A very short route to enantiomerically pure coumarin-bearing fluorescent amino acids

Article abstract of DOI:10.1002/anie.200454116

No bulkier than a tryptophan or a tyrosine residue are the fluorescent coumaryl amino acids described. These building blocks can be synthesized rapidly with high enantiomeric purity and introduced readily into a peptidic sequence. Their fluorescence prope

Full text of DOI:10.1002/anie.200454116

Communications
Fluorescent Amino Acids
AVery Short Route to Enantiomerically Pure
Coumarin-Bearing Fluorescent Amino Acids**
Marie-Priscille Brun, Laurent Bischoff, and
Christiane Garbay*
Fluorescent derivatives of biologically active peptides are
useful experimental tools for studying biological structure and
function[1] and for visualizing intracellular processes or
molecular interactions.[2] To achieve adequate detection of
peptide fluorescence, the monitored peptide must either
include a natural fluorescent amino acid, such as tyrosine
(Tyr) or tryptophan (Trp), or must be labeled with extrinsic or
intrinsic probes.[3] Since C- and N-terminal domains are often
important interactive sites, the incorporation of the fluores-
cent label into the peptide at a well-chosen position is critical
to avoiding changes in the conformational and biological
characteristics of the parent peptide.[4a] The use of an amino
acid as a fluorescent chromophore[4] offers at least two
advantages: first, peptides containing this amino acid can be
synthesized by solid-phase peptide synthesis (SPPS); and
second, since both ends of these peptides are free, other
residues may be attached to improve solubility or lability.[5] In
addition to the excitation and emission properties of a
[*] M.-P. Brun, Prof. Dr. C. Garbay
Laboratoire de Pharmacochimie MolØculaire et Cellulaire
INSERM U648 CNRS FRE 2718, UFR Biomedicale
45, Rue des Saints P›res, 75270 Paris Cedex 06 (France)
Fax : (+33)142-864-082
E-mail: christiane.garbay@univ-paris5.fr
Prof. Dr. L. Bischoff
IRCOF, UPRES-A 6014, INSA
B.P. 08, 76131 Mont-St-Aignan Cedex (France)
[**] This research was supported by “La Ligue Nationale contre le
Cancer”. We thank Dr. Veronique Gigoux for assistance in the
preparation of cell cultures and Dr. Nicolae Ghinea for confocal
microscopy studies.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
3432
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/anie.200454116
Angew. Chem. Int. Ed. 2004, 43, 3432 –3436

Angewandte
Chemie
fluorescent amino acid its steric hindrance must also be
considered. Thus, augmenting the conjugation in an amino
acid increases its fluorescence-emission properties. On the
other hand, however, such increased conjugation (as in the
case of polycycles) gives rise to more steric hindrance,
associated with lowered coupling yields. Furthermore, since
a higher chemical stability is expected if the fluorophore is
linked to the amino acids by a side-chain carbon–carbon
bond, we have focused on developing small and sensitive
fluorescent coumaryl amino acids that could be incorporated
efficiently into peptides in place of Tyr or Trp.
von Pechmann condensation in methanesulfonic acid
(MSA).[15] This condensation, carried out in the presence of
25 equivalents of MSA, proceeded smoothly and resulted in
the complete dissolution of starting material and the complete
conversion of the b-ketoester. Reaction times depended on
the substituents on the phenol ring: the greater the number of
electron-donating groups, the shorter the reaction time.
À
À
Since OCO benzyl and NCO2 benzyl bonds are not
stable under these conditions, running the reaction in the
presence of one equivalent of resorcinol produced polyben-
zylated coumarins. This could be avoided by using a large
excess of the resorcinol derivative, which acts as a scavenger
for the benzyl cations (Scheme 2). This allowed the formation
Coumarins are interesting fluorescent molecules due to
their relatively high quantum yields, low bleaching, reason-
able stability, and relative ease of synthesis. They have
been used extensively to label amines, thiols, and acids,[6]
thereby providing useful experimental tools.[7] To date,
enantioselective synthesis of amino acids,[8] especially
coumaryl amino acids, has been carried out by diaster-
eoselective alkylation of chiral glycine equivalents. For
example, the chiral auxiliaries of Oppolzer et al.[9] and
Williams[10] were used for the synthesis of l-(6,7-dime-
thoxycoumar-4-yl)alanine and l-(7-methoxycoumar-4-
yl)alanine (Mca, (S)-6), respectively.[4c,d] These methods
produced optically pure amino acids, but they required
multiple steps and the amino acids were obtained in poor
global yields. Herein we report a more rapid, cheap, and
versatile synthesis of coumaryl amino acids that led to
coumaryl alanines and coumaryl ethylglycines in either a
two- or a three-step process.
