View Article Online / Journal Homepage / Table of Contents for this issue
1859
ANALYST, DECEMBER 1992, VOL. 117
Published on 01 January 1992. Downloaded by CASE WESTERN RESERVE UNIVERSITY on 22/10/2014 17:23:24.
High-performance Liquid Chromatographic Determination of Fatty Acid Binding Proteins in Rat Liver With Fluorescence Detection Masatoshi Yarnaguchi, Kouichi Wada, Junichi lshida and Masaru Nakamura Faculty of Pharmaceutical Sciences, Fukuoka University, Nanakuma, Johnan-ku, Fukuoka 814-01, Japan
A highly sensitive, simple and reproducible method for the quantitative determination of fatty acid binding protein in rat liver by post-column high-performance liquid chromatography with fluorescence detection is presented. Fatty acid binding protein in rat liver cytosols is separated by gel-permeation column chromatography, which is followed by fluorescence detection. The detection makes use of the fluorescence enhancement observed when a fluorescent fatty acid probe, dansylundecanoic acid, binds to fatty acid binding protein. The method is rapid, simple to perform and highly sensitive. The method was applied to the determination of fatty acid binding protein in liver from control and hypolipidaemic drug treated rats. Keywords: Fatty acid binding protein; rat liver; post-column high-performance liquid chromatography; dansylundecanoic acid; fluorescence detection
Fatty acid binding protein (FABP) from rat liver is a protein of relative molecular mass (M,) 14 O00 that binds long-chain fatty acids and their corresponding coenzyme A (CoA) esters, as well as a number of non-polar organic anions,1 with relatively high affinity. It is thought to function in the intracellular transport and compartmentalization of long chain fatty acids, or it may protect specific enzymes against inhibition by long chain acyl-CoA esters;lJ however, the precise physiological role of this protein is still not known. This might be partially owing to the lack of a sensitive, simple and reproducible method. Radiochemical methods have been widely used for the detection and quantification of this protein in rat liver cytosols. Other methods for the separation of protein-bound and unbound labelled fatty acids include gel-permeation chromatography, affinity chromatography, electrophoretic techniques and modified equilibrium dialysis and/or their combinations.3-8 All these methods, however, require a radioactive compound and, therefore, subsequent studies on the protein are difficult to perform. A fluorimetric method for the detection and estimation of FABP in rat liver has been reported.9 In this method, FABP in rat liver cytosols was separated by conventional (or highperformance liquid) gel-permeation chromatography. Each fraction from the column was collected and detection was carried out fluorimetrically using dansylundecanoic acid as the fluorescent fatty acid probe for FABP; the probe has a high affinity for binding to FABP and shows considerable fluorescence enhancement. In this paper a highly sensitive, simple, rapid and reproducible high-performance liquid chromatographic (HPLC) method for the quantification of FABP in rat liver cytosols is described. The FABP was separated by gel-permeation HPLC and was determined fluorimetrically by post-column reaction with dansylundecanoic acid.
Female Wister Albino rats, weighing 150-200 g, were used throughout this study. Rat liver cytosols were prepared by the method of Wilkinson and Wilton.9 The concentration of protein in the cytosolic fraction was determined by the Biuret method with bovine serum albumin as standard.10 Purified FABP was prepared by the method of Wilkinson and Wilton.9 The concentration of protein in purified FABP preparations was determined by the spectrophotometric method of Whitaker and Granum.11
Procedure The rat liver cytosolic fraction was injected directly into the HPLC-fluorescence detection system, as shown in Fig. 1. Chromatography was performed with a CCPM high-performance liquid chromatograph (Tosoh, Tokyo, Japan) equipped with a Rheodyne Model 7125 syringe-loading sample injection valve (20 yl loop). The FABP was separated on a Tosoh TSK gel G2000SW XL column (300 X 7.8 mm i.d.), which was preceded by a TSK SW XL guard column (40 x 6 mm i.d.), by isocratic elution with 0.1 moll-' potassium phosphate buffer (pH 7.2) as eluent. The flow rate of the mobile phase was 0.5 ml min-1. The column temperature was ambient (18-37 "C). The eluate from the HPLC column was mixed with the probe solution delivered by a Tosoh CCPE-I1 pump using a T-type mixing device. The flow rate of the solution was 0.05 ml min-1. The fluorescence was monitored by a Hitachi FlOOO fluorescence spectrometer equipped with a 12 yl flow cell operated at an excitation wavelength of 350 nm and an
Column
& I
+
Waste
Experimental Reagents and FABP Samples All chemicals were of analytical-reagent grade unless stated otherwise. Dansylundecanoic acid was purchased from Molecular Probes (Junction City, OR, USA). A stock solution of dansylundecanoic acid (1 x 10-4 mol 1-1) was prepared in methanol and diluted further with 0.1 mol 1-1 potassium phosphate buffer (pH 7.4) to give an 8.0 pmol 1-1 solution. The resulting dansylundecanoic acid solution was used within 3 d of preparation. Clofibric acid and pravastatin sodium were obtained from Sigma (St. Louis, MO, USA) and Sankyo Pharmaceuticals (Tokyo, Japan), respectively.
