ARCHIVES
OF BIOCHEMISTRY
AND
BIOPHYSICS
Vol. 298, No. 1, October, pp. 254-258, 1992
An in vitro Reversible Interconversion of Rat Liver Fatty Acid Binding Protein Having Different Isoelectric Points by Virtue of the Fatty Acid Content’ Ming Li and Teruo Ishibashi2 Department
of Biochemistry,
Hokkaido
University
School of Medicine,
Sapporo 060, Japan
Received March 2, 1992, and in revised form June 11, 1992
Rat liver fatty acid binding protein (FABP) was purified to homogeneity by procedures including Sephadex G- 100 and DEAE-cellulose column chromatographies. FABP was resolved into two major peaks, A and B, by the first DEAE-cellulose column chromatography. Each of these two fractions exhibited apparent homogeneity upon polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate with a molecular weight of 14,000 Da and amino acid analysis of these fractions has revealed that they are virtually identical or closely resemble each other. However, their fatty acid content was significantly different and heterogeneity was clearly demonstrated in the patterns of isoelectric focusing. In this communication, a single isoform (~15.0) from peak B FABP was further purified by successive DEAE-cellulose column chromatography and used as the final preparation. When the final FABP was partly freed of fatty acids by a mild delipidation technique using Lipidex 1,000, the pI shifted upward from 5.0 to 7.0. However, the pI of the delipidated FABP returned to its original pI of 5.0 after recombining fatty acids. These in vitro manipulations of bound fatty acid content made clear its possible cause of the microheterogeneity of FABP. o 1~92 Academic
Press,
Inc.
Fatty acid binding protein (FABP)3 (also called “sterol carrier protein” (SCP) (1) and “Z protein” (2-4)) is an abundant, low-molecular-weight (around 14,000) cytosolic protein and contains long-chain fatty acids with Kd values of l-4 PM (5). FABP is thought to function in the intracellular transport and metabolism of long-chain fatty acids in liver and intestine, as well as in modulation of the r This work was supported by a Grant-in-Aid from the Ministry of Education, Science and Culture of Japan, and a grant from Eisai Co. ’ To whom correspondence should be addressed. 3 Abbreviation used: FABP, fatty acid binding protein.
rate of fatty acid uptake or utilization in response to nutritional and hormonal circumstances (5-8). In addition, FABP may protect other proteins and membranes from deleterious effects of higher concentrations of fatty acids (9). Although the primary structures of FABPs deduced from their cloned cDNA sequences (10, 11) and the crystal structure (12) are now known, the precise functional role of FABP remains an open question. FABP in amounts up to 5 to 8% of cytosolic proteins was found in rat liver (1, 13), yet an astonishing array of conflicting data with regard to isoelectric points, molecular masses, stability, and binding specificities of these proteins has been reported. For example, isoelectric focusing of binding proteins yielded several subfractions still capable of complexing fatty acids (4, 13, 14), a phenomenon explained by heterogeneity, or multiplicity, or concomitant occurrence of liganded and free forms of the binding proteins. As a result of these uncertainties, the true cellular function of FABP has yet to be determined. In the course of our studies on fatty acid-protein interactions (15, 16), we observed the occurrence of many discrete bands upon gel electrofocusing of FABP in rat liver cytosol, partly in accordance with other findings (4, 13,17-19). Then, we aimed to define the exact nature of the isoform occurrence in view of the possibility that bound fatty acids might make an important contribution to the observed isoform. The present study was undertaken in the hope that a highly sensitive thin-layer gel electrofocusing method might shed some light on these questions, because we lacked even an approximate estimation of the number of species involved. To this end, the relation of the number of fatty acids associated with FABP molecules to the heterogeneity was the major concern of this study and comparison of the population distribution of delipidated and relipidated FABPs was made with this focus in mind.
254 All
0003.9S61/92 $5.ocl Copyright 0 1992 by Academic Press, Inc. rights of reproduction in any form reserved.
INTERCONVERSION
EXPERIMENTAL
OF RAT
LIVER
FATTY
ACID
BINDING
255
PROTEIN
PROCEDURES
Materials. The following materials were all obtained from commercial sources: Sephadex G-100 from Pharmacia; DEAE-cellulose (DE52) from Whatman; sodium oleate and Lipidex 1,000 (hydroxyalcohol-dextran) from Sigma; thin-layer plate (silica gel 60) from Merck. All other chemicals were of analytical grade. Livers from Wistar Purification of rut liver fatty acid binding protein. rats, weighing about 250 g, were perfused with ice-cold 0.25 M sucrose, excised, and homogenized with a Potter-Elvehjem homogenizer in 2.5 vol of 0.25 M sucrose. The homogenate was centrifuged at 13,000g for 10 min and the supernatant was centrifuged at 105,OOOgfor 1 h. The resultant supernatant was the source of FABP. All the purification steps were performed at 4°C. The cytosol fraction (50 ml) was applied on a column of Sephadex G-100 (4 X 100 cm) equilibrated with 30 mM TrisHCl buffer (pH 9.0) (Fig. 1). Fractions possessing both special ultraviolet absorption spectra of FABP (1) and detected fatty acids were pooled. The pooled fractions were applied to a DEAE-cellulose column (2 X 14 cm) equilibrated with the same buffer (Fig. 2). The column was washed with the equilibration buffer until the unbound protein was nearly zero and then developed with a linear concentration gradient of NaCl from 0 to 0.2 M in the same buffer (150 ml each) to elute FABP. Two major peaks (A and B) of FABP were obtained. To further isolate the specified isoform of FABP, peak B was pooled and dialyzed against 30 mM Tris-HCI buffer (pH 9.0) and applied again to a DEAE-cellulose column (2 X 14 cm) equilibrated with the same buffer (Fig. 4). The procedures for developing this column were the same as with the first DEAE-cellulose column chromatography, except for a linear concentration gradient of NaCl from 0 to 0.1 M (300 ml each). The largest FABP fractions were pooled, dialyzed against 30 mM Tris-HCl buffer (pH 9.0), and used as the final preparation. Fatty acid content during column chromatography was determined calorimetrically by employing the NEFA C-test kit (Wako) (20). Protein was determined by the method of Lowry et al. (21), using bovine serum albumin as the standard. Lipids were extracted from the purified FABP Fatty acid analysis. by the method of Folch et al. (22) and separated by thin-layer chromatography on a silica gel 60 plate in petroleum ether/diethyl ether/ acetic acid (80/20/l, v/v). Appropriate zones were detected after spraying with 6% phosphomolybdate in ethanol and identified by means of a standard. Free fatty acids were extracted three times with hexane. The extracted samples were methylated with boron-trifluoride methanol reagent (23). The resultant fatty acid methyl esters were subjected to gas-
Fraction
Number
FIG. 2. The first DEAE-cellulose column of 1 ml were collected at a flow rate of 30 were used for the determination of the fatty peaks A and B indicates the fractions that the characterization of FABP.