Scheme 2. General synthesis of coumaryl amino acids.
Protected aspartic (Asp) and glutamic acids (Glu)
were used as chiral starting materials,[11] thus avoiding
high-cost chiral auxiliaries. They were transformed into side-
chain b-ketoesters (which we had chosen as coupling partners
for the condensation with phenols to obtain the coumarin
moiety[12]) by using the procedure of Masamune et al.[13]
(Scheme 1). Pure b-ketoesters 1–3 were obtained in good
yields after silica-gel column chromatography.
Coumarins have been synthesized by many pathways,
including von Pechmann, Perkin, Knoevenagel, Reformatsky,
and Wittig reactions.[14] The von Pechmann reaction is one of
the simplest and most straightforward methods. It involves a
phenol condensation with a b-ketoester in the presence of a
variety of Brønsted[14a,15] and Lewis acids.[12,16] We chose the
of compounds 5 and 14 in only one, direct step. For expensive
resorcinols, we first deprotected both amine and carboxy
groups to avoid side reactions (step c in Scheme 1). The
deprotected aspartic acid derived b-ketoester, (S)- or (R)-4,
was then treated with various polysubstituted phenols in the
presence of MSA to give compounds 5–13 (Scheme 3).
Hydrogenolytic conditions used for Asp are unsuitable for
Scheme 1. Synthesis of amino acid derived b-ketoesters. a) Carbonyl-
diimidazole, THF, room temperature; b) magnesium salt of monoethyl
malonic acid, THF, 08C and 12 h at room temperature, 50–80%; c) H2,
Pd/C 10%, HCl, AcOEt/EtOH, room temperature, 90–100% (and for
2, 1:1 TFA:CHCl3 at 08C and 1 h at room temperature). TFA=trifluoro-
acetic acid, Cbz=benzyloxycarbonyl, Boc=tert-butoxycarbonyl,
Bn=benzyl.
Scheme 3. Synthesis of aspartic acid derived coumaryl amino acids.
MSA (25 equiv), 08C, then room temperature, 2–4 h, 16% (8) to 71%
À
(6); XÀ =CH3SO3 ((S)-6) or CF3CO2À in case of purification by semi-
preparative HPLC on a Kromasil C8 column (A (H2O+0.1%TFA) and B
(70:30CH3CN:H2O+0.09%TFA), with an appropriate linear gradient
of B).
Angew. Chem. Int. Ed. 2004, 43, 3432 –3436
www.angewandte.org
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3433

Communications
Glu.[17] Therefore, coumarin formation was conducted on the
still-protected b-ketoester 3 with a large excess of 3-methoxy-
phenol (Scheme 4), which led to two products. In one product,
compound 14, the amine and acid functionalities were
Scheme 5. Synthesis of N-protected coumaryl amino acids. a) 10%
NaHCO3 at 08C, FmocCl at 08C, 1:1 dioxane:H2O, 1 h at 08C then 2 h
at room temperature, 80% ((S)-16) and 70% ((R)-16); b) NaOH at
08C, CbzCl at 08C, 1:1 dioxane:H2O, 1 h at room temperature, 91%;
c) Boc2O at 08C, 1:1 dioxane:5%NaHCO3, 1 h at 08C then 3 h at
room temperature, 84% (18) and 75% (19 and 20). Fmoc=9-fluore-
nylmethoxycarbonyl.
Scheme 4. Synthesis of glutamic acid derived coumaryl amino acids.
