenes include the Azo-coupling reaction, Mills
reaction, as well as metal=catalyzed coupling accompanied by
oxidation.32 An efficient synthesis of red-shifted azo-benzenes
was recently introduced by the coupling between ortho-
lithiated aromatic precursors and aryldiazonium salts.4 In
regard to heterocyclic azobenzene derivatives, a phenyl-
azoimidazole,19 azoheteroarenes with methylated five-mem-
bered heterocyclic rings,19,28 including imidazoles,21 have been
reported using diazonium couplings.
Scheme 1. Photoswitches: (a) Azobenzene Photoswitches
and (b) Synthetic Strategy of Azoheteroarene Molecules
Herein, we report on the synthesis and the photophysical
characterization of a series of azoheteroarenes, with Buch-
wald−Hartwig coupling or nucleophilic substitution, followed
by O2 oxidation as the key synthetic steps in the preparation of
the target compounds (Scheme 1b).
Pyridine and indole are among the most common aromatic
heterocycles in kinase inhibitors. Therefore, we opted for the
preparation of pyridine and indole-based azaanalogues,
including purine and quinazoline derivatives,33 to study how
the aromatic heterocycles influence the ease of synthesis and
photophysical properties of azoheteroarenes. All the selected
Received: June 27, 2018
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.8b02014
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
coupling reaction. Therefore, we decided to block the function
of the secondary amine with an appropriate protecting group.
The Boc group could be installed on 5-bromoindole efficiently,
but the Boc group was cleaved under the coupling conditions.
Instead, we turned to tosyl, and 1e-2, 1f-2, 1g-2, and 1h-2
could be obtained in good to excellent yields (see Supporting
Information (SI)).37 When 1f-2 and 1g-2 (entries 6 and 7,
Table 1) were treated under standard Buchwald−Hartwig
coupling conditions, 2e−2g were obtained in good yield
(above 80%). However, the tosyl group on 1h-2 was unstable
under these conditions. Compound 1h was therefore protected
with THP38 to give 1h-3 (entry 8, Table 1) and then subjected
to the coupling condition to generate 2h in 42% isolated yield
(data not shown). When XPhos was employed as the ligand
instead of t-Bu3P·HBF4, the reaction completed and provided
coupling product 3h with a yield of 54%.
Table 1. Synthesis of Azoheteroarenes by Buchwald−
Hartwig Coupling and Oxidation
Inspired by work from the groups of Szymanski and
Feringa,22 we aimed at forming the azo bond with a “one-
pot, two-steps strategy”, more specifically, removing the Boc
group by microwave-assisted heating (above 150 °C)39 and
then oxidizing the hydrazine to give azo products with O2.
Several rounds of optimization resulted in a protocol giving
generally excellent yields (Table 1).
40
The tosyl group on 3e−3g could be removed by Cs2CO3
to give the corresponding heterocyclic azoarenes (5e−5g; see
SI) with a yield above 80%. The THP group on 3h was
deprotected by 4 M HCl in dioxane39 to provide 5h in 74%
yield.
Applying our developed Pd-mediated coupling strategy to 6-
Cl-9H-purine (1i) and 4-Cl-quinazoline (1j) gave no
detectable coupling products (see SI). Therefore, we turned
to take nucleophilic substitution as the key step.41−44
Compound 1i in n-BuOH, under an N2 atmosphere, was
treated with phenylhydrazine in the presence of DIPEA, using
microwave heating (150 °C, 2 h)22 followed by O2 treatment
to form the azo compound 6 (see SI) in 65% yield. When 1j
was treated with the same condition, no target product was
1
detected by LC−MS or H NMR. However, when changing
from n-BuOH to i-PrOH as solvent42 and heating at a lower
temperature (MW 60 °C, 30 min), followed by the O2
treatment, 8 was obtained in moderate yield (23%; see SI).
