albumin (BSA) and with several other serum albumins. Materials and Methods: All serum albumin preparations were obtained from Sigma. Chemical Co. ,USA.
Vol. 43, No. 1, September 1997 BIOCHEMISTRYand MOLECULAR BIOLOGY INTERNATIONAL Pages 1-8
INTERACTION OF BROMOCRESOL GREEN W I T H DIFFERENT SERUM ALBUMINS STUDIED BY FLUORESCENCE QUENCHING V.D.Trivedi*, I. Saxena, M.U. Siddiqui, and M.A.Qasim Department of Biochemistry, LN.Medical College, A.M.U., Aligarh-202 002, India Received February 24, 1997
Received after revisionMay 6, 1997 Summary: The binding of bromocresol green to bovine serum albumin at micromolar concentrations leads to quenching of protein fluorescence. This property has been used here to study interaction of bromocresol green with bovine serum albumin as a function of pH and ionic strength. The transformation of fluorescence quench data obtained with bromocresol green into Scatchard plots yielded an association constant of 3.06X107 1M-1 and a binding capacity of about 1.0. The affinity of bromocresol green for bovine serum albumin remains virtually unchanged between pH 4.0 and 8.0 but decreases by about 7 fold with increase in ionic strength from 0.01 to 1.0. Six other serum albumins obtained from cat, dog, human, pig and sheep have also been studied for bromocresol green binding. Although all the albumins studied bind bromocresol green, they show considerable differences in their affinities towards the dye. It appears that despite a great degree of overall similarity in their structure and conformation, serum albumins from different species differ in their ligand binding properties. Introduction: A wide variety of exogenous and endogenous compounds are known to bind to serum albumin (1,2). This binding, in the case of endogenous compounds and for various therapeutic drugs, may be critical in the solubilization of these ligands (3), in making them non-toxic (2,4) and in their pharmacokinetic distribution (2,5). Several studies on aspects such as the molecular basis of ligand-albumin interactions (3,6-10), displacement of one ligand from its binding site on albumin by another (11,12) and structure and conformation of binding sites for these ligands stress the importance of albumin-ligand interactions. Although these interactions have been studied with diverse aims, the central problem in these interactions is the localization of the binding site in the protein and the molecular basis of binding. Of the various exogenous compounds that binds to albumin, bromocresol green (BCG) and a few other similar dyes (13-14) are of special interest because of their use in the quantitative estimation of albumin in sera. The molecular basis of this interaction is uncertain, though several reports suggest (6,15) that it binds at the primary bilirubin binding
Present address of MAQ: Department of Chemistry, Purdue University, West Lafayette, IN47907, USA. *Present address and correspondence: W-213, Centre for Cellular and Molecular Biology, Uppal Road, Hyderabad, AP-500 007 India, Fax -0091-40-671 195 1039-9712/97/010001-08505.00/0 Copyright 9 1997 by Academic Press Australia. All rights of reproduction in any form reserved.