chromatography. Fractions ml/h and aliquots of 0.1 ml acid content. The bar above were combined and used for
liquid chromatography and mass spectrometry using a JOEL TMCD300 equipped with a 3% SE-30 column under the following conditions: 160-270°C at a rate of 4’C/min; ion source, 180°C; ionizing voltage, 70 eV. Delipidution and relipidution of fatty acids. Both delipidation and assay of the fatty acid recombining activities of FABP were carried out as described previously using Lipidex 1,000 (16), which removes unbound as well as protein-bound fatty acid from aqueous solutions in a temperature-dependent manner according to the protein-lipid kinetics (24, 25). Electrophoretic and thin-layer gel electrofocusing procedures. Sodium dodecyl sulfate-polyacrylamide gel (12.5%) electrophoresis was performed according to the method of Weber and Osborn (26). Analytical isoelectric focusing gel electrophoresis on commercial PhastGel (IEF 39) was performed as described in the instruction manual using a Pharmacia PhastSystem. All steps were carried out at 15°C. Analysis of amino acid composition. The purified FABP was hydrolyzed with 6 N HCl at 1lO’C for 24 h in an evacuated sealed tube and analyzed with a Hitachi 835 amino acid analyzer.
RESULTS
FIG. 1. Sephadex G-100 column chromatography of rat liver cytosol. Fractions of 5 ml were collected at a flow rate of 20 ml/h and aliquots of 0.1 ml were used for the determination of the fatty acid content. The bar indicates fractions that were combined and used for the first DEAEcellulose column chromatography.
Fatty acid content of the two homogenous FABPs. When the FABP fraction obtained from the Sephadex G-100 column (Fig. 1) was subjected to DEAEcellulose column chromatography, it was further separated into two major peaks, A and B (Fig. 2). Peaks A and B were homogenous upon polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate with a molecular weight of 14,000 (Fig. 3, left) and had essentially identical amino acid compositions (Table I). The amino acid analysis results were comparable with those reported for band C-protein (14), sterol carrier protein (l), Z-protein (27), and FABP (13). It is noteworthy that no tryptophan was detected. Their endogenous fatty acid contents, however, were strikingly different, 0.6 mol fatty acid/m01 protein for A and 2.3 mol fatty acid/m01 protein for B (Table II). These results indicate that these two proteins of A and B FABP were essentially the same protein but had different fatty acid contents.
256
LI AND
ISHIBASHI TABLE
-8
II
Fatty Acid Content and Isoelectric Point of FABP Purified on the First DEAE-Cellulose Column Chromatography Fatty acids Peaks
-8
16:0
l&O
PH
18:l (mol/mol
A B
0.22
0.92
0.25 0.40
0.04 0.64
20:o
22:o
Total
0.07 0.18
0.58 2.30
pl
FABP) N.D.” 0.16
6.1-7.0 5.0-5.6
-5 ’ Not detected. -4
FIG. 3. (Left) Electrophoresis of FABP peaks A and B upon polyacrylamide disc gel (12.5%) in the presence of 0.1% sodium dodecyl sulfate. The FABP fractions of peaks A (15 pg) and B (10 pg) obtained by the first DEAE-cellulose column chromatography were subjected to electrophoresis. (Right) Thin-layer gel electrofocusing (pH 3-9) of peaks A (0.5 pg) and B (0.5 pg).