The photophysical properties of 3a−8 were then charac-
terized. The experimental results are shown only for 5e (Figure
1a, see SI for “enlarged” versions of all absorption spectra),
whereas the evaluated photophysical parameters for the whole
series are collected in Table 2. Water/DMSO has been used as
solvent. When the heterocyclic azo-switches get integrated
with phramacophores, water solubility could increase, thereby
excluding or minimizing the need of DMSO as cosolvent.
The initial absorption spectra, ascribed to 100% E-isomer,
were recorded after thermal adaption at 50 °C for 72 h. The
absorption maxima of the π → π* transition for the respective
compounds are observed at λ between 317 and 387 nm,
whereas the n → π* transitions at longer wavelengths are very
weak and barely seen in the absorption spectra. There is a clear
trend for the π → π* transition to redshift with increasing size
of the heterocycle. Leaving aside the protected compounds
(3e−3h) the monocyclic compounds display absorption
maxima between 317 and 328 nm, the bicyclic between 325
and 362 nm, and the tricyclic at 387 nm.
a
b
Using MW at 110 °C, the reaction was completed in 1.5 h. Boc
c
deprotection and oxidation combined. XPhos was used as ligand.
d
e
f
MW, 200 °C, 30 min. MW, 180 °C, 1 h, then 200 °C, 8 min; MW,
160 °C, 2 h, then 180 °C, 30 min.
target compounds contain heterocycles that are found in
reported kinase inhibitors (1a−1h; Table 1).
Bromoheterocycles 1a−1d (entries 1−3; Scheme 1) were
initially tested out using Buchwald−Hartwig conditions.34
Both 1a, 1b, and 1d substrates gave the target coupling
products 2a, 2b, and 3d in good yields. However, 1c
underwent significant reduction of the aryl halide by β-
elimination35 during the coupling,36 and only a low yield of 2c
was obtained.
However, when 1e was submitted to this condition, no
product formation was observed. This was probably because
the secondary amine in the heterocycle interfered with the
Introduction of the electron-withdrawing tosyl group results
in a blue shift of the absorption maxima (cf. 3e−3g and 5e−
5g) as well as significant hyperchromism (see absorption
B
DOI: 10.1021/acs.orglett.8b02014
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Table 2. Photophysical Properties
a
b
Wavelengths of the π → π* transition of the E-isomer. λmax in the n → π* region at the photostationary distribution (PSD) after UV irradiation.
Irradiation wavelengths used to enrich the samples in the Z- (longer wavelength light) and E-isomers (shorter wavelength light), respectively.
c
d
Photostationary distribution after photoisomerization to enrich the samples in the Z- and E-isomers, respectively. Tabulated as the fraction of the
e
Z-isomer. Determined in pure DMSO-d6. Time constants for the photoisomerization reactions, normalized to a common light density (365 nm:
f
2.5 mW/cm2, 405 nm: 4.0 mW/cm2, 460 nm: 3.0 mW/cm2). See SI for the original conditions used. Fraction of the sample isomerized in the
thermal Z → E reaction over 24 h. When isomerization is substantial, the corresponding time constant of the reaction is given.
spectra in the SI). Electron donation from a THP group (cf. 3h
and 5h) also results in hyperchromism but has no effect on the
spectral position. The spectral shifts are supported by quantum
mechanical (DFT) calculations (see SI). Subsequent exposure
to UV light resulted in an increased absorption of the n → π*
transition at the expense of the π → π* transition. All processes
proceeded with isosbestic points, which implies clean
conversion between the two implicated isomers. The wave-
lengths of the absorption maxima in the n → π* region are
collected in Table 2, together with the isomerization kinetics
and the photostationary distribution (PSD). Although no
attempts were made toward determining the isomerization
quantum yields, it is seen that the E → Z isomerizations readily
proceed at the modest light intensities used in the experiments
(2−4 mW/cm2). At the PSD, samples 3a−5g are enriched in
the Z-form to between ca. 60% and almost full conversion.
Compounds 5h and 8 showed no detectable isomerization (see
below), while the thermal reversion for 6 was too fast to allow
for the determination of any reliable value.
Note that the PSDs were assessed in pure DMSO (Figure
1b) as opposed to water/DMSO. The values should, however,
be similar for the two solvent systems (Figure S39).