Vol. 43, No. 1, 1997
BIOCHEMISTRY and MOLECULAR BIOLOGY INTERNATIONAL
site of human serum albumin (HSA). In this communication we describe a fluorescence quenching method for studying the interaction of bromocresol green with bovine serum albumin (BSA) and with several other serum albumins. Materials and Methods: All serum albumin preparations were obtained from Sigma Chemical Co. ,USA. BCG was a product of BDH, Poole, England. Other chemicals were of analytical grade. The concentration of various albumin preparations was determined by the method of l o w r y et al. (16). An extinction coefficient of 6.67 for a 1% solution of BSA at 279 nm was used to estimate the concentration of BSA sample (17). Light absorption measurements were made on a Shimadzu double beam spectrophotometer, UV-150-02. Binding of BCG to serum albumin: Fluorescence quench titration was employed to study the binding of BCG to different serum albumin preparations. The method is based on the observation that the addition of BCG, at micro molar concentrations, to a serum albumin preparation causes quenching of protein fluorescence, the extent of which depends both on protein concentration as well as on the dye concentration. Fluorescence measurements were made on a Shimadzu spectrofluoro photometer, model RF-540, equipped with a data recorder, DR-3. The other accessories attached with the fluorometer were a water jacketed cell holder equipped with a mini magnetic stirrer and a constant temperature water circulator (TB-85). All fluorescence measurements were made at 28~ with constant stirring of the solution. The fluorescence quench titration was performed by taking a constant concentration of albumin solution (2.5/zM) in a series of test tubes and adding increasing concentrations (0-25 tzM) of BCG. Fluorescence measurements were performed after 15 minutes of incubation of albumin with BCG. The titration curves were analysed in the same ways as described earlier for bilirubin-albumin interaction (18,19). The binding constant, Ka, was obtained by transforming the titration curve into Scatchard equation: n Ka- Ka Q = Q/[D] = Q/{R-Q} {albumin}T
(1)
where {albumin}T is the total albumin concentration, and R is the molar ratio between dye and total albumin at any point in the titration curve. [D] is the free dye concentration and Q is the fractional quench. The interaction of BCG with BSA was studied at three pH values pH 4.0, 6.0 and 8.0, at an ionic strength of 0.15. The procedure was the same as described above. The buffers used were 150 mM sodium acetate buffer, pH 4.0; 120 mM sodium phosphate buffer, pH 6.0; and 52 mM sodium phosphate buffer, pH 8.0. The effect of ionic strength was studied in 9 mM sodium phosphate buffer (pH 6.0) by including the requisite amounts of NaC1 in the buffer. Results and Discussion: The fluorescence emission spectra of BSA in the presence of BCG are shown in Fig. 1. The emission spectra clearly show that BCG acts as a strong quencher. The titration curve obtained from these data is shown in the inset of Fig. 1. This quenching cannot be attributed to dynamic quenching, firstly because dynamic quenching is usually observed at a much higher concentration of the quencher than used in this study and because no quenching was observed when BSA and BCG were mixed'in the presence of 9M urea. It is believed that BCG binds at a number of sites on BSA (15). However, the extrapolation of the linear plot of the titration curve gives a stoichiometry of about 1, suggesting that a
BIOCHEMISTRY and MOLECULAR BIOLOGY INTERNATIONAL
Vol. 43, NO. 1, 1997
10( ~0
13
8( iO
tn
c
t~
6(
,0
C
!0
r u
0
,7
2
t,
6
8
10
Molar Ratio~ BCGIBSA
2C
300 Wavelr
400 h (nm)
Fig. 1. Fluorescence emission spectra of BSA in the absence and in the presence of different concentrations of BCG in phosphate buffer, pH 6.0, I 0.15 at 28~ The excitation wavelength was 282 nm. The emission spectra from top to bottom were obtained at BCG/BSA molar ratio of 0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0, 2.0, 3.0, 4.0, 5.0, 10.0. The inset shows a plot of percent fluorescence against BCG/BSA molar ratio. The protein concentration was 2.5/zM.
single strong binding site on BSA is involved during the titration. It, therefore, appears that the affinity of BCG for other sites is several times weaker and hence no appreciable binding occurs at these sites until the strong binding site is occupied. The titration data were transformed in the Scatchard equation (eqn. 1) by the method of Levine (1977). The plot between Q/[D] and [Q] is shown in Fig.2. The linear least squares analysis of this plot gives an association constant of 3.06X107 IM -1 and a binding capacity (n) of 0.94. Thus the strength of binding of BCG is nearly the same as that of bilirubin at its primary binding site on BSA and HSA (15, 20).
Vol. 43, No. 1, 1997
BIOCHEMISTRYand MOLECULAR BIOLOGY INTERNATIONAL
q,D
0
6
O x
"" O"
4
~./z~
I 0.7
I 0.8
I 0.9
Q
Fig.2. Scatchard plot for the interaction of BCG with BSA at pH 6.0, I 0.15 at 28~ Scatchard plot was obtained from the data of Fig. 1. by using equation 1.