3-Methoxyphenol (10 equiv), MSA (25 equiv), 08C then room tempera-
ture, 2 h, 60% (ratio 14:15=62:38).
deprotected, whereas the other, compound 15, still bore the
benzyl ester group. Coumaryl amino acids in which the amino
group was protected with a widely used protecting group, such
as Fmoc ((S)- and (R)-16), Cbz (17), or Boc (18–20), were also
synthesized (Scheme 5).
Since a stereogenic carbon atom is present in these amino
acids, we measured optical rotation (Table 1) and also
determined the enantiomeric purity of one of the compounds.
We assumed that the other compounds would exhibit the
same sensitivity towards racemization, given that the most
probable source of racemization should be the earlier
formation of the b-ketoester in a slightly alkaline medium.
To verify the optical purity, the amino groups of (S)- and (R)-6
were protected with an Fmoc group and the compounds
coupled with tert-butylalaninate. The purity of the corre-
sponding diastereomeric products 21 and 22 was examined by
using appropriate HPLC isocratic conditions (Figure 1). In
the case of 21, only one diastereomer could be detected. For
22, a diastereomeric ratio of 98:2 was observed for the crude
reaction mixture. We can conclude that no racemization
occurred during the b-ketoester formation, in the course of
the cyclization, or during the protection and coupling
processes (BOP (= benzotriazol-1-yloxytris(dimethylamino)-
phosphonium hexafluorophosphate)/DIEA (= N,N-diisopro-
pylethylamine)). These observations suggest that the N-
protected amino acids obtained can be incorporated by
SPPS into peptides without risk of racemization.
Absorption and fluorescence wavelength maxima, extinc-
tion coefficients, and fluorescence quantum yields (QY) were
measured for each compound (Figure 2, Table 1). The optical
Table 1: Physicochemical properties of coumaryl amino acids.[a]
n
R1
R2
R3
Yield [%]
[a]2D0[a]
lAbs. [nm][b]
lEm. [nm][b]
e [cmÀ1 mÀ1 [c]
]
QY[d]
Trp
Tyr
5
(S)-6
(R)-6
7
–
–
–
–
1
1
1
1
1
1
1
1
1
1
2
2
–
–
–
–
–
–
H
H
H
H
H
H
OH
OMe
OMe
OH
H
–
–
–
–
8.4
278[18a]
274[18a]
329
324
323
345
331
325
329
323
324
327
321
352[18a]
303[18a]
464
383
386
423
465
387
421
421
419
463
380
5300[18a]
1400[18a]
9000
12000
14000
12000
10300
13300
6800
12700
11500
11800
12500
14600
0.12[18b]
0.13[18b]
0.49
0.18
0.15
0.60
0.54
0.17
0.50
(2S)
(2S)
(2R)
(2S)
(2S)
(2S)
(2S)
(2S)
(2S)
(2S)
(2S)
(2S)
OH
OMe
OMe
OMe
OH
OEt
OMe
OH
OMe
OH
OMe
OMe
H
H
H
OMe
Cl
H
H
H
H
H
H
H
47
71
70
45
16
39
36
18
34
48
37
23
12.9
À12.1
n.d.[e]
3.2
15.7
n.d.[e]
10.1
12.0
13.2
19.4
n.d.[e]
8
9
10
11
12
13
14
15
0.18
0.39
n.d.[e]
0.21
H
322
381
0.17
[a] Determined in 1n HCl. [b] Determined in 95% ethanol. [c] Extinction coefficient. [d] Quantum yield using quinine sulfate in sulfuric acid as a
standard reference. [e] Not determined.
3434
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 3432 –3436

Angewandte
Chemie
Figure 1. HPLC chromatograms of diastereoisomers 21 and 22. HPLC
isocratic conditions on a Vydac C18 column (38% A (H2O+0.1%TFA)
and 62% B (3:1:1 CH3CN:EtOH:MeOH)).
Figure 3. Peptide visualization using confocal microscopy. HeLa cells
grown in DMEM (=Dulbecco’s minimum essential medium) were
treated with 10 mm peptide 2. After 2 h incubation at 378C, the cells
were washed with PBS (=phosphate-buffered saline), fixed with 3%
PFA (=paraformaldehyde) for 20 min at room temperature and
mounted for examination. Images were obtained with a Zeiss LSM510
confocal scanning microscope using an argon ion laser set at 488 nm.
amino acids as intrinsic fluorescent labels.
