Palladium-Mediated Approach to Coumarin-Functionalized Amino Acids

Article abstract of DOI:10.1021/acs.orglett.7b00854

Incorporation of the fluorogenic l-(7-hydroxycoumarin-4-yl)ethylglycine into proteins is a valuable biological tool. Coumarins are typically accessed via the Pechmann reaction, which requires acidic conditions and lacks substrate flexibility. A Pd-mediated coupling is described between o-methoxyboronic acids and a glutamic acid derived (Z)-vinyl triflate, forming latent coumarins. Global deprotection with BBr₃ forms the coumarin scaffold in a single step. This mild and scalable route yielded five analogues, including a probe suitable for use at lower pH.

Full text of DOI:10.1021/acs.orglett.7b00854

Letter
pubs.acs.org/OrgLett
Palladium-Mediated Approach to Coumarin-Functionalized Amino
Acids
Lindon W. K. Moodie, Samy Chammaa, Tomas Kindahl,† and Christian Hedberg*,†
†
‡,§
†
Umea
̊
University, Department of Chemistry, 90187 Umea,
Department of Chemical Biology, Max Planck Institute of Molecular Physiology, Otto-Hahn-Strasse 11, 44227 Dortmund, Germany
S Supporting Information
̊
Sweden
‡
*
ABSTRACT: Incorporation of the fluorogenic L-(7-hydroxycou-
marin-4-yl)ethylglycine into proteins is a valuable biological tool.
Coumarins are typically accessed via the Pechmann reaction, which
requires acidic conditions and lacks substrate flexibility. A Pd-
mediated coupling is described between o-methoxyboronic acids
and a glutamic acid derived (Z)-vinyl triflate, forming latent
coumarins. Global deprotection with BBr forms the coumarin
3
scaffold in a single step. This mild and scalable route yielded five analogues, including a probe suitable for use at lower pH.
he site-specific incorporation of fluorophore-containing
unnatural amino acids (UAAs) into a protein of interest
T
by genetic code expansion has emerged as a valuable alternate
strategy to expressible fluorescent protein tags, for example,
green fluorescent protein (GFP). When compared with large,
protein-derived reporters (often greater than 20 kDa),
fluorescent UAAs are significantly smaller in size and can be
incorporated at the internal positions of a protein, whereas GFP
1
is limited to the C- and N-termini. Practically, UAAs can be
introduced into a protein via engineered orthogonal tRNA/
aminoacyl-tRNA synthetase pairs in response to the amber stop
2
codon (UAG), strategically placed in a gene of interest. The
3
4
method is applicable to prokaryotic and eukaryotic systems,
5
as well as living organisms. Pioneering work by Schultz and co-
workers demonstrated that L-(7-hydroxycoumarin-4-yl)-
ethylglycine (1) could be selectively incorporated into a
modified protein and used to probe urea-dependent denatura-
Figure 1. An alternate approach to the Schultz amino acid 1.
6
tion. UAA 1 exhibits favorable photophysical properties such
as solvent- and pH-dependent fluorescence (dictated by the
We envisioned a practical synthesis of L-coumarin-4-
ylethylglycines, where both the nature and position of aryl
ring substituents could be varied outside of the strict
prerequisites of the Pechmann condensation while conserving
a similar disconnection. Obtaining access to analogues of 1 will
expand the scope of coumarins as tools for chemical biology.
Retrosynthetically, the α,β-unsaturated ester (4) was identified
as the key intermediate toward these goals. We envisaged it
could be accessed via a palladium-mediated cross coupling
between a suitably substituted arylboronic acid (5) and a vinyl
triflate (6) (Figure 1). Compound 6 can be derived from the
commercially available N-Cbz-L-glutamic acid benzyl ester.
Careful consideration of the phenolic and amino acid
protecting groups would enable global deprotection and
pK of the phenolic functionality: 7.8) and a reasonable
a
quantum yield and Stokes shift. Consequently, it has proven a
7
useful tool in chemical and molecular biology. Given the
application of 1, it is somewhat surprising that analogues with
varied aryl substitution patterns have not been extensively
explored. This can be attributed to the typical methodology for
accessing coumarin-4-yl-ethylglycines: the Pechmann conden-
sation of resorcinol (2) and a glutamic acid-derived β-keto ester
8
(
3) (Figure 1). In this approach, resorcinol is the only practical
reaction partner for the keto ester, limiting the availability of
analogues as a consequence. The strongly acidic and
dehydrating conditions and requirement of superstoichiometric
amounts of the aryl partner enforce further limitations.
