- Email: [email protected]
Purification and characterization of fatty acid-binding protein from rat kidney
ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS Vol. 254, No. 2, May 1, pp. 552-558,1987
Purification SATOSHI
and Characterization of Fatty Acid-Binding from Rat Kidney’
FUJII,
HIDEAKI
Department of Cardiovascular Received
KAWAGUCHI,
AND
HISAKAZU
Protein YASUDA
Medicine, Hokkaido University School of Medicine, Sappwro 060, Japan August
15, 1986, and in revised
form
January
13, 1987
We detected the presence of a fatty acid-binding protein (FABP) in rat kidney cytosols. This protein was eluted and purified 9.3-fold by sequential gel filtration and anionexchange chromatography. Homogeneity was shown by a single band on polyacrylamide gel with a molecular weight of about 15,500. It had an optimum binding pH of 7.4. The binding of palmitate to the protein was saturable. Examination of fatty acid binding revealed the presence of a single class of fatty acid-binding sites. The apparent dissociation constant was 1.0 PM and the maximal binding capacity was 48 nmol/mg of protein. This protein showed similar binding characteristics for palmitate, oleate, and arachidonate. Rabbit antibody to this cytosolic FABP gave a single precipitin line with the antigen and selectively inhibited [14C]palmitate binding to the protein. 0 1987 Academic Press, Inc.
In the cytosol of many tissues which use and transport long-chain fatty acids, free fatty acids bind to a cytoplasmic protein that aids their distribution to various pools and pathways, affecting many enzyme reactions in the fatty acid metabolism. This protein, with a molecular weight of about 12,000, has been identified in various tissues; for example, intestinal mucosa, liver, myocardium, and kidney (l-4). The best characterized of these binding proteins is that of rat liver (5-lo), and is called Z protein. The precise physiological function of this protein is not known in detail, but it seems to play an important role as a carrier protein in intracellular lipid metabolism because it is capable of binding fatty acids and possibly even transporting them through cytosol. Although the existence of this protein also has been reported previously in kidney (l), no definitive study of its precise char’ Supported by Grant-in-Aid 59570346 from the Ministry Culture of Japan. ‘To whom correspondence 6693-9861/87 Copyright All rights
acter and function has been carried out. Because of our interest in the fatty acid metabolism in kidney, we set out to isolate renal fatty acid-binding protein (FABP)3 and to evaluate its binding characteristics. Here, the purification and partial characterization of FABP from the rat kidney cytosolic fraction is reported. The possible relationship of this protein to Z protein is discussed also. MATERIALS
AND
METHODS
Materials. The following chemicals were obtained from commercial sources. [1-i4C]Palmitic acid (58 mCi/ mmol), [I-“Cloleic acid (57 mCi/mmol), and [l“Clarachidonic acid (55 mCi/mmol) were purchased from Amersham (UK); nonradiolabeled palmitic acid (99%) and bovine serum albumin were from Sigma (St. Louis, MO); Sephadex G-75 was from Pharmacia (Uppsala, Sweden); DEAE-cellulose (DE-52) was from Whatman (Springfield, UK); and charcoal and dextran were from Wako Pure Chemical Industries (Tokyo, Japan). All other chemicals were of analytical grade. Rabbit anti-rat liver FABP DE II IgG (7) was kindly donated by Professor T. Ono (Niigata, Japan).
for Scientific Research of Education, Science and should
$3.66
0 1987 hy Academic Press, Inc. of reproduction in any form resewed.
‘Abbreviations used: FABP, fatty protein; SDS, sodium dodecyl sulfate.
be addressed.