Fig. 1 Schematic diagram of the HPLC system with fluorescence detection. PI and P2, HPLC pumps (Tosoh CCPM and CCPE-11, respectively); I, injection valve (Rheodyne 7125, 20 PI); D1?UV detector (Hitachi 635M; 280 nm); D2, fluorescence detector (Hitachi F-1000); G, guard column (TSK gel SW XL); Column, TSK gel G2000SW XL (300 x 7.8 mm id.); M, mixing tee; R, reaction coil (F'TFE tube, 5 m x 0.5 mm i.d.); Rec, recorder; Reg; pressure regulator; E, mobile phase [0.1 mol 1-1 potassium phosphate buffer (pH 7.2)]; R, 8 pmol 1-1 dansylundecanoic acid. Flow rates: E, 0.5 ml min-1; R, 0.05 ml min-1
View Article Online
Published on 01 January 1992. Downloaded by CASE WESTERN RESERVE UNIVERSITY on 22/10/2014 17:23:24.
1860
ANALYST, DECEMBER 1992, VOL. 117
emission wavelength of 500 nm. The length of the reaction coil (0.5 mm i d . ) between the fluorescence detector and the column was 5 m. A back-pressure regulator was placed at the fluorescence detector outlet to eliminate out-gassing problems. When FABP was monitored spectrophotometrically at 280 nm, a detector (Hitachi Model 635M;15 pl flow cell) was set between the analytical column and the reaction coil. The FABP content was calibrated using a calibration graph of peak height. The graph was established by passing various amounts of purified FABP through the HPLC system.
peak heights. However, the use of these solvents resulted in evolution of gas in the reaction coil and/or flow cell, and interfered with the determination of FABP. Although 0.1 moll-' potassium phosphate buffer (pH 7.2) gave less intense peak heights (approximately 60% of that for methanol and dimethyl sulfoxide), the solvent did not cause the occurrence of gas. Thus, 8 pmol 1-1 dansylundecanoic acid in the phosphate buffer was used in the procedure. Maximum peak height was attained with a 4-6 m length of tubing coil; a 5 m length was employed. The temperature of the reaction coil (10-37 "C) did not influence the peak height.
Results and Discussion
Chromatography The HPLC conditions were the same as those reported by Wilkinson and Wilton.9 Fig. 2(a) shows a typical chromatogram obtained with rat liver cytosol. For ultraviolet (UV) detection, a small peak due to FABP was observed at a retention time of 21 min. However, many peaks were observed in the chromatogram and hindered the sensitive determination of FABP. On the other hand, a very simple chromatogram was given by fluorescence detection. Peaks 1 and 2 in Fig. 2(a) were identified to be FABP in rat liver cytosols by comparing a chromatogram [Fig. 2(b)] obtained with the purified FABP. Peaks 3 and 4 in Fig. 2(a) might be due to rat hepatic albumin.