Furthermore, preparative HPLC is commonly used for the
7c
purification of 1, which is hampered by solubility problems.
Received: March 22, 2017
Revised: May 4, 2017
Several alternate strategies have been reported that provide 1,
9
albeit as racemates.
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.7b00854
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
coumarin formation to be effected in a single reaction step. In
this regard, the N-Cbz, benzyl ester, and phenolic methyl ether
groups were chosen due to their lability upon exposure to
boron tribromide. In order to avoid purification by preparative
HPLC, we aimed that the product obtained in the final step
Elution of the boric acid byproduct with water, followed by a
gradient of aq NH OH, provided the coumarin amino acid 1 in
4
60% isolated yield after lyophilization, avoiding preparative
HPLC. The Mosher’s amide of 1 was prepared, and no
extensive epimerization was observed (see the Supporting
Information).
7c,i
could be purified by ion exchange chromatography.
The synthesis commenced with the formation of vinyl triflate
from β-keto ester 3 (Scheme 1), derived from the
To further elucidate the scope of the coumarin synthesis, we
performed the cross coupling of 6 with a number of
commercially available o-methoxy-substituted arylboronic
acids (Table 1, entries 1−4). Here, all derivatives employed
6
Scheme 1. Synthesis of the Schultz Coumarin 1
Table 1. Substrate Scope of Coumarin Synthesis
corresponding glutamic acid in 85% yield (see Supporting
Information). As the Z isomer of 6 was desired for ring closure
to form the coumarin motif, we applied the modified
Schotten−Baumann-type conditions of Babinski et al., where
the enolate geometry, and therefore stereoselectivity of the enol
10
triflate formation, is controlled by the choice of base. Initial
experiments using LiOH and trifluoromethanesulfonic anhy-
dride provided crude 6 as a 9:1 mixture of the Z/E isomers,
respectively. Further tuning of the reaction conditions,
including slow addition and lowering the internal reaction
temperature to 0−5 °C, resulted in a 99:1 Z/E ratio and a yield
of 90% after chromatography. The importance of temperature
control was most pronounced during large-scale (≥20g)
preparations of 6. Attempts to selectively synthesize the
10
isomeric E-6 ((Me) NOH, hexanes ) to test if the E isomer
4
could isomerize and undergo coumarin formation during BBr3
treatment were not successful.
With intermediate Z-6 in hand, we subjected it to cross
coupling with 2,4-dimethoxyphenylboronic acid (5). A number
of conditions and reagent combinations were screened, and the
operationally simple and economical conditions of Pd(PPh3)4
(
5 mol %) and K PO in 1,4-dioxane/water at 50 °C gave the
3 4
most satisfying results. Catalyst loading could be further
reduced to 2.5 mol % without diminishing coupling perform-
ance and could be performed on gram scale (1.11 g of triflate,
8
5% yield). As we could access the linear coumarin precursor 4
on a practical scale, we sought to identify conditions for the
global deprotection/cyclization step to complete the synthesis.
The challenges of this reaction involved optimizing the amount
of BBr required, formation of the coumarin scaffold, and
3
developing a practical purification method. Solutions of 4 in
dichloromethane were cooled to −78 °C, treated with varying
performed well, with the yields for 7−10 in the range of 80−
86%. The BBr -initiated deprotection/cyclization of these linear
3
amounts of BBr , and slowly warmed to ambient temperature
motifs furnished the coumarins 12−15. Varying the position
and presence of the phenolic substituent provides analogues to
probe the structure−activity relationship (SAR) of the different
aminoacyl-tRNA synthetase mutants during amber suppression
screening and mutagenesis.
3
before reverse quenching into ice water. It was found that 5.4
equiv of BBr3 was sufficient for global deprotection and
cyclization, resulting in a sole product, 1. Purification of the
reaction product was complicated by the large amounts of boric
acid formed during reaction quenching. To remedy this, the
biphasic mixture from the quench was concentrated to a small
volume and the resulting slurry applied directly to a Dowex 50
As the fluorescent properties of coumarin 1 are influenced by
the acidity of the C-7 phenol, it would be useful to access
9b,11
analogues where the phenolic pK is lowered.