C
DOI: 10.1021/acs.orglett.8b02014
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
detectable Z-isomers is due to thermal isomeriation Z → E it
must be faster than the time resolution of the experiment, i.e., a
few nanoseconds.
The photoinduced Z → E isomerizations were studied by
exposing the samples enriched in the Z-form to light in the
visible region, yielding PSD containing between 65% and 88%
of the E-isomer. The rates at which the isomerizations occur
show efficient isomerization already at moderate light
intensities (1−4 mW/cm2). Except for 3f, no significant
photodecomposition was observed upon photocycling (Figures
S15−S26).
In conclusion, we have synthesized a set of azoheteroarenes
mainly with a strategy using Buchwald−Hartwig coupling and
O2 oxidation as the key steps and commercially available
heterocyclic halogen as the starting substrates. The majority of
the compounds can be enriched in the Z-isomer to more than
90% using light at 365 or 405 nm, whereas the reverse
photoisomerization yields between 65% and 85% E-isomer at
the photostationary state using longer wavelength of light. The
thermal stability of the Z-isomer ranges between minutes to
months, and nearly all compounds have excellent photo-
stability. The relatively easy availability by synthesis together
with favorable photophysical properties make some of these
compounds promising candidates for future application in
photopharmacology, e.g., photoswitchable kinase inhibitors.
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.8b02014.
Experimental procedures and UV−vis and NMR spectra
(PDF)
Figure 1. (a) Absorption spectra of compound 5e at 30 μM
concentration in aqueous solution (5% DMSO). Black line: At
thermal equilibrium. Red line: After exposure to 365 nm UV to yield
the PSD. Magenta line: After exposure to 405 nm light to yield the
PSD. Blue line: After a second round of 365 nm UV to yield the PSD.
Green line: Spectra recorded 24 h (at room temperature in the dark)
subsequent to the second round of 365 nm UV. The inset shows the
absorption changes during 10 cycles of UV and Vis exposure. (b) H
NMR spectra of 5e at the PSD after exposure to 365 nm light and 460
nm light allowing quantification of Z/E ratios.
AUTHOR INFORMATION
Corresponding Authors
■
*E-mail: a-son@chalmers.se.
*E-mail: grotli@chem.gu.se.
ORCID
1
́
Joakim Andreasson: 0000-0003-4695-7943
Morten Grøtli: 0000-0003-3621-4222
Notes
The majority of the compounds displays good thermal
stability, as less than 10% is isomerized back to the E- form
after 24 h in the dark. Exceptions are 3c, 3g, 3h, 5e, and 6
(Table 2). Compound 5g decomposes rapidly, tentatively
ascribed to hydrolysis. With the exception of 6, these are the
compounds with the lowest activation energies for thermal Z
→ E isomerization as estimated by DFT calculations assuming
thermal isomerization through inversion28,45 (see SI). The
discrepancy for 6 is tentatively ascribed to a different
isomerization mechanism, rotation rather than inversion, as 6
together with 8 in the Z-form display the largest twisting angles
between the plane of the heterocycle and the N−N double
bond (see SI).
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by the Swedish Research Council
(VR, grant No. 2015-05642 for M.G. and 2016-0360 for J.A.).
We thank Astrid Nilsen-Moe, Department of Chemistry
̈
Ångstrom Laboratory, Uppsala University, for valuable
assistance with the ns transient absorption measurements.
REFERENCES
■
(1) Molecular Switches; Feringa, B. L., Browne, W. R., Eds.; Wiley-
VCH: Weinheim, 2011.
(2) Dong, M.; Babalhavaeji, A.; Samanta, S.; Beharry, A. A.; Woolley,
G. A. Acc. Chem. Res. 2015, 48, 2662.
(3) Russew, M.-M.; Hecht, S. Adv. Mater. 2010, 22, 3348.
(4) Hansen, M. J.; Lerch, M. M.; Szymanski, W.; Feringa, B. L.
Angew. Chem., Int. Ed. 2016, 55, 13514.