552
acid-binding
FATTY
ACID-BINDING
PROTEIN
Preparation of rat kidney c@aaol Kidneys from male albino Wistar rats were excised and perfused immediately with cold 0.25 M sucrose. The kidneys were minced with scissors and homogenized with a Polytron homogenizer (PT-10, Kinematica, Switzerland) at setting ‘7 for 60 s in 10 ~0110 mM Tris-HCl (pH 7.4), 10 mM KCl, 1 mM EDTA, 1 mM dithiothreitol (buffer A) and then with a Potter-Elvehjem homogenizer filled with a Teflon pestle. The total homogenate was centrifuged at 3000g for 10 min at 4°C. The supernatant was centrifuged again at 105,OOOg for 90 min at 4°C. The resultant supernatant, exclusive of floating fat, was used as the cytosolic fraction of the FABP. Purlficatim of fatty acid-binding protein from rat kidney cytosol The 105,000~ supernatant of rat kidney was concentrated by ultrafiltration (UK-lo, Toyoroshi, Tokyo, Japan) and 30-35 mg of the protein was applied to a Sephadex G-75 column (2.6 X 70 cm) equilibrated with buffer A. The fractions with high binding activity for palmitate were designated as FABP. They were then combined, dialyzed against 30 mM Tris-HCl (pH 8.5), and applied to a DEAE-cellulose column (2.1 X 15 cm) equilibrated with 30 mM Tris-HCl (pH 8.5). The column was initially eluted with the equilibrating buffer until unbound proteins were completely eluted. Then the retained protein was eluted with a linear gradient prepared from 100 ml each of the equilibrating buffer and 0.3 M NaCl in the same buffer. Preparation of antZnn+ to fatty acid-bindiw protein Outbred New Zealand white rabbits were immunized with the purified FABP. Purified FABP (200 ag) was dissolved in 0.5 ml phosphate-buffered sodium chloride, mixed with an equal volume of complete Freund’s adjuvant and administered. After two booster injections over a period of 2 months, anti-FABP antisera were obtained. IgG fractions of the antisera were prepared by standard techniques (11). The presence and purity of the antibody were examined against purified FABP by radial double-immunodiffusion on agar plates by the method of Oucbterlony (12). Assay of fatty acid binding. [l-‘%]Fatty acid binding to the protein was measured by a slight modification of the method of Lee and Wiggert (13). Unless otherwise stated, incubation mixtures contained [l‘*C]palmitic acid and unlabeled palmitic acid (total 1 pM), 30 mM Tris-HCI (pH 7.4; buffer B), and cytosolic protein in a final volume of 500 pl. [1-‘%]Palmitic acid dissolved in ethanol in a 2.0 ml polyethylene vial was evaporated prior to the addition of other assay components. After incubation with shaking at 37’C for 10 min, the reaction was stopped by adding 50 ~1 of a charcoal-dextran mixture (5% charcoal, 1% dextran (w/v), 10 mM Tris, 10 rnrc KCl, 1 mM EDTA). Our preliminary study revealed that the recovery of protein was 95-98s and that unbound fatty acids were effectively adsorbed in this assay system. Following vig-
FROM
RAT
KIDNEY
553
orous mixing and centrifugation at 15,000~ for 5 min at 4°C 400-pl aliquots of the supernatant were withdrawn and the radioactivity was measured in a liquid scintillation counter. LMipidation and amino acid ana&ia Some samples of the purified protein were delipidated by the method of Glatz (14). For determination of the amino acid compositions, proteins were hydrolyzed with 6 M HCl at 110°C for 24 h in evacuated sealed tubes. Amino acids were analyzed with a Hitachi KLA 3B amino acid analyzer, using norleucine as the internal standard. Other methods. Sodium dodecyl sulfate (SDS)-polyacrylamide slab gel electrophoresis was carried out according to the procedure of Maize1 (15). Each lane was loaded with lo-25 pg of protein. Protein concentration was determined (i) by the method of Lowry et al. (16) with bovine serum albumin as the standard, or (ii) from amino acid analysis (pure preparations only).