An UAA with
a
+
column (H form, 300 mesh, 40 g wet resin per g product).
these properties would be advantageous for studies of biological
B
DOI: 10.1021/acs.orglett.7b00854
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
components that operate under acidic conditions, i.e., the
endosomal and lysosomal pathways (pH ≈ 5). Fluorine
substituents at the 6 and 8 positions of 7-OH-coumarin have
with a glutamic acid derived vinyl triflate 6 (99:1 Z/E) to
access the latent precoumarin 4. Due to the careful selection of
the phenolic and amino acid protecting groups, treatment of
been shown to significantly reduce the phenolic pK while
the linear compound 4 with BBr initiated the removal of all
a
3
11
improving photostability. Toward these goals, the difluori-
nated boronic acid pinacol ester 17 was synthesized from
protecting groups and effected lactonization to construct the
coumarin motif. To demonstrate the versatile and mild nature
of this reaction sequence, coumarins 12−16, which differ by the
phenolic substitution of their aryl rings, were synthesized. Of
importance, coumarins 1 and 12−16 could be obtained on a
useful scale without relying on preparative HPLC purification.
It is envisaged these analogues will provide insight into SAR
involved in aminoacyl-tRNA synthetase mutagenesis and be
useful as fluorescent probes to investigate biological processes
that occur in lower pH environments.
2
,3,4,5-tetrafluoronitrobenzene (35% over four steps, see the
Supporting Information). Both the coupling and coumarin
formation reactions proceeded smoothly (86 and 77%,
respectively) to afford the fluorinated coumarin 16.
The spectroscopic properties of coumarins 1 and 16 were
investigated in phosphate buffer at pH 2.8−8.0. Monitoring the
effect of pH on absorption (360 nm) allowed us to determine
the pK of compound 16 as 4.8 (see the SI; pK of 1: 7.8).
a
a
While the absorption spectra of both compounds changed at
the different pH’s (Figure S1), the shape and wavelength range
of the fluorescence spectra were independent of pH throughout
the measured range, indicating that emission only occurs from
one ionization form of the compounds (Figures S2 and S3).
Both compounds displayed high quantum yields, relatively long
lifetimes, and well-separated emission peaks (Table 2). In
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.7b00854.
Experimental procedures; H, 13C and 19F NMR spectra
1
for novel compounds; spectroscopic studies (PDF)
Table 2. Spectroscopic Data for Compounds 1 and 16 in
Phosphate Buffer
AUTHOR INFORMATION
compd pH λmax (nm) ε (M− cm−1
1
)
emmax (nm)
Φf
τ (ns)
■
Corresponding Author
*E-mail: christian.hedberg@chem.umu.se.
ORCID
1
2.8
.0
2.8
.0
320
362
320
362
11900
11500
10000
14900
456
455
458
457
0.68
0.72
0.68
0.67
5.5
5.5
5.5
5.6
8
16
8
Tomas Kindahl: 0000-0002-6548-6158
Christian Hedberg: 0000-0002-0832-8276
Present Address
contrast to compound 1 that showed spectroscopic differences
only at pH ≥ 6, difluorinated compound 16 showed marked
differences in absorbance at pH 4−6, resulting in varied
fluorescence intensity (Figure 2). This demonstrates its
potential as a wavelength-ratiometric pH probe for application
in low pH environments (Figure S5).
§
(S.C.) Taros Chemicals GmbH & Co. KG, Emil-Figge-Strasse
76a, 44227 Dortmund, Germany.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
We thank the Max-Planck society (Germany), Umea
Sweden), and Knut and Alice Wallenberg Foundation
Sweden) for generous support. L.W.K.M. thanks the Kempe
■
(
(
̊
University
Foundation (Sweden) for a postdoctoral fellowship.
REFERENCES
■
(
1) (a) Krueger, A. T.; Imperiali, B. ChemBioChem 2013, 14, 788−
7
8
2
99. (b) Harkiss, A. H.; Sutherland, A. Org. Biomol. Chem. 2016, 14,
911−8921. (c) Sinkeldam, R. W.; Greco, N. J.; Tor, Y. Chem. Rev.
010, 110, 2579−2619.
(
2) (a) Lang, K.; Chin, J. W. Chem. Rev. 2014, 114, 4764−4806.
(
b) Kim, C. H.; Axup, J. Y.; Schultz, P. G. Curr. Opin. Chem. Biol. 2013,
1
(
2
(
7, 412−419.
3) Wang, L.; Brock, A.; Herberich, B.; Schultz, P. G. Science 2001,
92, 498−500.
4) Chin, J. W.; Cropp, T. A.; Anderson, J. C.; Mukherji, M.; Zhang,
Z. W.; Schultz, P. G. Science 2003, 301, 964−967.
5) (a) Greiss, S.; Chin, J. W. J. Am. Chem. Soc. 2011, 133, 14196−
4199. (b) Chin, J. W. Annu. Rev. Biochem. 2014, 83, 379−408.