Optimization of Fluorescence Detector Response
The optimum conditions were examined using a purified FABP solution (1.2 mg ml-1). Dansylundecanoic acid gave the most intense peak for FABP at probe (dansylundecanoic acid) concentrations in the solution over the range 6 1 0 pmol 1-1. Of the three solvents used for the preparation of the dansylundecanoic acid solution, methanol and dimethyl sulfoxide gave the maximum
Calibration Graph, Precision and Detection Limit
A calibration graph was established by determining various amounts (1.2-40 pg) of purified FABP by HPLC. The relationship between the peak height and the amount of FABP was linear up to at least 40 pg per 20 pl injection volume with a linear correlation coefficient of 0.998. Thus, the proposed method permits the determination of FABP over wide concentration ranges. The detection limit was 0.06 pg of FABP per 20 p1 injection volume with a signal-to-noise ratio of 5. The sensitivity is about 10 times higher than with the conventional fluorimetric met hod .9 The precision was established by repeated determinations (n = 7) using a rat liver cytosol sample (FABP concentration, 32.1 pg mg-1 of cytosol protein). The relative standard deviation was 4.7%. The precision is better than that for the conventional fluorimetric method.9
t
Concentration of Fatty Acid Binding Protein in Liver Cytosol From Control and Drug-treated Rats
The validity of the method was demonstrated by measuring the concentration of FABP in liver from control and hypolipidaemic drug treated rats (Table 1). Clofibric acid and pravastatin sodium, which are both hypolipidaemic drugs, increased the concentration of FABP in rat liver. The concentrations in control and clofibric acid treated rats were in good agreement with those reported by other workers.8.9 The hepatic FABP level in pravastatin sodium treated rats was first determined by the proposed method.
0
10
20
30
40
50
Table 1 Effect of administration of clofibric acid and pravastatin sodium on the concentration of FABP in rat liver cytosols. The rats were fed ad libitum on a commercial diet. Rats of drug-treated groups were fed ad libitum for 7 d before death on a diet containing either 0.5% clofibric acid or 0.5% pravastatin sodium. (Values given are mean f standard deviation; numbers in parentheses are the number of animals used)
Retention timelmin Fig. 2 Chromatograms obtained with ( a ) rat liver cytosolic fraction and ( b ) purified FABP. A portion (20 p1) of the cytosol fraction
(protein, 40 mg ml-l) or the purified FABP solution (FABP, 1.2 mg ml-1) was applied onto the HPLC column. A, UV (280 nm) detection; €3, fluorescence detection (hex = 350 nm; he, = 500 nm). Peaks: 1 and 2, FABP; 3 and 4, albumin; other peaks, unknown
Treatment Control (3) Pravastatin sodium (3) Clofibric acid (3)
FABP content/ pg mg-1 of cytosolic protein 31.5 f 2.2 54.7 5.9 51.0 k 5.8
*
View Article Online
ANALYST, DECEMBER 1992, VOL. 117
Published on 01 January 1992. Downloaded by CASE WESTERN RESERVE UNIVERSITY on 22/10/2014 17:23:24.
This work provides the first fluorimetric HPLC method for the quantification of FABP in rat liver cytosols. This method is sufficiently sensitive to determine 0.06 pg of hepatic FABP. The method is rapid and simple to perform and can, therefore, be applied to routine analyses in biological and biomedical investigations. This research was partly supported by a Grant-in-Aid for Scientific Research (No. 04671337) from the Ministry of Education, Science and Culture of Japan. The authors are grateful to Dr. S. Yamano €or his helpful suggestions and K. Abe for his skilful assistance.
References 1 Glatz, J. F. C., and Veerkamp, J. H., Znt. J . Biochem., 1985,17, 13. 2 Grinstead, G. F., Trzaskos, J. M., Billheimer, J. T., and Gaylor, J. L., Biochim. Biophys. Acta, 1983, 751, 41.
1861 3 Morrow, F. D., and Martin, R. J., J. Lipid Res., 1983, 24,324. 4 Glatz, J. F. C., and Veerkamp, J. H., Anal. Biochem., 1983, 132,89. 5 Glatz, J. F. C., Baerwaldt, C. C. F., Veerkamp, J. H., and Kempen, H. J. M., J . Biol. Chem., 1984,259,4259. 6 Smith, F. M. M., Dils, R. R., and Gurr, M. I., Biochem. Soc. Trans., 1983, 11, 308. 7 Keuper, H. J . K., Klein, R. A., and Spener, F., Chem. Phys. Lipids, 1983,32, 153. 8 Kawashima, Y., Nakagawa, S., Tachibana, Y . , and Kozulka, H., Biochim. Biophys. Acta, 1983, 754,21. 9 Wilkinson, T. C. I., and Wilton, D. C., Biochem. J., 1986,238, 419. 10 Gornall, A. G., Bardawill, F. J., and David, M. M., J. Biol. Chem., 1949, 177, 751. 11 Whitaker, J. R., and Granum, P. E., Anal. Biochem., 1980, 109, 156.