RESULTS
Purification of Fatty Acid-Binding Protein from Rat Kidney In the first step of purification by Sephadex G-75 gel filtration, fatty acid-binding activities were present mainly in fractions corresponding to the molecular weights of 12,000-17,000 (Fig. 1A). Fatty acid-binding activity was also present in the high molecular weight region (>50,000) due to residual albumin. Some fatty acid-binding activity was also present in the region of molecular weights 1500-2000. These fractions, which were considered to contain a fatty acid-binding peptide (17-l@, were not studied further. The activity peaks of the residual albumin and the fatty acid-binding peptide were completely separated from the FABP fractions in this step. For further purification anion-exchange chromatography was used. The FABP fraction was then further purified. Part of the fraction was adsorbed weakly to the DEAEcellulose and eluted with the equilibrating buffer (30 mM Tris-HCl, pH 8.5), but the remainder was eluted by applying a linear gradient of NaCl (Fig. 1B). The fraction with high-binding activity gave rise to a single band with a molecular weight of 15,500 on 15% polyacrylamide gel electrophoresis in the presence of 0.1% SDS (Fig. 2). The FABP fractions were dialyzed
554
FUJII,
KAWAGUCHI,
AND
YASUDA
FIG. 1. Purification of FABP from rat kidney. (A) Gel filtration of rat kidney 105,OOOg supernatant on a Sephadex G-75 column (2.6 X 70 cm). The supernatant (32.4 mg protein) was chromatographed on a Sephadex G-75 column equilibrated with 10 mM Tris-HCl (pH 7.4), 10 mM KCl, 1 mM EDTA, 1 mM dithiothreitol (buffer A). Fraction volumes of 7 ml each were collected at a flow rate of 21 ml/ h. (B) Chromatography on DEAE-cellulose of FABP. The FABP fractions of the Sephadex G-75 column (13.2 mg protein) were chromatographed on a DEAE-cellulose column with 30 mM TrisHCl (pH 8.5), and subsequently with a linear gradient prepared from 100 ml each of the equilibrating buffer and 0.3 M NaCl in the same buffer. Fraction volumes of 5 ml each were collected at a flow rate of 15 ml/h. In A and B, fractions were assayed for protein by measuring& ( * * *) and palmitate binding (0 - 0). Binding activity was measured as described under Materials and Methods. Results of representative experiments are shown.
against 30 mM Tris-HCl (pH 7.4) and the binding activity of each fraction of the DEAE-cellulose eluate was measured. The fractions with the highest binding activity were pooled and used as the final rat kidney FABP. FABP (10 pg) was labeled by a tracer amount of [14C]palmitate and resolved in the sample buffer (it was not treated by a heat-denaturation procedure for the electrophoresis) and subjected to 0.1% SDS-polyacrylamide gel electrophoresis. Our preliminary experiments revealed that the presence of up to 0.3% SDS did not interfere with the binding of fatty acid to FABP. Approximately 60% of the total recovered radioactivity was found at the position corresponding to the band of molecular weight 15,500 (results not shown). The rest of the radioactivity was found to have comigrated with the tracking dye, suggesting that part of the [‘“Clpalmitate was released from the protein during the electrophoresis procedure. A summary of typical purification procedures, as routinely carried out, is given in Table I. To calculate the specific binding precisely, protein concentration was determined by using amino acid analysis of pure preparations. The rat kidney FABP
was finally purified approximately with a 50% yield.
9.3-fold
Pmpertks of Pur$ed Rat Kidney Acid-Binding Protein
Fatty
Fatty acid binding to the freshly prepared delipidated binding protein was studied at 3’7°C. Under our standard assay conditions [14Clpalmitate binding increased depending on the amount of FABP (results not shown). Scatchard analysis revealed a single class of high-affinity binding sites with an apparent dissociation constant (Kd) of 1.0 PM. The maximal binding capacity (B,& was 48 nmol/mg of protein (Fig. 3). The Kd values for oleate and arachidonate, calculated in the same way, were 1.0 and 0.9 j&M, respectively. The maximal binding capacities for oleate and arachidonate were 54.4 and 44.5 nmol/mg protein, respectively. Binding of [14C]palmitate to purified FABP was a time-dependent process; the binding activity rose rapidly until 5 min, reached its maximum after 10 min, and gradually declined thereafter (results not shown). The optimum pH range of binding was between 7.2 and 7.6, with the maximum binding observed at pH
FATTY
ACID-BINDING
PROTEIN
Mr (x103)
RAT
555
KIDNEY
20 and 0°C markedly decreased binding activity (to 3.2 and 1.4 nmol/mg of protein, respectively). A decrease of binding activity of purified FABP occurred gradually with storage at -70°C (about 10% loss after 2 months). Under our standard assay conditions fatty acid binding by FABP was about 30% lower without delipidation when assayed with 0.3 PM palmitate, but not when assayed using 1 PM palmitate (results not shown). Competition from endogenous fatty acid was observed to be more marked at low palmitate concentrations. To evaluate the effect of delipidation on fatty acid binding, delipidation was repeated twice in some experiments. Fatty acid-binding activity was not altered further after the second delipidation.