À
À
À
peptide 1 H2N RQIKIWFQNRRMKWKK Mca COOH
À
À
peptide 2 H2N Hca-RQIKIWFQNRRMKWKK COOH
In conclusion, we have developed an extremely direct
synthesis of optically pure fluorescent amino acids, starting
from inexpensive chiral materials. Our synthesis represents a
useful tool in the preparation of coumaryl amino acids, which
can be easily incorporated into peptides and used in biological
studies. Potential applications of these molecules include the
wide range of investigating cellular processes and other
aspects of biological mechanism and function.
Figure 2. Fluorescence properties of the coumaryl amino acids 5–15.
Experimental Section
Synthesis of peptides 1 and 2: SPPS by using small-scale Fmoc
chemistry on a HMP (= 4-hydroxymethylphenoxy) resin, DCC/HOBt
coupling method (BOP/HOBt/DIEA for Hca), removal of Fmoc
groups with piperidine (20% in NMP), resin cleavage with TFA/
TIPS/H2O (9.5:0.25:0.25 in volume) and HPLC semipreparative
purification on a Vydac C18 column (A (H2O+0.1%TFA) and B
(70:30 CH3CN:H2O+0.09%TFA), using a linear gradient of B from
10 to 90% in 30 min). DCC = N,N’-dicyclohexylcarbodiimide,
HOBt = 1-hydroxybenzotriazole, NMP = N-methylpyrrolidinone,
TIPS = triisopropylsilane.
properties of the natural fluorescent amino acids Trp and Tyr
were included for comparison. The novel coumaryl amino
acids 5–14 had higher wavelength values and higher absorp-
tion values than Trp or Tyr, and they displayed higher
quantum yields than Trp or Tyr, especially in the case of
compounds 5, 7, 8, 10 and 12, with values ranking from 0.39 to
0.60 versus 0.12 and 0.13 for Trp and Tyr, respectively. These
characteristics make compounds 5–14 good fluorophores for
biological studies.
See Supporting Information for detailed reaction conditions and
compound characterization.
The use of such fluorescent amino acids in SPPS was
validated by adding N-Fmoc-protected (S)-6 (Mca) to the C-
terminal end of penetratin, a 16 amino acid peptide corre-
sponding to the third helix of the homeodomain of the
Antennapedia protein (to give peptide 1, see Experimental
Section).[19] To investigate whether the coumaryl amino acid
could be used in biological studies, compound 5 (Hca) was
added to the N-terminal end of penetratin (to give peptide 2,
see Experimental Section). After incubation of this peptide in
HeLa cells, we visualized its internalization by confocal
microscopy (Figure 3), thus validating the use of coumaryl
Received: February 26, 2004 [Z54116]
Keywords: amino acids · coumarin · cyclization · fluorescent
.
probes · solid-phase synthesis
[1] a) C. G. Bradshaw, K. Ceszkowski, G. Turcatti, I. J. Beresford, A.
Chollet, J. Med. Chem. 1994, 37, 1991 – 1995; b) J.-M. Soleilhac,
F. Cornille, L. Martin, C. Lenoir, M.-C. FourniØ-Zaluski, B. P.
Roques, Anal. Biochem. 1996, 241, 120 – 127.
Angew. Chem. Int. Ed. 2004, 43, 3432 –3436
www.angewandte.org
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3435

Communications
[2] a) G. Turcatti, H. Vogel, A. Chollet, Biochemistry 1995, 34,
3972 – 3980; b) M. R. Tota, C. D. Strader, J. Biol. Chem. 1990,
265, 16891 – 16897; c) D. J. Cowley, A. J. Schulze, J. Pept. Res.
1997, 49, 444 – 454.
[3] a) T. Licher, E. Kellner, H. Lill, FEBS Lett. 1998, 431, 419 – 422;
b) G. Drin, M. Mazel, P. Clair, D. Mathieu, M. Kaczorek, J.
Temsamani, Eur. J. Biochem. 2001, 268, 1304 – 1314; c) G. Jin, X.