Microheterogeneity of FABP by fatty acid content. On the other hand, the thin-layer gel electrofocusing spectra of the two homogenous A and B FABPs exhibited multiple bands in both preparations (Fig. 3, right). Peak A was focused ranging from pl 6.1 to 7.0 with a main band of 7.0, while peak B was from ~15.0 to 5.6 with a main band
TABLE
1
Amino Acid Composition of FABP Purified on the First DEAE-Cellulose Column Chromatography Peaks Amino acid
A
B (mol/mol
protein)
Met Asx Thr Ser Glx GUY Ala Val Ile Leu Tyr Phe LYS His Pro Trp Arg
0.9 6.3 12.1 10.5 3.5 17.5 12.5 3.4 11.5 8.5 7.4 1.3 5.8 16.5 2.1 2.7 0 2.4
0.9 6.4 12.0 10.6 3.8 17.5 12.9 3.1 12.1 9.1 7.4
Total
124.9
127.6
CYS
of ~15.0. As shown in Table II, the peak A FABP had much less fatty acid than the peak B FABP. That is, as fatty acid content decreased the pI increased. Consequently, electrofocusing of FABP was shown to be influenced dramatically by the fatty acid content. Purification of a specified isoform of FABPs. The peak B fraction, which was the major peak in the first DEAEcellulose column chromatography (Fig. 2), was applied on a second DEAE-cellulose column to resolve many isoforms as shown in Fig. 4. The biggest peak was combined and used as the final FABP preparation. The final preparation was homogenous even on electrofocusing, and its plwas calculated to be 5.0 (Fig. 5). The final single isoform (~15.0) of FABP was purified 17.7-fold with a 2.7% yield (Table III). In vitro confirmation of heterogenous FABP occurrence by removing and recombining fatty acids. Delipidation and relipidation experiments were carried out in an attempt to elucidate the influence of fatty acid content on the electrofocusing pattern of FABP. When the final FABP preparation (~15.0) was partially delipidated using
1.1 6.3 17.1 2.2 2.7 0 2.4
Fraction
Number
FIG. 4. The second DEAE-cellulose column chromatography. Fractions of 1 ml were collected at a flow rate of 30 ml/h and aliquots of 0.1 ml were used for the determination of the fatty acid content. The bar indicates the fractions which were combined and used as the final FABP preparation.
INTERCONVERSION
Purification
OF RAT
LIVER
FATTY
TABLE
III
of a Single
Isoform
ACID
(~15.0)
BINDING
257
PROTEIN
of FABP
Fractions
Vol (ml)
Protein (ms)
Fatty acids (wzlmg protein)
Yield (%I
105,OOOgSupernatant Sephadex G-100 First DEAE-cellulose Second DEAE-cellulose
120 270 75 36
2760 61.3 10.5 4.2
2.2 23.8 37.4 38.9
100 39.3 6.7 2.7
Lipidex 1,000, its pl shifted from 5.0 to 6.6 and 7.1 (Fig. 5). Furthermore, the pl of the delipidated FABP returned to its original pI of 5.0 from the higher pl by recombining fatty acid. These relationships between the fatty acid content of FABP and its pl are shown in Table IV. Thus, the isoforms of FABP, differing in pl, were at least partly interconvertible and possibly depended on the amount of bound endogenous ligand. Furthermore, these manipulations of the fatty acid content induced significant effects on the secondary structure of FABP (details not shown). DISCUSSION
Rat liver cytosol contains one or more fractions capable of binding fatty acid (Fig. 1) and the presently investigated FABP is a member of the family. There are several reasons why FABP in general represents an attractive model system for examining long-chain fatty acid-protein interactions. FABP is made up of small proteins composed of 127-132 amino acids and contains long-chain fatty acids (5). The heterogeneity of FABP can now be explained by the fact that many isoforms of M, 14,000 occur in the cytosol of rat liver; they differ in pl (Fig. 3), but resemble each other in their amino acid compositions (Table I). In this communication, to gain insights about the heterogenous behavior of FABP, isoelectric focusing was chosen since molecules could be separated according to their ef-
FIG. 5. Thin-layer gel electrofocusing (pH 3-9) of the final FABP. The electrophoretic conditions were the same as those used in Fig. 4. (a) Native (0.1 pg), (b) delipidated (0.2 pg), (c) relipidated (0.1 pg) FABP.
Purification (-fold) 1 10.8 17.0 17.7
fective net charges which ought to be determined by the chemical and conformational integrity of the protein molecules and the number and type of noncovalently bound ligands. Now it has been demonstrated that the fatty acid components of FABP clearly made a major contribution to the heterogeneity, although unique species with 1, 2, or 3 fatty acids bound cannot be separated. Because the pK, of fatty acids is about 4.8 (17), their carboxyl groups will charge negatively under physiological conditions, which creates bonds to positively charged amino groups of the protein by ionic binding. With delipidation the positive charges on the protein surface, where negatively charged fatty acids were bound before, would be exposed. The positive charges on the protein surface change markedly depending on the amount of the bound fatty acids, which is thought to be the principal cause of the microheterogeneity of FABP. Fluctuations in free fatty acid levels within the physiological range seem to have a major modulatory effect on the electrofocusing pattern. In the present study, we have developed a procedure for the isolation of a single isoform (~15.0) of FABP involving successive DEAE-cellulose column chromatographies to get further information on ligand-protein interaction. As a result, it has been found that the pl of the delipidated FABP shifted to become more basic than the native one (Fig. 5). This was proved by experiments recombining fatty acids to the delipidated FABP, which brought about a return to the original pl of 5.0. Such a pI shift upon binding fatty acid was observed earlier for serum albumin, the major extracellular longchain fatty acid binding protein (28). When two or more fatty acids are bound to human serum albumin the pl decreases by approximately 0.75 pH unit. On the other hand, Spencer and King demonstrated that neutralizing one charge only caused a 0.07 shift in the pl value for bovine albumin (29). Therefore, the change in the pl unit upon the binding of fatty acid molecules cannot be attributed entirely to their negative charges. The present results do not strictly exclude the possibility that the change in pl of FABP complexed with fatty acid was caused by conformational changes altering the pK, of several ionizable groups on the protein.