92.5 66.2 45
31
21.5
14.4
FROM
1
1
2
3
Immunological Studies of Fatty AcidBinding Protein
4
FIG. 2. Polyacrylamide slab gel electrophoresis in 0.1% SDS of FABP-containing fractions from rat kidney at various stages of purification, Lane 1, proteins of known subunit molecular weights used for calibration in the gel system (with M,): lysozyme (14,400), soybean trypsin inhibitor (21,500), carbonic anhydrase (31,000), ovalbumin (45,000), bovine serum albumin (66,200), and phosphorylase b (92,500); lane 2, rat kidney cytosol (25 gg protein); lane 3, FABP fractions combined after Sephadex G-75 gel filtration (10 pg protein); lane 4, fraction from DEAE-cellulose column (10 pg protein).
7.4. Fatty acid binding was virtually unobserved at pH values below 6.8 or above 8.4. Lowering of the incubation temperature to TABLE PURIFICATION
Fraction 105,OOOg Supernatant Sephadex G-75 DEAE-cellulose
OF FATTY
ACID-BINDING
In a double-immunodiffusion test, rabbit anti-rat kidney FABP IgG formed a single precipitin line with purified rat kidney FABP and with rat kidney cytosol (Fig. 4), whereas no precipitin line was observed between the rabbit anti-rat liver Z protein IgG and purified rat kidney FABP, indicating that rat kidney FABP can be discriminated immunologically from rat liver Z protein. In studies employing various concentrations of rabbit anti-rat kidney FABP IgG with constant amounts of FABP and palmitate, antibody against the FABP inhibited binding of [14C]palmitate to FABP, I PROTEIN
FROM RAT KIDNEY
Total activity
Yield (%)
32.4'
71280
100
13.2” 1.8”
60720 36720
85 50
Protein (mg)
a Results are representative of four purifications. * Measured with 1 NM [1-‘“Cbalmitate at 37°C. “Determined by the method of Lowry. *Determined with part of the supernatant after dealbuminzation. e Determined from amino acid analysis.
CYTOSOL~
Specific activity* (pm011 P&d 2.2* 4.6 20.4
Purification (fold) 1 2.1 9.3
556
FUJII,
KAWAGUCHI,
AND
YASUDA
controls. In incubations containing the same amounts of rabbit anti-rat kidney FABP IgG and palmitate but various amounts of FABP, a linear relationship was observed between the logarithm of the FABP concentration and the percentage inhibition of palmitate binding, as determined by comparison with control incubations containing the control IgG fraction (results not shown). O0
20
10 BOUND
(pmol/pg
30
40
1
50
protein)
FIG. 3. Scatchard plot of the binding of palmitate by purified FABP from rat kidney. Freshly purified FABP (5 pg) was incubated with palmitate in a total volume of 500 ~1 buffer B solution at 37°C. After equilibration, protein-bound and unbound palmitate were separated by using charcoal-dextran. The ratio of bound to free (B/F) [“C]palmitate against the concentration of bound [14C]palmitate is shown. The concentration of free fatty acids was calculated from the initial concentration and the amount of protein-bound fatty acid. Protein was determined from amino acid analysis. Data shown are mean values from four incubations (SD < f5% in all cases).
whereas the control IgG fraction (obtained from nonimmune rabbit sera) did not affect palmitate binding to FABP (Fig. 5). The anti-FABP antibody inhibited binding by a maximum of 45% compared with the
Amino
Acid Anal&s
The amino acid composition of kidney FABP (Table II) has not been reported until now and appears to be different from that of liver protein and heart protein (7, 19). DISCUSSION
Although the presence of FABP in the kidney has been previously indicated (l), its precise nature has remained unstudied. This is especially true of the kinetic aspects of the interactions between renal FABP and fatty acids. The results of this study showed that there is FABP in rat kidney cytosol and that this FABP is closely related to, but essentially different from, rat liver Z protein. FABPs studied so far have similar molecular weights but show different isoelectric points. Rat liver FABP is
FIG. 4. Double-immunodiffusion test. The test was carried out at 4°C for 24 h in 1.2% agar (Difco Lab.) containing 0.02 M sodium phosphate buffer, pH 8.5, and 0.02% NaNs. Well A, rabbit anti-rat kidney FABP IgG (1 mg); well B, rabbit anti-rat liver FABP IgG (1 mg); wells 1 and 3, purified rat kidney FABP (50 fig); wells 2 and 4, rat kidney cytosol(250 pg).