Paper 2t02440C Received May 12, 1992 Accepted July 2, 1992
1859
ANALYST, DECEMBER 1992, VOL. 117
Published on 01 January 1992. Downloaded by CASE WESTERN RESERVE UNIVERSITY on 22/10/2014 17:23:24.
High-performance Liquid Chromatographic Determination of Fatty Acid Binding Proteins in Rat Liver With Fluorescence Detection Masatoshi Yarnaguchi, Kouichi Wada, Junichi lshida and Masaru Nakamura Faculty of Pharmaceutical Sciences, Fukuoka University, Nanakuma, Johnan-ku, Fukuoka 814-01, Japan
A highly sensitive, simple and reproducible method for the quantitative determination of fatty acid binding protein in rat liver by post-column high-performance liquid chromatography with fluorescence detection is presented. Fatty acid binding protein in rat liver cytosols is separated by gel-permeation column chromatography, which is followed by fluorescence detection. The detection makes use of the fluorescence enhancement observed when a fluorescent fatty acid probe, dansylundecanoic acid, binds to fatty acid binding protein. The method is rapid, simple to perform and highly sensitive. The method was applied to the determination of fatty acid binding protein in liver from control and hypolipidaemic drug treated rats. Keywords: Fatty acid binding protein; rat liver; post-column high-performance liquid chromatography; dansylundecanoic acid; fluorescence detection
Fatty acid binding protein (FABP) from rat liver is a protein of relative molecular mass (M,) 14 O00 that binds long-chain fatty acids and their corresponding coenzyme A (CoA) esters, as well as a number of non-polar organic anions,1 with relatively high affinity. It is thought to function in the intracellular transport and compartmentalization of long chain fatty acids, or it may protect specific enzymes against inhibition by long chain acyl-CoA esters;lJ however, the precise physiological role of this protein is still not known. This might be partially owing to the lack of a sensitive, simple and reproducible method. Radiochemical methods have been widely used for the detection and quantification of this protein in rat liver cytosols. Other methods for the separation of protein-bound and unbound labelled fatty acids include gel-permeation chromatography, affinity chromatography, electrophoretic techniques and modified equilibrium dialysis and/or their combinations.3-8 All these methods, however, require a radioactive compound and, therefore, subsequent studies on the protein are difficult to perform. A fluorimetric method for the detection and estimation of FABP in rat liver has been reported.9 In this method, FABP in rat liver cytosols was separated by conventional (or highperformance liquid) gel-permeation chromatography. Each fraction from the column was collected and detection was carried out fluorimetrically using dansylundecanoic acid as the fluorescent fatty acid probe for FABP; the probe has a high affinity for binding to FABP and shows considerable fluorescence enhancement. In this paper a highly sensitive, simple, rapid and reproducible high-performance liquid chromatographic (HPLC) method for the quantification of FABP in rat liver cytosols is described. The FABP was separated by gel-permeation HPLC and was determined fluorimetrically by post-column reaction with dansylundecanoic acid.
Female Wister Albino rats, weighing 150-200 g, were used throughout this study. Rat liver cytosols were prepared by the method of Wilkinson and Wilton.9 The concentration of protein in the cytosolic fraction was determined by the Biuret method with bovine serum albumin as standard.10 Purified FABP was prepared by the method of Wilkinson and Wilton.9 The concentration of protein in purified FABP preparations was determined by the spectrophotometric method of Whitaker and Granum.11
Procedure The rat liver cytosolic fraction was injected directly into the HPLC-fluorescence detection system, as shown in Fig. 1. Chromatography was performed with a CCPM high-performance liquid chromatograph (Tosoh, Tokyo, Japan) equipped with a Rheodyne Model 7125 syringe-loading sample injection valve (20 yl loop). The FABP was separated on a Tosoh TSK gel G2000SW XL column (300 X 7.8 mm i.d.), which was preceded by a TSK SW XL guard column (40 x 6 mm i.d.), by isocratic elution with 0.1 moll-' potassium phosphate buffer (pH 7.2) as eluent. The flow rate of the mobile phase was 0.5 ml min-1. The column temperature was ambient (18-37 "C). The eluate from the HPLC column was mixed with the probe solution delivered by a Tosoh CCPE-I1 pump using a T-type mixing device. The flow rate of the solution was 0.05 ml min-1. The fluorescence was monitored by a Hitachi FlOOO fluorescence spectrometer equipped with a 12 yl flow cell operated at an excitation wavelength of 350 nm and an
Column
& I
+
Waste
Experimental Reagents and FABP Samples All chemicals were of analytical-reagent grade unless stated otherwise. Dansylundecanoic acid was purchased from Molecular Probes (Junction City, OR, USA). A stock solution of dansylundecanoic acid (1 x 10-4 mol 1-1) was prepared in methanol and diluted further with 0.1 mol 1-1 potassium phosphate buffer (pH 7.4) to give an 8.0 pmol 1-1 solution. The resulting dansylundecanoic acid solution was used within 3 d of preparation. Clofibric acid and pravastatin sodium were obtained from Sigma (St. Louis, MO, USA) and Sankyo Pharmaceuticals (Tokyo, Japan), respectively.