Compounds 5h and 8 were subjected to ns transient
absorption measurements. After the excitation pulse at 355 or
410 nm, the ground-state bleach recovery in the π → π* region
was monitored and found to contain no detectable transients.
Thus, the only thing we can conclude is that if the absence of
́
(5) Bleger, D.; Hecht, S. Angew. Chem., Int. Ed. 2015, 54, 11338.
D
DOI: 10.1021/acs.orglett.8b02014
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(6) Velema, W. A.; Szymanski, W.; Feringa, B. L. J. Am. Chem. Soc.
2014, 136, 2178.
(7) Li, G.; Ma, L.; Liu, G.; Fan, C.; Pu, S. RSC Adv. 2017, 7, 20591.
(8) Banghart, M.; Borges, K.; Isacoff, E. Y.; Trauner, D.; Kramer, R.
H. Nat. Neurosci. 2004, 7, 1381.
(40) Bajwa, J. S.; Chen, G.-P.; Prasad, K.; Repic, O.; Blacklock, T. J.
Tetrahedron Lett. 2006, 47, 6425.
(41) Rooney, L.; Vidal, A.; D’Souza, A. M.; Devereux, N.; Masick,
B.; Boissel, V.; West, R.; Head, V.; Stringer, R.; Lao, J.; Petrus, M. J.;
Patapoutian, A.; Nash, M.; Stoakley, N.; Panesar, M.; Verkuyl, J. M.;
Schumacher, A. M.; Petrassi, H. M.; Tully, D. C. J. Med. Chem. 2014,
57, 5129.
́
(9) Ferreira, R.; Nilsson, J. R.; Solano, C.; Andreasson, J.; Grøtli, M.
Sci. Rep. 2015, 5, 09769.
(42) Van Horn, K. S.; Burda, W. N.; Fleeman, R.; Shaw, L. N.;
Manetsch, R. J. Med. Chem. 2014, 57, 3075.
(10) Kienzler, M. A.; Reiner, A.; Trautman, E.; Yoo, S.; Trauner, D.;
Isacoff, E. Y. J. Am. Chem. Soc. 2013, 135, 17683.
(43) Mantione, D.; Aizpuru, O. O.; Memeo, M. G.; Bovio, B.;
Quadrelli, P. Eur. J. Org. Chem. 2016, 2016, 983.
(44) Too, K.; Brown, D. M.; Bongard, E.; Yardley, V.; Vivasb, L.;
Loakes, D. Bioorg. Med. Chem. 2007, 15, 5551.
(45) Yin, T. T.; Zhao, Z.-X.; Zhang, H. X. New J. Chem. 2017, 41,
1659.
(11) Al-Atar, U.; Fernandes, R.; Johnsen, B.; Baillie, D.; Branda, N.
R. J. Am. Chem. Soc. 2009, 131, 15966.
(12) Velema, W. A.; van der Berg, J. P.; Hansen, M. J.; Szymanski,
W.; Driessen, A. J. M.; Feringa, B. L. Nat. Chem. 2013, 5, 924.
(13) Polosukhina, A.; Litt, J.; Nemargut, J.; Sychev, Y.; Trauner, D.;
Van Gelder, R. N.; Kramer, R. H. Neuron 2012, 75, 271.
(14) Gorostiza, P.; Isacoff, E. Y. Science 2008, 322, 395.
(15) Szymanski, W.; Beierle, J. M.; Kistemaker, H. A. V.; Velema, W.
A.; Feringa, B. L. Chem. Rev. 2013, 113, 6114.
(16) (a) Gelebart, A. H.; Mulder, D. J.; Varga, M.; Konya, A.;
Vantomme, G.; Meijer, E. W.; Selinger, R. L. B.; Broer, D. J. Nature
2017, 546, 632. (b) Zhang, J.; Wang, J.; He, T. Mater. Horiz. 2014, 1,
̈
169. (c) Gostl, R.; Senf, A.; Hecht, S. Chem. Soc. Rev. 2014, 43, 1982.