6) Wang, J. Y.; Xie, J. M.; Schultz, P. G. J. Am. Chem. Soc. 2006, 128,
738−8739.
7) (a) Cottam Jones, J. M.; Harris, P. W. R.; Scanlon, D. B.; Forbes,
Figure 2. Emission at 460 nm of compounds 1 and 16 at pH 2.8−8.0
when excited at 360 nm. A.U. = arbitrary units.
(
1
(
8
In conclusion, we describe a novel route to coumarin-
functionalized amino acids for application in genetic code
expansion technologies. Traditionally, the Pechmann cycliza-
tion has been used to access non-natural amino acids such as 1,
which operates under harsh conditions and is not amenable to
analogue exploration. We circumvented these problems via the
palladium-mediated coupling of 2-methoxyarylboronic acid
(
B. E.; Brimble, M. A.; Abell, A. D. Org. Biomol. Chem. 2016, 14, 2698−
2705. (b) Dean, S. F.; Whalen, K. L.; Spies, M. A. ACS Cent. Sci. 2015,
1, 364−373. (c) Amaro, M.; Brezovsky, J.; Kovacova, S.; Sykora, J.;
C
DOI: 10.1021/acs.orglett.7b00854
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Bednar, D.; Nemec, V.; Liskova, V.; Kurumbang, N. P.; Beerens, K.;
Chaloupkova, R.; Paruch, K.; Hof, M.; Damborsky, J. J. Am. Chem. Soc.
2
015, 137, 4988−4992. (d) Chen, S. X.; Fahmi, N. E.; Wang, L.;
Bhattacharya, C.; Benkovic, S. J.; Hecht, S. M. J. Am. Chem. Soc. 2013,
35, 12924−12927. (e) Ho, D.; Lugo, M. R.; Lomize, A. L.;
Pogozheva, I. D.; Singh, S. P.; Schwan, A. L.; Merrill, A. R. Biochemistry
1
2
011, 50, 4830−4842. (f) Charbon, G.; Brustad, E.; Scott, K. A.;
Wang, J. Y.; Lobner-Olesen, A.; Schultz, P. G.; Jacobs-Wagner, C.;
Chapman, E. ChemBioChem 2011, 12, 1818−1821. (g) Saraogi, I.;
Zhang, D. W.; Chandrasekaran, S.; Shan, S. O. J. Am. Chem. Soc. 2011,
1
33, 14936−14939. (h) Lacey, V. K.; Parrish, A. R.; Han, S. L.; Shen,
Z. X.; Briggs, S. P.; Ma, Y. G.; Wang, L. A. Angew. Chem., Int. Ed. 2011,
0, 8692−8696. (i) Kuhn, S. M.; Rubini, M.; Muller, M. A.; Skerra, A.
5
J. Am. Chem. Soc. 2011, 133, 3708−3711. (j) Ugwumba, I. N.; Ozawa,
K.; Xu, Z. Q.; Ely, F.; Foo, J. L.; Herlt, A. J.; Coppin, C.; Brown, S.;
Taylor, M. C.; Ollis, D. L.; Mander, L. N.; Schenk, G.; Dixon, N. E.;
Otting, G.; Oakeshott, J. G.; Jackson, C. J. J. Am. Chem. Soc. 2011, 133,
3
26−333. (k) Mendes, K. R.; Martinez, J. A.; Kantrowitz, E. R. ACS
Chem. Biol. 2010, 5, 499−506.
8) Brun, M. P.; Bischoff, L.; Garbay, C. Angew. Chem., Int. Ed. 2004,
3, 3432−3436.
9) (a) Braun, M.; Dittrich, T. Beilstein J. Org. Chem. 2010,
(
4
(
DOI: 10.3762/bjoc.6.69. (b) Xu, X. Y.; Hu, X. S.; Wang, J. Y. Beilstein
J. Org. Chem. 2013, 9, 254−259.
(
10) Babinski, D.; Soltani, O.; Frantz, D. E. Org. Lett. 2008, 10,
2
(
1
901−2904.
11) Sun, W. C.; Gee, K. R.; Haugland, R. P. Bioorg. Med. Chem. Lett.
998, 8, 3107−3110.
D
DOI: 10.1021/acs.orglett.7b00854
Org. Lett. XXXX, XXX, XXX−XXX