Fig. 1 Schematic diagram of the HPLC system with fluorescence detection. PI and P2, HPLC pumps (Tosoh CCPM and CCPE-11, respectively); I, injection valve (Rheodyne 7125, 20 PI); D1?UV detector (Hitachi 635M; 280 nm); D2, fluorescence detector (Hitachi F-1000); G, guard column (TSK gel SW XL); Column, TSK gel G2000SW XL (300 x 7.8 mm id.); M, mixing tee; R, reaction coil (F'TFE tube, 5 m x 0.5 mm i.d.); Rec, recorder; Reg; pressure regulator; E, mobile phase [0.1 mol 1-1 potassium phosphate buffer (pH 7.2)]; R, 8 pmol 1-1 dansylundecanoic acid. Flow rates: E, 0.5 ml min-1; R, 0.05 ml min-1
View Article Online
Published on 01 January 1992. Downloaded by CASE WESTERN RESERVE UNIVERSITY on 22/10/2014 17:23:24.
1860
ANALYST, DECEMBER 1992, VOL. 117
emission wavelength of 500 nm. The length of the reaction coil (0.5 mm i d . ) between the fluorescence detector and the column was 5 m. A back-pressure regulator was placed at the fluorescence detector outlet to eliminate out-gassing problems. When FABP was monitored spectrophotometrically at 280 nm, a detector (Hitachi Model 635M;15 pl flow cell) was set between the analytical column and the reaction coil. The FABP content was calibrated using a calibration graph of peak height. The graph was established by passing various amounts of purified FABP through the HPLC system.
peak heights. However, the use of these solvents resulted in evolution of gas in the reaction coil and/or flow cell, and interfered with the determination of FABP. Although 0.1 moll-' potassium phosphate buffer (pH 7.2) gave less intense peak heights (approximately 60% of that for methanol and dimethyl sulfoxide), the solvent did not cause the occurrence of gas. Thus, 8 pmol 1-1 dansylundecanoic acid in the phosphate buffer was used in the procedure. Maximum peak height was attained with a 4-6 m length of tubing coil; a 5 m length was employed. The temperature of the reaction coil (10-37 "C) did not influence the peak height.
Results and Discussion
Chromatography The HPLC conditions were the same as those reported by Wilkinson and Wilton.9 Fig. 2(a) shows a typical chromatogram obtained with rat liver cytosol. For ultraviolet (UV) detection, a small peak due to FABP was observed at a retention time of 21 min. However, many peaks were observed in the chromatogram and hindered the sensitive determination of FABP. On the other hand, a very simple chromatogram was given by fluorescence detection. Peaks 1 and 2 in Fig. 2(a) were identified to be FABP in rat liver cytosols by comparing a chromatogram [Fig. 2(b)] obtained with the purified FABP. Peaks 3 and 4 in Fig. 2(a) might be due to rat hepatic albumin.