FATTY
ACID-BINDING
Anti-FABP
PROTEIN
IgG
o0
25 IgG
50 added
75
100
Cpg,
FIG. 5. Effect of the IgG fraction of rabbit antibody on the FABP (anti FABP) (0) and a control IgG fraction (made from nonimmune rabbit sera) (0) on total [Wlpalmitate binding to FABP. FABP (5 pg) was incubated in a total volume of 500 pl buffer B solution for 30 min at 37°C with various amounts of IgG and then incubated for an additional 10 min with 1 pM [i4Cjpalmitate. Data shown are mean values from four incubations (SD < +5% in all cases).
reported to consist of three proteins with the same molecular weight but with different pIvalues (5, ‘7,20). The acidic forms may be denatured protein or may be due to various ligands bound to the protein (5, 21). Therefore, delipidation of FABP appears to be necessary for characterization. The Kd values of rat kidney FABP, to our knowledge, have not yet been reported. With freshly purified kidney FABP, our value of&for palmitate (1.0 PM) correlates well with the value reported for heart and liver FABP (19). Due to the loss of binding activity by partial denaturation during purification steps, it is probable that the maximal binding capacity of rat kidney FABP might be underestimated, and therefore is lower than the value of liver FABP and similar to the value of the heart FABP (19). The binding capacities for other fatty acids were also slightly variable, probably due to the difficulty of protein determination and to the small amounts of purified FABP available for binding assays. FABP from rat kidney has a molecular weight and palmitate binding property similar to those from heart and liver
FROM
RAT
557
KIDNEY
FABP. However, the amino acid composition of kidney FABP (Table II) seems different and more acidic than that from liver FABP, judging from the relatively large number of acidic residues. The non-identity of Z protein and rat brain FABP has also been reported (22). The utilization of our fatty acid-binding assay system made a sensitive and precise detection of the functional activity of FABP in column fractions possible. This is better than the methods by which FABP is detected by coelution of labeled fatty acids or sulphobromophthalein from a gel filtration column, since in these techniques fatty acids also bind to the gel itself. Furthermore, when our system is used, the relative activity of FABP is estimated more easily and the characterization of FABP is made more easily. Interference of charcoal against protein recovery has been reported (23); however, when small but sufficient quantities of charcoal-dextran were used, interference was negligible and protein recovery was not affected in our assay system. TABLE AMINO
Amino
ACID COMPOSITION ACID-BINDING acid
II OF RAT KIDNEY PROTEIN ’
Residues/1000
amino
acids
11 144 58 37 140 8 72 38 55 22 6’7 90 44 69 85 20 53 0
CYS Asx Thr Ser Glx Pro GUY Ala Val Met Ile Leu TY~ Phe LYS His Arg Tw a Values represent rations; SD i f5%
FATTY
means of four in all cases.
different
prepa-
558
FUJII,
KAWAGUCHI,
Our present study shows that rat kidney contains a distinct type of FABP which is essentially different from rat liver Z protein. It remains to be established, however, whether other FABPs such as those purified from rat heart (19) and brain (22) are different proteins. The various FABPs may perform various functions in the cellular metabolism of fatty acids, though these functions have not yet been established. As to kidney, it has been reported that prostaglandin synthesis and phospholipase activity are increased in spontaneously hypertensive rats (24,25). These results suggest that a unique pattern of fatty acid metabolism may be ongoing in spontaneously hypertensive rats and that FABP in kidney might be related to altered fatty acid metabolism through arachidonic acid transport. A functional assay system for FABP will help to establish its physiological and pathophysiological role in kidney and other tissues. ACKNOWLEDGMENTS We are grateful to Professor T. Ono (Department of Biochemistry, Niigata University School of Medicine) for supplying the anti-rat liver Z protein IgG, and to Dr. K. Kameda (Department of Biochemistry, Hokkaido University School of Medicine; head, Dr. T. Ishibashi) for stimulating discussions. The technical assistance of Dr. H. Okamoto is appreciated. REFERENCES 1. OCKNER, R. K., MANNING, J. A., POPPENHAUSEN, R. B., AND Ho, W. K. L. (1972) Science (Washington DC) 177,56-58. 2. MISHKIN, S., AND TURCO~, R. (1974) Biochem Biophys Res. Commun 60,376-381. 3. OCKNER, R. K., AND MANNING, J. A. (1974) J. Clin Invest. 54,326-338. 4. SAID, B., AND SCHULZ, H. (1984) J. BioL &em. 259, 1155-1159.
