but no information on the relative binding affinities of the
isomers is reported. Preferential binding of -phenylalanine to
L
BSA was observed in ultrafiltration experiments,10b and in all
the above examples, the observed selectivity was only moder-
ate. The larger discrimination observed with Py-Phe, compared
to the naphthyl or the dansyl amino acids, can be attributed to
the increased hydrophobic surface area of the pyrenyl chromo-
phore. Larger, rigid, hydrophobic surfaces of correct configura-
tion can interact better with the protein, while a wrong isomer
will interact poorly, thus widening the gap between the two
binding constants.
The chiral center present in Py-Phe is five atoms away from
the pyrenyl chromophore and yet it dramatically influences the
binding behavior/spectral properties of the pyrenyl chromo-
phore. Both isomers may bind at the same site or may bind at
different sites on BSA. These two models can be distinguished
in photocleavage experiments in which the pyrenyl probe can be
activated with light to cleave the protein backbone at the probe
binding site. Such experiments are in progress. Hydrophobic
Fig. 2 Plot of the fluorescence intensities (I0/I) of Py-
circles) and Py- -Phe (2 mM, filled circles) recorded at increasing
concentrations of BSA. Note that larger concentrations of BSA are needed
to saturate binding of Py- -Phe. Both samples were excited at 344 nm.
D
-Phe (2 mM, open
L
D
to be different. These details are examined in fluorescence
experiments. The protein–probe mixtures are excited at 344 nm
(isosbestic point) and the probe emission was monitored at
increasing concentrations of the protein. The ratio of initial
intensity (I0) to that in the presence of BSA (I) is plotted as a
function of BSA concentration (Fig. 2). The fluorescence
burial of Py- -Phe at domain II, subdomain C of BSA at
residues 346 and 347, was indicated from the photocleavage
experiments from this laboratory, and current results are
L
consistent with site specific binding of
D
and
L
isomers of Py-
Phe to BSA.5 The marked differences observed in the spectral
properties of the bound enantiomers are surprising, and they are
primarily due to differences in the residues that line the binding
cavity. These results imply that the location of the pyrenyl
chromophore, the photoactive moiety used for protein cleavage,
is different for the two enantiomers.
intensity of the
concentration, and the ratio, I0/I, reaches a plateau at higher
protein concentrations. The fluorescence from the -isomer, in
D-isomer increases initially with protein
L
contrast, is strongly quenched by BSA, and the corresponding
plot shows an accidental near-mirror image behavior. These
results highlight the differences in the local environment
surrounding the fluorophore when the two isomers bind to
BSA.
The accessibility of the two isomers to the solvent was probed
using the fluorescence quencher, hexamminecobalt(III) chlo-
ride. The extent of protection offered by the protein can be
readily distinguished in these simple experiments. The emission
Such differences in the binding environment will be exploited
in photocleavage experiments to direct the photochemical
reagents to different sites on proteins. The large differences in
the spectral properties of Py-Phe isomers, in addition, provide a
signal transduction mechanism for the translation of the chiral
recognition information into an easily measurable spectroscopic
quantity for application in chiral biosensors.
Financial support of this work by the National Science
Foundation (DMR-9729178), and the sponsors of the Petroleum
Research Fund, are gratefully acknowledged.
from the BSA-bound
CoHA (Fig. 3) whereas for the
D
-isomer was enhanced by the addition of
L
-isomer the emission was
quenched, confirming the differences in the binding behavior of
the two optical antipodes shown in Figs. 1 and 2.
Notes and references
1 C. J. Suckling, Enzyme Chemistry: Impact and Applications, Chapman
and Hall, London, 2nd edn., 1990; A. Lehninger, D. L. Nelson and
M. M. Cox, Principles of Biochemistry, Worth, New York, 1993; F. H.
Westheimer, H. Fisher, E. E. Conn and B. Vennesland, J. Am. Chem.
Soc., 1951, 73, 2043.