The
The Ka values for the BSA-BCG interaction at various pH values (at a contstant ionic strength of 0.15) and at various ionic strengths 0.01, 0.05, 0.15, 0.5 and 1.0 in 9 mM sodium phosphate buffer (pH 6.0) are listed in Table I. The Ka values were determined at least three times. The maximum deviation from average value was + 10%. With this degree of error, it can be said that the Ka values between pH 4.0 and 8.0 are nearly constant. Thus the well-known pH induced N ~ F transition (1) is without any significant effect on the binding BCG with BSA. In contrast to pH, change in ionic strength produces significant changes in Ka. Increase in ionic strength from 0.01 to 1.0 resulted in a 7 fold decrease in the Ka (Table I). The decrease in Ka with increasing ionic strength clearly suggests that electrostatic interactions are important in the binding of BCG to BSA. The interaction of BCG with six other serum albumins namely cat serum albumin (CSA), dog serum albumin (DSA), goat serum albumin (GSA), human serum albumin (HSA), pig serum albumin (PSA) and sheep serum albumin (SSA) was studied in 120 mM sodium phosphate buffer, pH 6.0, ionic strength 0.15. The titration curves are shown in Fig.3 and the values of Ka obtained by Scatchard analysis are depicted in Table II. Of the seven serum albumin preparations used in this study, BSA hadthe highest Ka and CSA and DSA had about 70 fold lower values of Ka. GSA was the only other serum albumin with a Ka value comparable with BSA. Clearly, serum albumins from different sources, despite
Vol. 43, No. 1, 1997
BIOCHEMISTRY and MOLECULAR BIOLOGY INTERNATIONAL
Table I Effect of pH and ionic strength on the binding affinity of BCG with BSA
Conditions
Ka* (1M-1) X 10-7
(a) Ionic strength 0.15, T e m p 28~ pH pH pH (b)
4.0 6.0 8.0 pH 6.0, T e m p 28~
I 0.01 I 0.05 I 0.15 I 0.5 I 1.0
2.29 3.06 2.65
5.03 4.09 3.06 1.62 0.72
D
*Ka values were obtained from fluorescence quench titration data as described in the text.
being similar in structure and conformation, appear to differ vastly in their ligand binding properties. This is an important aspect and should be taken into consideration while interpreting interactions of ligand with different serum albumins in terms of their interactions with HSA. There is some dispute about the location of the primary BCG binding site on serum albumin. It has been proposed that the BCG binding and the primary bilirubin binding sites are common (15). However, some other investigations suggest that the two sites are different. Bush and Reed (14) have found that covalent linkage of bilirubin to HSA decreases bromocresol purple (BCP) binding to HSA but not BCG. Similarly Maguire and Price (21) observed an underestimation of HSA in patients of renal insufficiency by BCP method, but a nearly normal estimate with the BCG method. It has been shown earlier that the binding of bilirubin is 2-3 times stronger to HSA as compared to its binding to BSA (4). Our present studies show that the binding of BCG to BSA is - 20 times stronger than its binding to HSA. These results taken together, show that BCG and bilirubin bind to different sites on BSA, in accordance with the data of Bush and Reed (14) and Maguire and Price (21). These differences in the binding of bilirubin and BCG to BSA and HSA may be explained as follows: The primary structure of the proposed strong bilirubin binding site in
BIOCHEMISTRY and MOLECULAR BIOLOGY INTERNATIONAL
Vol. 43, No. 1, 1997
100i
80
60
u tu
O
40
It o~
20
I
I
I
2
4
6
--
~
--
8
$10
Molar Ratio~ BCG/ALI~UMIN
Fig.3. BCG-albumin interaction isotherms, obtained from the fluorescence quench data at pH 6.0, I 0.15 at 28~ Various albumin preparations were: bovine (~---~) , goat (o---~), sheep ( ~ - - e ) , pig ( ~ - - o ) , human (~---~), dog ( ~ ~), cat (~-----a) serum albumin. The albumin concentration was 2.5 ~M.
Vol. 43, No. 1, 1997
BIOCHEMISTRYond MOLECULAR BIOLOGY INTERNATIONAL
Table II
Interaction of BCG with different serum albumins
Serum albumins
Ka* (1M1) X 10-7
Bovine Cat Dog Goat Human Pig Sheep
3.06 0.05 0.06 2.91 0.16 0.57 0.68
*Ka values were obtained in 120 mM phosphate buffer, pH 6.0, by fluorescence quench titration.