Huang, R. Black, M. Wolfson, C. Rauch, H. McGregor, G.
Ellestad, R. Cowling, Anal. Biochem. 2002, 302, 269 – 275;
d) E. H. Pap, T. B. Dansen, K. W. Wirtz, Trends Cell Biol. 2001,
11, 10 – 12.
[4] a) H. Mihara, S. Lee, Y. Shimohigashi, H. Aoyagi, T. Kato, N.
Izumiya, T. Costa, FEBS Lett. 1985, 193, 35 – 38; b) M. Taki, T.
Hohsaka, H. Murakami, K. Taira, M. Sisido, FEBS Lett. 2001,
507, 35 – 38; c) F. A. Bennett, D. J. Barlow, A. N. O. Dodoo, R. C.
Hider, A. B. Lansley, M. J. Lawrence, C. Marriott, S. S. Bansal,
Tetrahedron Lett. 1997, 38, 7449 – 7452; d) P. Kele, G. Sui, Q.
Huo, R. M. Leblanc, Tetrahedron Asymmetry 2000, 11, 4959 –
4963.
[5] W. Wang, T. C. Liang, Biochem. Biophys. Res. Commun. 1994,
201, 835 – 840.
[6] a) H. Wendt, C. Berger, A. Baici, R. M. Thomas, H. R. Bosshard,
Biochemistry 1995, 34, 4097 – 4107; b) J. H. Wolf, J. Korf, J.
Pharm. Biomed. Anal. 1992, 11, 99 – 107.
[7] a) R. OꢀKennedy in Coumarins, Wiley, New York, 1997; b) R. M.
Christie, Rev. Prog. Color. Relat. Top. 1993, 23, 1.
[8] a) R. O. Duthaler, Tetrahedron 1994, 50, 1539 – 1650; b) R. M.
Williams, Aldrichimica Acta 1992, 25, 11 – 25.
[9] W. Oppolzer, A. J. Kingma, G. Poli, Tetrahedron Lett. 1989, 30,
6009.
[10] W. Zhai, Tetrahedron 1988, 44, 5425 – 5430.
[11] R. M. Adlington, J. E. Baldwin, D. Catterick, G. J. Pritchard, J.
Chem. Soc. Perkin Trans. 1 2001, 668 – 679.
[12] S. P. Chavan, K. Shivasankar, R. Sivappa, R. Kale, Tetrahedron
Lett. 2002, 43, 8583 – 8586.
[13] D. W. Brooks, L. D. Lu, S. Masamune, Angew. Chem. 1979, 91,
76 – 77; Angew. Chem. Int. Ed. Engl. 1979, 18, 72 – 74.
[14] a) H. von Pechmann, C. Duisberg, Ber. Dtsch. Chem. Ges. 1883,
16, 2119; b) J. R. Johnson, Org. React. 1942, 1, 210; c) G. Brufola,
F. Fringuelli, O. Piermatti, F. Pizzo, Heterocycles 1996, 43, 127;
d) R. L. Shriner, Org. React. 1942, 1, 1; e) I. Yavari, R. Hekmat-
Shoar, A. Zonouzi, Tetrahedron Lett. 1998, 39, 2391.
[15] W. C. Sun, K. R. Gee, R. P. Haugland, Bioorg. Med. Chem. Lett.
1998, 8, 3107 – 3110.
[16] D. S. Bose, A. P. Rudradas, M. H. Babu, Tetrahedron Lett. 2002,
43, 9195 – 9197.
[17] M. P. Brun, A. S. Martin, C. Garbay, L. Bischoff, Tetrahedron
Lett. 2003, 44, 7011 – 7013.
[18] a) G. D. Fasman in Handbook of Biochemistry and Molecular
Biology, Proteins, Vol. I, CRC, Boca Raton, FL, 1976, pp. 183 –
203; b) R. Chen, J. Res. Natl. Bur. Stand. Sect. A 1976, 76, 593 –
606.
[19] L. Theodore, D. Derossi, G. Chassaing, B. Llir, M. Kubes, P.
Jordan, H. Chneiweiss, P. Godement, A. Prochiantz, J. Neurosci.
1995, 15, 7158 – 7167.
3436
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 3432 –3436