258
LI AND
ISHIBASHI
TABLEIV Effect of Delipidation
and Relipidation
on the Electrofocusing
Spectrum of a Single Isoform of FABP
Fatty acids Treatments
16:0
16:l
l&O
l&l (mol/mol
None Delipidation Relipidation”
0.98 0.41 0.41
a Oleic acid was reconstituted * Not detected.
0.08 0.05 0.05 to the delipidated
0.43 0.11 0.11
form as described under Experimental
ACKNOWLEDGMENT We are grateful to M. Kim Barrymore the manuscript.
for his linguistic
correction
of
REFERENCES 1. Dempsey, M. E., McCoy, K. E., Baker, H. N., Dimitriadou-Vafiadou, A., Lorsbach, T., and Howard, J. B. (1981) J. Biol. Chem. 256, 1867-1873. 2. Levi, A. J., Gatmaitan, Z., and Arias, I. M. (1969) J. Clin. Inuest. 48,2156-2167. 3. Mishkin, S., Stein, L., Gatmaitan, Z., and Arias, I. M. (1972) Biochem. Biophys. Res. Commun. 47,997-1003. 4. Trulzsch, D., and Arias, I. M. (1981) Arch. Biochem. Biophys. 209, 433-440. 5. Fleischner, G., Meijer, D. K. F., Levine, W. G., Gatmaitan, Z., Gluck, R., and Arias, I. M. (1975) Biochem. Biophys. Res. Commun. 67, 1401-1407. 6. Burnett, D. A., Lysenko, N., Manning, J. A., and Ockner, R. K. (1979) Gastroenterology 77, 241-249. 7. Ockner, R. K., Manning, J. A., Poppenhausen, R. B., and Ho, W. K. L. (1972) Science 177, 56-58. 8. Ockner, R. K., Lysenko, N., Manning, J. A., Monroe, S. E., and Burnett, D. A. (1980) J. Clin. Inuest. 65, 1013-1023. 9. Lunzer, M. A., Manning, J. A., and Ockner, R. K. (1977) J. Biol. Chem. 252.5483-5487.
Total
0.31 N.D.* N.D.
2.66 0.78 1.77
FABP) 0.86 0.21 1.20
A question of general interest is, of course, whether the mechanism at work during isoelectric focusing of FABP is also valid in other proteins. Even weak binding to a protein might change the final electrofocusing pattern and lead to false indications of heterogeneity or of equilibria among different conformations. No heterogeneity of proteins should therefore be accepted before accurate analysis of the protein itself. In order to understand the structural basis for multiplicity and functional properties of FABP, we need further studies of molecular bases.
20:o
5.0 6.6, 7.1 5.0
Procedures.
10. Gordon, J. I., Alpers, D. H., Ockner, (1983) J. Biol. C&m. 268,3356-3363.
R. K., and Strauss, A. W.
11. Schulenberg-Schell, H., Schafer, P., Keuper, H. J. K., Stanislawski, B., Hoffmann, E., Ruterjans, H., and Spener, F. (1988) Eur. J. Biochem. 170,565-574. 12. Sacchettini, J. C., Gordon, J. I., and Banaszak, L. J. (1989) J. Mol. Biol. 208,327-339. 13. Ockner, R. K., Manning, J. A., and Kane, J. P. (1982) J. Biol. Chem. 257, 7872-7878. 14. Billheimer, J. T., and Gaylor, J. L. (1980) J. Biol. Chem. 255,81288135. 15. Senjo, M., Ishibashi, T., Imai, Y., Takahashi, K., andOno, T. (1985) Arch. B&hem. Biophys. 236,662-668. 16. Ming, L., and Ishibashi, T. (1990) J. Biochem. 108, 462-465. 17. Haunerland, N., Jagschies, G., Schulenberg, H., and Spener, F. (1984) Hoppe-Seyler’s 2. Physiol. Chem. 365, 365-376. 18. Takahashi, K., Odani, S., and Ono, T. (1983) Eur. J. Biochem. 136, 589-601. 19. Matarese, V., Stone, R. L., Waggoner, D. W., and Bernlohr, D. A. (1989) Prog. Lipid Res. 28, 245-272. 20. Shimizu, S., Yasui, K., Tani, Y., and Yamada, H. (1979) Biochem. Biophys. Res. Commun. 91, 108-113. 21. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193,265-275. 22. Folch, J., Lees, M., and Sloane-Stanley, G. H. (1957) J. Biol. Chem. 226, 497-509. 23. Glass, R. L. (1971) Lipids 6, 919-925. 24. Glatz, J. F. C., and Veerkamp, J. H. (1983) Anal. Biochem. 132,
89-95. 25. Glatz, J. F. C., Baerwaldt, C. C. F., Veerkamp, J. H., and Kempen, H. J. M. (1984) J. Biol. Chem. 259,4295-4300. 26. Weber, K., and Osborn, M. (1969) J. Biol. Chem. 244,4406-4412. 27. Ketterer, B., Tipping, E., Hackney, J. F., and Beale, D. (1976) Bi0chem.J. 155,511-521. 28. Basu, S. P., Rao, S. N., and Hartsuck, J. A. (1978) Biochim. Biophys. Acta 533, 66-73. 29. Spencer, E. M., and King, T. P. (1971) J. Biol. Chem. 246, 201208.