(d) Erbas-Cakmak, S.; Kolemen, S.; Sedgwick, A. C.; Gunnlaugsson,
T.; James, T. D.; Yoon, J.; Akkaya, E. U. Chem. Soc. Rev. 2018, 47,
2228.
́
(17) Garcia-Amoros, J.; Cuadrado, A.; Reig, A.; De Waele, V.;
Poizat, O. Chem. - Eur. J. 2015, 21, 14292.
(18) Vitaku, E.; Smith, D. T.; Njardarson, J. T. J. Med. Chem. 2014,
57, 10257.
(19) Wendler, T.; Schu
2012, 77, 3284.
̈
tt, C.; Nather, C.; Herges, R. J. Org. Chem.
̈
(20) Weston, C. E.; Richardson, R. D.; Haycock, P. R.; White, A. J.
P.; Fuchter, M. J. J. Am. Chem. Soc. 2014, 136, 11878.
(21) Weston, C. E.; Richardson, R. D.; Fuchter, M. J. Chem.
Commun. 2016, 52, 4521.
(22) Kolarski, D.; Szymanski, W.; Feringa, B. L. Org. Lett. 2017, 19,
5090.
(23) Whittaker, N.; Jones, T. S. G J. Chem. Soc. 1951, 1565.
(24) Ragon, F.; Yaksi, K.; Sciortino, N. F.; Chastanet, G.; Kepert, C.
J.; Neville, S. M.; et al. Aust. J. Chem. 2014, 67, 1563.
(25) Wang, Y.-T.; Liu, X.-Y.; Cui, G.; Fang, W.-H.; Thiel, W. Angew.
Chem., Int. Ed. 2016, 55, 14009.
(26) Otsuki, J.; Suwa, K.; Sarker, K.-K.; Sinha, C. J. Phys. Chem. A
2007, 111, 1403.
(27) Simeth, N. A.; Crespi, S.; Fagnoni, M.; Konig, B. J. Am. Chem.
Soc. 2018, 140, 2940.
̈
(28) Calbo, J.; Weston, C. E.; White, A. J. P.; Rzepa, H. S.;
Contreras-García, J.; Fuchter, M. J. J. Am. Chem. Soc. 2017, 139, 1261.
(29) Stricker, L.; Fritz, E.-C.; Peterlechner, M.; Doltsinis, N. L.;
Ravoo, B. J. J. Am. Chem. Soc. 2016, 138, 4547.
(30) Brown, E. V. J. Heterocycl. Chem. 1969, 6, 571.
(31) Kyziol, J. B. J. Heterocycl. Chem. 1985, 22, 1301.
(32) Merino, E. Chem. Soc. Rev. 2011, 40, 3835.
(33) Sharma, V.; Kamal, R.; Kumar, V. Curr. Top. Med. Chem. 2017,
17, 2482.
(34) Lim, Y.-K.; Lee, K. S.; Cho, C. G. Org. Lett. 2003, 5, 979.
(35) Muci, A. R.; Buchwald, S. L. Top. Curr. Chem. 2002, 219, 131.
(36) When 1c was stirred at 100 °C under microwave or normal
heating, an isomer of 2c was obtained as 2c-2 (see SI); when this
byproduct was heated at 120−150 °C, it converted back to 2c.
(37) Belliotti, T. R.; Wustrow, D. J.; Brink, W. A.; Zoski, K. T.; Shih,
Y.-H.; Whetzel, S. Z.; Georgic, L. M.; Corbin, A. E.; Akunne, H. C.;
Heffner, T. G.; Pugsley, T. A.; Wise, L. D. J. Med. Chem. 1999, 42,
5181.
(38) Lohou, E.; Collot, V.; Stiebing, S.; Rault, S. Synthesis 2011, 16,
2651.
(39) Hasan, I.; Marnelli, E. R.; Lin, L. C.; Fowler, F. W.; Levy, A. B.
J. Org. Chem. 1981, 46, 157.
E
DOI: 10.1021/acs.orglett.8b02014
Org. Lett. XXXX, XXX, XXX−XXX