Optimization of Fluorescence Detector Response
The optimum conditions were examined using a purified FABP solution (1.2 mg ml-1). Dansylundecanoic acid gave the most intense peak for FABP at probe (dansylundecanoic acid) concentrations in the solution over the range 6 1 0 pmol 1-1. Of the three solvents used for the preparation of the dansylundecanoic acid solution, methanol and dimethyl sulfoxide gave the maximum
Calibration Graph, Precision and Detection Limit
A calibration graph was established by determining various amounts (1.2-40 pg) of purified FABP by HPLC. The relationship between the peak height and the amount of FABP was linear up to at least 40 pg per 20 pl injection volume with a linear correlation coefficient of 0.998. Thus, the proposed method permits the determination of FABP over wide concentration ranges. The detection limit was 0.06 pg of FABP per 20 p1 injection volume with a signal-to-noise ratio of 5. The sensitivity is about 10 times higher than with the conventional fluorimetric met hod .9 The precision was established by repeated determinations (n = 7) using a rat liver cytosol sample (FABP concentration, 32.1 pg mg-1 of cytosol protein). The relative standard deviation was 4.7%. The precision is better than that for the conventional fluorimetric method.9
t
Concentration of Fatty Acid Binding Protein in Liver Cytosol From Control and Drug-treated Rats
The validity of the method was demonstrated by measuring the concentration of FABP in liver from control and hypolipidaemic drug treated rats (Table 1). Clofibric acid and pravastatin sodium, which are both hypolipidaemic drugs, increased the concentration of FABP in rat liver. The concentrations in control and clofibric acid treated rats were in good agreement with those reported by other workers.8.9 The hepatic FABP level in pravastatin sodium treated rats was first determined by the proposed method.
0
10
20
30
40
50
Table 1 Effect of administration of clofibric acid and pravastatin sodium on the concentration of FABP in rat liver cytosols. The rats were fed ad libitum on a commercial diet. Rats of drug-treated groups were fed ad libitum for 7 d before death on a diet containing either 0.5% clofibric acid or 0.5% pravastatin sodium. (Values given are mean f standard deviation; numbers in parentheses are the number of animals used)
Retention timelmin Fig. 2 Chromatograms obtained with ( a ) rat liver cytosolic fraction and ( b ) purified FABP. A portion (20 p1) of the cytosol fraction
(protein, 40 mg ml-l) or the purified FABP solution (FABP, 1.2 mg ml-1) was applied onto the HPLC column. A, UV (280 nm) detection; €3, fluorescence detection (hex = 350 nm; he, = 500 nm). Peaks: 1 and 2, FABP; 3 and 4, albumin; other peaks, unknown
Treatment Control (3) Pravastatin sodium (3) Clofibric acid (3)
FABP content/ pg mg-1 of cytosolic protein 31.5 f 2.2 54.7 5.9 51.0 k 5.8
*
View Article Online
ANALYST, DECEMBER 1992, VOL. 117
Published on 01 January 1992. Downloaded by CASE WESTERN RESERVE UNIVERSITY on 22/10/2014 17:23:24.
This work provides the first fluorimetric HPLC method for the quantification of FABP in rat liver cytosols. This method is sufficiently sensitive to determine 0.06 pg of hepatic FABP. The method is rapid and simple to perform and can, therefore, be applied to routine analyses in biological and biomedical investigations. This research was partly supported by a Grant-in-Aid for Scientific Research (No. 04671337) from the Ministry of Education, Science and Culture of Japan. The authors are grateful to Dr. S. Yamano €or his helpful suggestions and K. Abe for his skilful assistance.
References 1 Glatz, J. F. C., and Veerkamp, J. H., Znt. J . Biochem., 1985,17, 13. 2 Grinstead, G. F., Trzaskos, J. M., Billheimer, J. T., and Gaylor, J. L., Biochim. Biophys. Acta, 1983, 751, 41.
1861 3 Morrow, F. D., and Martin, R. J., J. Lipid Res., 1983, 24,324. 4 Glatz, J. F. C., and Veerkamp, J. H., Anal. Biochem., 1983, 132,89. 5 Glatz, J. F. C., Baerwaldt, C. C. F., Veerkamp, J. H., and Kempen, H. J. M., J . Biol. Chem., 1984,259,4259. 6 Smith, F. M. M., Dils, R. R., and Gurr, M. I., Biochem. Soc. Trans., 1983, 11, 308. 7 Keuper, H. J . K., Klein, R. A., and Spener, F., Chem. Phys. Lipids, 1983,32, 153. 8 Kawashima, Y., Nakagawa, S., Tachibana, Y . , and Kozulka, H., Biochim. Biophys. Acta, 1983, 754,21. 9 Wilkinson, T. C. I., and Wilton, D. C., Biochem. J., 1986,238, 419. 10 Gornall, A. G., Bardawill, F. J., and David, M. M., J. Biol. Chem., 1949, 177, 751. 11 Whitaker, J. R., and Granum, P. E., Anal. Biochem., 1980, 109, 156.
Paper 2t02440C Received May 12, 1992 Accepted July 2, 1992