OF BIOCHEMISTRY
AND
BIOPHYSICS
Vol. 298, No. 1, October, pp. 254-258, 1992
An in vitro Reversible Interconversion of Rat Liver Fatty Acid Binding Protein Having Different Isoelectric Points by Virtue of the Fatty Acid Content’ Ming Li and Teruo Ishibashi2 Department
of Biochemistry,
Hokkaido
University
School of Medicine,
Sapporo 060, Japan
Received March 2, 1992, and in revised form June 11, 1992
Rat liver fatty acid binding protein (FABP) was purified to homogeneity by procedures including Sephadex G- 100 and DEAE-cellulose column chromatographies. FABP was resolved into two major peaks, A and B, by the first DEAE-cellulose column chromatography. Each of these two fractions exhibited apparent homogeneity upon polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate with a molecular weight of 14,000 Da and amino acid analysis of these fractions has revealed that they are virtually identical or closely resemble each other. However, their fatty acid content was significantly different and heterogeneity was clearly demonstrated in the patterns of isoelectric focusing. In this communication, a single isoform (~15.0) from peak B FABP was further purified by successive DEAE-cellulose column chromatography and used as the final preparation. When the final FABP was partly freed of fatty acids by a mild delipidation technique using Lipidex 1,000, the pI shifted upward from 5.0 to 7.0. However, the pI of the delipidated FABP returned to its original pI of 5.0 after recombining fatty acids. These in vitro manipulations of bound fatty acid content made clear its possible cause of the microheterogeneity of FABP. o 1~92 Academic
Press,
Inc.
Fatty acid binding protein (FABP)3 (also called “sterol carrier protein” (SCP) (1) and “Z protein” (2-4)) is an abundant, low-molecular-weight (around 14,000) cytosolic protein and contains long-chain fatty acids with Kd values of l-4 PM (5). FABP is thought to function in the intracellular transport and metabolism of long-chain fatty acids in liver and intestine, as well as in modulation of the r This work was supported by a Grant-in-Aid from the Ministry of Education, Science and Culture of Japan, and a grant from Eisai Co. ’ To whom correspondence should be addressed. 3 Abbreviation used: FABP, fatty acid binding protein.
rate of fatty acid uptake or utilization in response to nutritional and hormonal circumstances (5-8). In addition, FABP may protect other proteins and membranes from deleterious effects of higher concentrations of fatty acids (9). Although the primary structures of FABPs deduced from their cloned cDNA sequences (10, 11) and the crystal structure (12) are now known, the precise functional role of FABP remains an open question. FABP in amounts up to 5 to 8% of cytosolic proteins was found in rat liver (1, 13), yet an astonishing array of conflicting data with regard to isoelectric points, molecular masses, stability, and binding specificities of these proteins has been reported. For example, isoelectric focusing of binding proteins yielded several subfractions still capable of complexing fatty acids (4, 13, 14), a phenomenon explained by heterogeneity, or multiplicity, or concomitant occurrence of liganded and free forms of the binding proteins. As a result of these uncertainties, the true cellular function of FABP has yet to be determined. In the course of our studies on fatty acid-protein interactions (15, 16), we observed the occurrence of many discrete bands upon gel electrofocusing of FABP in rat liver cytosol, partly in accordance with other findings (4, 13,17-19). Then, we aimed to define the exact nature of the isoform occurrence in view of the possibility that bound fatty acids might make an important contribution to the observed isoform. The present study was undertaken in the hope that a highly sensitive thin-layer gel electrofocusing method might shed some light on these questions, because we lacked even an approximate estimation of the number of species involved. To this end, the relation of the number of fatty acids associated with FABP molecules to the heterogeneity was the major concern of this study and comparison of the population distribution of delipidated and relipidated FABPs was made with this focus in mind.
254 All
0003.9S61/92 $5.ocl Copyright 0 1992 by Academic Press, Inc. rights of reproduction in any form reserved.
INTERCONVERSION
EXPERIMENTAL
OF RAT
LIVER
FATTY
ACID
BINDING
255
PROTEIN
PROCEDURES
Materials. The following materials were all obtained from commercial sources: Sephadex G-100 from Pharmacia; DEAE-cellulose (DE52) from Whatman; sodium oleate and Lipidex 1,000 (hydroxyalcohol-dextran) from Sigma; thin-layer plate (silica gel 60) from Merck. All other chemicals were of analytical grade. Livers from Wistar Purification of rut liver fatty acid binding protein. rats, weighing about 250 g, were perfused with ice-cold 0.25 M sucrose, excised, and homogenized with a Potter-Elvehjem homogenizer in 2.5 vol of 0.25 M sucrose. The homogenate was centrifuged at 13,000g for 10 min and the supernatant was centrifuged at 105,OOOgfor 1 h. The resultant supernatant was the source of FABP. All the purification steps were performed at 4°C. The cytosol fraction (50 ml) was applied on a column of Sephadex G-100 (4 X 100 cm) equilibrated with 30 mM TrisHCl buffer (pH 9.0) (Fig. 1). Fractions possessing both special ultraviolet absorption spectra of FABP (1) and detected fatty acids were pooled. The pooled fractions were applied to a DEAE-cellulose column (2 X 14 cm) equilibrated with the same buffer (Fig. 2). The column was washed with the equilibration buffer until the unbound protein was nearly zero and then developed with a linear concentration gradient of NaCl from 0 to 0.2 M in the same buffer (150 ml each) to elute FABP. Two major peaks (A and B) of FABP were obtained. To further isolate the specified isoform of FABP, peak B was pooled and dialyzed against 30 mM Tris-HCI buffer (pH 9.0) and applied again to a DEAE-cellulose column (2 X 14 cm) equilibrated with the same buffer (Fig. 4). The procedures for developing this column were the same as with the first DEAE-cellulose column chromatography, except for a linear concentration gradient of NaCl from 0 to 0.1 M (300 ml each). The largest FABP fractions were pooled, dialyzed against 30 mM Tris-HCl buffer (pH 9.0), and used as the final preparation. Fatty acid content during column chromatography was determined calorimetrically by employing the NEFA C-test kit (Wako) (20). Protein was determined by the method of Lowry et al. (21), using bovine serum albumin as the standard. Lipids were extracted from the purified FABP Fatty acid analysis. by the method of Folch et al. (22) and separated by thin-layer chromatography on a silica gel 60 plate in petroleum ether/diethyl ether/ acetic acid (80/20/l, v/v). Appropriate zones were detected after spraying with 6% phosphomolybdate in ethanol and identified by means of a standard. Free fatty acids were extracted three times with hexane. The extracted samples were methylated with boron-trifluoride methanol reagent (23). The resultant fatty acid methyl esters were subjected to gas-
Fraction
Number
FIG. 2. The first DEAE-cellulose column of 1 ml were collected at a flow rate of 30 were used for the determination of the fatty peaks A and B indicates the fractions that the characterization of FABP.