2 Y. Abe, S. Fukui, Y. Koshiji, M. Kobayashi, T. Shoji, S. Sugata, H.
Nishizawa, H. Suzuki and K. Iwata, Biochim. Biophys. Acta, 1999,
1433, 188; A. Haque and J. T. Stewart, J. Liq. Chromatogr. Relat.
Technol., 1998, 21, 2675; C. J. Stefan, M. Ubbink, G. W. Canters and
H. P. J. M. Dekkers, J. Phys. Chem., 1996, 100, 17 957; G. Felix and V.
Descorps, Chromatographia, 1999, 49, 595; Y. Yan and M. L. Myrick,
Anal. Chem., 1999, 71, 1958.
3 S. Allenmark and S. Andersson, Chirality, 1992, 4, 24; T. Cserhati and
E. Forgacs, Int. J. Bio-Chromatogr., 1999, 4, 203; L. Lepri, V. Coas and
M. Del Bubba, J. Planar, Chromatogr.-Mod., TLC, 1999, 12, 221; T.
Kitae, T. Nakayama and K. Kano, J. Chem. Soc., Perkin Trans. 2, 1998,
207.
Fig. 3 Chiral selectivity for the quenching of fluorescence emission from the
Py-D-Phe (open circles) and Py-L-Phe (filled circles) bound to bovine serum
albumin by hexamminecobalt(III) chloride.
4 K. B. Lipkowitz, Acc. Chem. Res., 2000, 33, 555.
The large difference in the affinities of the optical antipodes
arising due to a single chiral center is unexpected. For example,
the binding of dansyl labeled amino acids to BSA shows only a
5 (a) A. Buranaprapuk, C. V. Kumar, S. Jockusch and N. J. Turro,
Tetrahedron, 2000, 56, 8311; (b) C. V. Kumar and A. Buranaprapuk,
J. Am. Chem. Soc., 1999, 121, 4262; (c) C. V. Kumar, A. Buranaprapuk,
G. J. Opiteck, M. B. Moyer, S. Jockusch and N. J. Turro, Pro. Natl.
Acad. Sci., 1998, 95, 10361; (d) C. V. Kumar and A. Buranaprapuk,
Angew. Chem., Int. Ed. Engl., 1997, 36, 2085.
3+1 discrimination between the corresponding
D
- and
L
-
isomers.2 Treatment of racemic potassium tris(oxalato)cobalta-
te(III) with BSA resulted in the enrichment of the D isomer with
an enantiomeric excess of 18%6 and only minor differences
were reported for the binding of the ketoprofen enantiomers to
BSA.7 The first report on the chiral recognition by a de novo
6 T. Taura, Inorg. Chim. Acta, 1996, 252, 1.
7 M. Levi and M. Zandomeneghi, Gazz. Chim. Ital., 1966, 126, 599.
8 K. S. Broo, H. Nilsson, J. Nilsson and L. Baltzer, J. Am. Chem. Soc.,
1988, 120, 10287.
9 G. Massolini, A. F. Aubry, A. McGann and I. W. Wainer, Biochem.
Pharmacol., 1993, 46, 1285.
10 (a) D. S. Sampath and P. Balaram, Biochim. Biophys. Acta, 1986, 882,
183; (b) A. Higuchi, T. Hashimoto, M. Yonehara, N. Kubota, K.
Wanatabe, S. Uemiya, T. Kojima and M. Hara, J. Membr. Sci., 1997,
130, 31.
designed peptide using -norleucine derivative was reported to
D
be 2+1,8 and (R)-warfarin was retained to a greater extent on an
immobilized BSA column than (S)-warfarin9 while no enantio-
selectivity was observed for the binding of gossypol to BSA.10
The
D-amino acids, on a liquid chromatography column with
immobilized BSA as the chiral phase, eluted at different times,
298
Chem. Commun., 2001, 297–298
Typeset and printed by Black Bear Press Limited, Cambridge, England