BSA
[residues 186-238 (1)] and HSA upon alignment shows remarkable similarity in
sequence. If K--,R and S--,T substitutions are overlooked, then there are differences at only 8 of the 52 amino acids present in this region. Of these 8, only four residues located at the N terminal of the bilirubin binding site show substitutions involving differences in charge and hydrophobicity. With this high degree of homology the nearly identical bilirubin binding affinity of these two proteins is understandable. Assuming that the BCG binding site in the two albumins is superimposable with the bilirubin binding site, BCG would be expected to bind BSA and HSA with a similar binding affinity. The fact that BCG binds to BSA with a binding constant 20 times stronger than that of HSA shows that the primary bilirubin and BCG binding sites are different. The fluorescence quenching method described here can be of general application in studying and quantitating the interaction of a number of structurally related dyes with albumins. Secondly, the interaction of BCG with different serum albumins clearly shows that despite a great degree of overall structural and conformational similarity in different albumins (10), there are considerable differences among them, presumably at the subdomain level and/or in the ligand binding regions. Finally, our results suggest that the BCG binding site and the primary bilirubin binding site on BSA are not identical. Acknowledgements: This work was supported by research from Council of Scientific and Industrial Research, New Delhi. Facilities were provided by Aligarh M. University. VDT wishes to thank Dr. T. Ramakrishna Murti, B. Raman and Sushil Chandani for discussions and critical comments.
Vol. 43, No. 1, 1997
BIOCHEMISTRYand MOLECULAR BIOLOGY INTERNATIONAL
References: 1. Carter, D.C., and Ho, J.X. (1994) Adv. Protein Chem. 45, 153-203. 2. Peters, T. Jr. (1996) All about albumin: Biochemistry, Genetics and Medical applications, Academic Press, New York. 3. Peters, T.,Jr. and Reed, R.G. (1977) proceeding of the 1lth FEBS meeting, pp. 11-20. 4. Brodersen, R. (1979) Crit. Rev. Clin. Lab. Sci. 11,305-399. 5. Sjoholm, I., Ekman, B., Kober, A., Ljungstedt-Pahlman, I., Seiving, B., and Sjodin, T. (1979) Mol. Pharmacol. 16, 767-777. 6. Kragh-Hansen, U. (1981) Pharmacological Rev. 33,17-52. 7. Sudlow, G, Birkett, D.J., and Wode, B.N. (1976) Mol. Pharmacol. 12, 1052-1061. 8. Lightner, D.A., Reisingere, M., and Landen, G.L. (1986) J. Biol. Chem. 261, 60346038. 9. Bos, O. J. M., Fischer, M.J.E., Wilting, J., and Janssen, L.H.M. (1988) Biochim. Biophys. Acta 953, 37-47. 10. He, X.M., and Carter, D.C. (1992) Nature 358, 209-215. 11. Brodersen, R., and Ebbesen, F. (1983) J. Pharm. Sci. 72, 248-253. 12. Brodersen, R., Friis-Hansen, B., and Stem, L. (1983) Dev. Pharmacol. Ther. 6, 217229. 13. Hill, P.G., and wells, T.N.C. (1983) Ann. Clin. Biochem. 20, 264-270. 14. Bush, V., and Reed, R.G. (1987) Clin. Chem. 33, 821-823. 15. Reed, R.G., Feldhoff, R.C., Clute, O.L., and Peters. T. Jr. (1975) Biochemistry 14, 4578-4583. 16. Lowry, O.H., Rosebrough, N.J., Farr, A.L., and Randall, R.J. (1951) 1. Biol. Chem. 193, 265-275. 17. Janatova, J., Fuller, J.K., and Hunter, M.J. (1968) J. Biol. Chem. 243,3612-3622. 18. Levine, R.L. (1977) Clin. Chem. 23, 2292-2301. 19, Tayyab, S., and Trivedi, V.D. (1995) Biochem. Edu. 23, 98-101. 20. Berde, C.B., Hudson, B.S., Simoni, R.D., and Skaler, L.A. (1979) J. Biol. Chem. 254, 391-400. 21. Maguire, G. A., and Price, C.P. (1986) Clin. Chim. Acta 155, 83-88.