chromatography. Fractions ml/h and aliquots of 0.1 ml acid content. The bar above were combined and used for
liquid chromatography and mass spectrometry using a JOEL TMCD300 equipped with a 3% SE-30 column under the following conditions: 160-270°C at a rate of 4’C/min; ion source, 180°C; ionizing voltage, 70 eV. Delipidution and relipidution of fatty acids. Both delipidation and assay of the fatty acid recombining activities of FABP were carried out as described previously using Lipidex 1,000 (16), which removes unbound as well as protein-bound fatty acid from aqueous solutions in a temperature-dependent manner according to the protein-lipid kinetics (24, 25). Electrophoretic and thin-layer gel electrofocusing procedures. Sodium dodecyl sulfate-polyacrylamide gel (12.5%) electrophoresis was performed according to the method of Weber and Osborn (26). Analytical isoelectric focusing gel electrophoresis on commercial PhastGel (IEF 39) was performed as described in the instruction manual using a Pharmacia PhastSystem. All steps were carried out at 15°C. Analysis of amino acid composition. The purified FABP was hydrolyzed with 6 N HCl at 1lO’C for 24 h in an evacuated sealed tube and analyzed with a Hitachi 835 amino acid analyzer.
RESULTS
FIG. 1. Sephadex G-100 column chromatography of rat liver cytosol. Fractions of 5 ml were collected at a flow rate of 20 ml/h and aliquots of 0.1 ml were used for the determination of the fatty acid content. The bar indicates fractions that were combined and used for the first DEAEcellulose column chromatography.
Fatty acid content of the two homogenous FABPs. When the FABP fraction obtained from the Sephadex G-100 column (Fig. 1) was subjected to DEAEcellulose column chromatography, it was further separated into two major peaks, A and B (Fig. 2). Peaks A and B were homogenous upon polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate with a molecular weight of 14,000 (Fig. 3, left) and had essentially identical amino acid compositions (Table I). The amino acid analysis results were comparable with those reported for band C-protein (14), sterol carrier protein (l), Z-protein (27), and FABP (13). It is noteworthy that no tryptophan was detected. Their endogenous fatty acid contents, however, were strikingly different, 0.6 mol fatty acid/m01 protein for A and 2.3 mol fatty acid/m01 protein for B (Table II). These results indicate that these two proteins of A and B FABP were essentially the same protein but had different fatty acid contents.
256
LI AND
ISHIBASHI TABLE
-8
II
Fatty Acid Content and Isoelectric Point of FABP Purified on the First DEAE-Cellulose Column Chromatography Fatty acids Peaks
-8
16:0
l&O
PH
18:l (mol/mol
A B
0.22
0.92
0.25 0.40
0.04 0.64
20:o
22:o
Total
0.07 0.18
0.58 2.30
pl
FABP) N.D.” 0.16
6.1-7.0 5.0-5.6
-5 ’ Not detected. -4
FIG. 3. (Left) Electrophoresis of FABP peaks A and B upon polyacrylamide disc gel (12.5%) in the presence of 0.1% sodium dodecyl sulfate. The FABP fractions of peaks A (15 pg) and B (10 pg) obtained by the first DEAE-cellulose column chromatography were subjected to electrophoresis. (Right) Thin-layer gel electrofocusing (pH 3-9) of peaks A (0.5 pg) and B (0.5 pg).
Microheterogeneity of FABP by fatty acid content. On the other hand, the thin-layer gel electrofocusing spectra of the two homogenous A and B FABPs exhibited multiple bands in both preparations (Fig. 3, right). Peak A was focused ranging from pl 6.1 to 7.0 with a main band of 7.0, while peak B was from ~15.0 to 5.6 with a main band
TABLE
1
Amino Acid Composition of FABP Purified on the First DEAE-Cellulose Column Chromatography Peaks Amino acid
A
B (mol/mol
protein)
Met Asx Thr Ser Glx GUY Ala Val Ile Leu Tyr Phe LYS His Pro Trp Arg
0.9 6.3 12.1 10.5 3.5 17.5 12.5 3.4 11.5 8.5 7.4 1.3 5.8 16.5 2.1 2.7 0 2.4
0.9 6.4 12.0 10.6 3.8 17.5 12.9 3.1 12.1 9.1 7.4
Total
124.9
127.6
CYS
of ~15.0. As shown in Table II, the peak A FABP had much less fatty acid than the peak B FABP. That is, as fatty acid content decreased the pI increased. Consequently, electrofocusing of FABP was shown to be influenced dramatically by the fatty acid content. Purification of a specified isoform of FABPs. The peak B fraction, which was the major peak in the first DEAEcellulose column chromatography (Fig. 2), was applied on a second DEAE-cellulose column to resolve many isoforms as shown in Fig. 4. The biggest peak was combined and used as the final FABP preparation. The final preparation was homogenous even on electrofocusing, and its plwas calculated to be 5.0 (Fig. 5). The final single isoform (~15.0) of FABP was purified 17.7-fold with a 2.7% yield (Table III). In vitro confirmation of heterogenous FABP occurrence by removing and recombining fatty acids. Delipidation and relipidation experiments were carried out in an attempt to elucidate the influence of fatty acid content on the electrofocusing pattern of FABP. When the final FABP preparation (~15.0) was partially delipidated using
1.1 6.3 17.1 2.2 2.7 0 2.4
Fraction
Number
FIG. 4. The second DEAE-cellulose column chromatography. Fractions of 1 ml were collected at a flow rate of 30 ml/h and aliquots of 0.1 ml were used for the determination of the fatty acid content. The bar indicates the fractions which were combined and used as the final FABP preparation.
INTERCONVERSION
Purification
OF RAT
LIVER
FATTY
TABLE
III
of a Single
Isoform
ACID
(~15.0)
BINDING
257
PROTEIN
of FABP
Fractions
Vol (ml)
Protein (ms)
Fatty acids (wzlmg protein)
Yield (%I
105,OOOgSupernatant Sephadex G-100 First DEAE-cellulose Second DEAE-cellulose
120 270 75 36
2760 61.3 10.5 4.2
2.2 23.8 37.4 38.9
100 39.3 6.7 2.7
Lipidex 1,000, its pl shifted from 5.0 to 6.6 and 7.1 (Fig. 5). Furthermore, the pl of the delipidated FABP returned to its original pI of 5.0 from the higher pl by recombining fatty acid. These relationships between the fatty acid content of FABP and its pl are shown in Table IV. Thus, the isoforms of FABP, differing in pl, were at least partly interconvertible and possibly depended on the amount of bound endogenous ligand. Furthermore, these manipulations of the fatty acid content induced significant effects on the secondary structure of FABP (details not shown). DISCUSSION
Rat liver cytosol contains one or more fractions capable of binding fatty acid (Fig. 1) and the presently investigated FABP is a member of the family. There are several reasons why FABP in general represents an attractive model system for examining long-chain fatty acid-protein interactions. FABP is made up of small proteins composed of 127-132 amino acids and contains long-chain fatty acids (5). The heterogeneity of FABP can now be explained by the fact that many isoforms of M, 14,000 occur in the cytosol of rat liver; they differ in pl (Fig. 3), but resemble each other in their amino acid compositions (Table I). In this communication, to gain insights about the heterogenous behavior of FABP, isoelectric focusing was chosen since molecules could be separated according to their ef-
FIG. 5. Thin-layer gel electrofocusing (pH 3-9) of the final FABP. The electrophoretic conditions were the same as those used in Fig. 4. (a) Native (0.1 pg), (b) delipidated (0.2 pg), (c) relipidated (0.1 pg) FABP.
Purification (-fold) 1 10.8 17.0 17.7
fective net charges which ought to be determined by the chemical and conformational integrity of the protein molecules and the number and type of noncovalently bound ligands. Now it has been demonstrated that the fatty acid components of FABP clearly made a major contribution to the heterogeneity, although unique species with 1, 2, or 3 fatty acids bound cannot be separated. Because the pK, of fatty acids is about 4.8 (17), their carboxyl groups will charge negatively under physiological conditions, which creates bonds to positively charged amino groups of the protein by ionic binding. With delipidation the positive charges on the protein surface, where negatively charged fatty acids were bound before, would be exposed. The positive charges on the protein surface change markedly depending on the amount of the bound fatty acids, which is thought to be the principal cause of the microheterogeneity of FABP. Fluctuations in free fatty acid levels within the physiological range seem to have a major modulatory effect on the electrofocusing pattern. In the present study, we have developed a procedure for the isolation of a single isoform (~15.0) of FABP involving successive DEAE-cellulose column chromatographies to get further information on ligand-protein interaction. As a result, it has been found that the pl of the delipidated FABP shifted to become more basic than the native one (Fig. 5). This was proved by experiments recombining fatty acids to the delipidated FABP, which brought about a return to the original pl of 5.0. Such a pI shift upon binding fatty acid was observed earlier for serum albumin, the major extracellular longchain fatty acid binding protein (28). When two or more fatty acids are bound to human serum albumin the pl decreases by approximately 0.75 pH unit. On the other hand, Spencer and King demonstrated that neutralizing one charge only caused a 0.07 shift in the pl value for bovine albumin (29). Therefore, the change in the pl unit upon the binding of fatty acid molecules cannot be attributed entirely to their negative charges. The present results do not strictly exclude the possibility that the change in pl of FABP complexed with fatty acid was caused by conformational changes altering the pK, of several ionizable groups on the protein.
258
LI AND
ISHIBASHI
TABLEIV Effect of Delipidation
and Relipidation
on the Electrofocusing
Spectrum of a Single Isoform of FABP
Fatty acids Treatments
16:0
16:l
l&O
l&l (mol/mol
None Delipidation Relipidation”
0.98 0.41 0.41
a Oleic acid was reconstituted * Not detected.
0.08 0.05 0.05 to the delipidated
0.43 0.11 0.11
form as described under Experimental
ACKNOWLEDGMENT We are grateful to M. Kim Barrymore the manuscript.
for his linguistic
correction
of
REFERENCES 1. Dempsey, M. E., McCoy, K. E., Baker, H. N., Dimitriadou-Vafiadou, A., Lorsbach, T., and Howard, J. B. (1981) J. Biol. Chem. 256, 1867-1873. 2. Levi, A. J., Gatmaitan, Z., and Arias, I. M. (1969) J. Clin. Inuest. 48,2156-2167. 3. Mishkin, S., Stein, L., Gatmaitan, Z., and Arias, I. M. (1972) Biochem. Biophys. Res. Commun. 47,997-1003. 4. Trulzsch, D., and Arias, I. M. (1981) Arch. Biochem. Biophys. 209, 433-440. 5. Fleischner, G., Meijer, D. K. F., Levine, W. G., Gatmaitan, Z., Gluck, R., and Arias, I. M. (1975) Biochem. Biophys. Res. Commun. 67, 1401-1407. 6. Burnett, D. A., Lysenko, N., Manning, J. A., and Ockner, R. K. (1979) Gastroenterology 77, 241-249. 7. Ockner, R. K., Manning, J. A., Poppenhausen, R. B., and Ho, W. K. L. (1972) Science 177, 56-58. 8. Ockner, R. K., Lysenko, N., Manning, J. A., Monroe, S. E., and Burnett, D. A. (1980) J. Clin. Inuest. 65, 1013-1023. 9. Lunzer, M. A., Manning, J. A., and Ockner, R. K. (1977) J. Biol. Chem. 252.5483-5487.
Total
0.31 N.D.* N.D.
2.66 0.78 1.77
FABP) 0.86 0.21 1.20
A question of general interest is, of course, whether the mechanism at work during isoelectric focusing of FABP is also valid in other proteins. Even weak binding to a protein might change the final electrofocusing pattern and lead to false indications of heterogeneity or of equilibria among different conformations. No heterogeneity of proteins should therefore be accepted before accurate analysis of the protein itself. In order to understand the structural basis for multiplicity and functional properties of FABP, we need further studies of molecular bases.
20:o
5.0 6.6, 7.1 5.0
Procedures.
10. Gordon, J. I., Alpers, D. H., Ockner, (1983) J. Biol. C&m. 268,3356-3363.
R. K., and Strauss, A. W.
11. Schulenberg-Schell, H., Schafer, P., Keuper, H. J. K., Stanislawski, B., Hoffmann, E., Ruterjans, H., and Spener, F. (1988) Eur. J. Biochem. 170,565-574. 12. Sacchettini, J. C., Gordon, J. I., and Banaszak, L. J. (1989) J. Mol. Biol. 208,327-339. 13. Ockner, R. K., Manning, J. A., and Kane, J. P. (1982) J. Biol. Chem. 257, 7872-7878. 14. Billheimer, J. T., and Gaylor, J. L. (1980) J. Biol. Chem. 255,81288135. 15. Senjo, M., Ishibashi, T., Imai, Y., Takahashi, K., andOno, T. (1985) Arch. B&hem. Biophys. 236,662-668. 16. Ming, L., and Ishibashi, T. (1990) J. Biochem. 108, 462-465. 17. Haunerland, N., Jagschies, G., Schulenberg, H., and Spener, F. (1984) Hoppe-Seyler’s 2. Physiol. Chem. 365, 365-376. 18. Takahashi, K., Odani, S., and Ono, T. (1983) Eur. J. Biochem. 136, 589-601. 19. Matarese, V., Stone, R. L., Waggoner, D. W., and Bernlohr, D. A. (1989) Prog. Lipid Res. 28, 245-272. 20. Shimizu, S., Yasui, K., Tani, Y., and Yamada, H. (1979) Biochem. Biophys. Res. Commun. 91, 108-113. 21. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193,265-275. 22. Folch, J., Lees, M., and Sloane-Stanley, G. H. (1957) J. Biol. Chem. 226, 497-509. 23. Glass, R. L. (1971) Lipids 6, 919-925. 24. Glatz, J. F. C., and Veerkamp, J. H. (1983) Anal. Biochem. 132,
89-95. 25. Glatz, J. F. C., Baerwaldt, C. C. F., Veerkamp, J. H., and Kempen, H. J. M. (1984) J. Biol. Chem. 259,4295-4300. 26. Weber, K., and Osborn, M. (1969) J. Biol. Chem. 244,4406-4412. 27. Ketterer, B., Tipping, E., Hackney, J. F., and Beale, D. (1976) Bi0chem.J. 155,511-521. 28. Basu, S. P., Rao, S. N., and Hartsuck, J. A. (1978) Biochim. Biophys. Acta 533, 66-73. 29. Spencer, E. M., and King, T. P. (1971) J. Biol. Chem. 246, 201208.