AND
YASUDA
5. TRIJLZSCH, D., AND ARIAS, I. M. (1981) A&L. B&hem Biophys 209,433-440. 6. OCKNER, R. K., MANNING, J. A., AND KANE, J. P. (1982) J. BioL Ch.em. 257,7872-7878. 7. TAKAHASHI, K., ODANI, S., AND ONO, T. (1983) Eur. J B&hem 136.589-601. 8. TAKAHASHI, K., ODANI, S., AND ONO, T. (1982) Biochem Biophys. Res. Commun. 106, 10991105. 9. TAKAHASHI, K., ODANI, S., AND ONO, T. (1982) FEBS Leti 140,63-66. 10. LEVI, A., GATMAITAN, Z., AND ARIAS, I. M. (1969) J. Clin Invest. 48,2156-2167. 11. KOCHWA, S. (1961) J. CZin. Invest. 40,874-883. 12. OUCHTERLONY, 0. (1958) Prog. AUe~gg 5,1-78. 13. LEE, L., AND WIGGERT, B. (1984) J. Neurochxm. 42, 47-53. 14. GLATZ, J. F. C., BAERWALDT, C. C. F., VEERKAMP, J. H., AND KEMPEN, H. J. M. (1984) J. BioL Chem 259,4295-4300. 15. MAIZEL, J. V. (1971) lMetb3.s ViroL 5,179~246. 16. LOWRY, 0. H., ROSENBROUGH, N. J., FARR, A. L., AND RANDALL, R. J. (1951) J. BioL Chem 193, 265-275. 17. SUZUE, G., AND MARCEL, Y. L. (1975) Canoo! J. Biochem 53,804-809. 18. R~STOW, B., HODI, J., KUNZE, D., REICHMANN, G., AND EGGER, E. (1978) FEBS Lett. 95,225-228. 19. GLATZ, J. F. C., JANSSEN, A. M., BAERWALDT, C. C. F., AND VEERKAMP, J. H. (1985) Biochim Biophys. Acta 837,57-66. 20. KETTERER, B., TIPPING, E., HACKNEY, J. F., AND BEALE, D. (1976) Biochem J. 155,511-521. 21. DEMPSEY, M. E., MCCOY, K. E., BAKER, H. N.,DIMITRIADOU-VARFIADOU, A., LORSBACH, T., AND HOWARD, J. B. (1981) J. BioL Chem 256,18671873. 22. SENJO, M., ISHIBASHI, T., IMAI, Y., TAKAHASHI, K., AND ONO, T. (1985) Arch Biochem Biophys 236, 662-668. 23. GLATZ, J. F. C., AND VEERKAMP, J. H. (1983)Anal. Bioch,em 132,89-95. 24. LIMAS, C., AND LIMA% C. J. (1979) Amer. J. PhpioL 5, H65-H72. 25. KAWAGUCHI, H., ISHIBASHI, T., AND IMAI, Y. (1981) tipids 16,37-41.
Purification SATOSHI
and Characterization of Fatty Acid-Binding from Rat Kidney’
FUJII,
HIDEAKI
Department of Cardiovascular Received
KAWAGUCHI,
AND
HISAKAZU
Protein YASUDA
Medicine, Hokkaido University School of Medicine, Sappwro 060, Japan August
15, 1986, and in revised
form
January
13, 1987
We detected the presence of a fatty acid-binding protein (FABP) in rat kidney cytosols. This protein was eluted and purified 9.3-fold by sequential gel filtration and anionexchange chromatography. Homogeneity was shown by a single band on polyacrylamide gel with a molecular weight of about 15,500. It had an optimum binding pH of 7.4. The binding of palmitate to the protein was saturable. Examination of fatty acid binding revealed the presence of a single class of fatty acid-binding sites. The apparent dissociation constant was 1.0 PM and the maximal binding capacity was 48 nmol/mg of protein. This protein showed similar binding characteristics for palmitate, oleate, and arachidonate. Rabbit antibody to this cytosolic FABP gave a single precipitin line with the antigen and selectively inhibited [14C]palmitate binding to the protein. 0 1987 Academic Press, Inc.
In the cytosol of many tissues which use and transport long-chain fatty acids, free fatty acids bind to a cytoplasmic protein that aids their distribution to various pools and pathways, affecting many enzyme reactions in the fatty acid metabolism. This protein, with a molecular weight of about 12,000, has been identified in various tissues; for example, intestinal mucosa, liver, myocardium, and kidney (l-4). The best characterized of these binding proteins is that of rat liver (5-lo), and is called Z protein. The precise physiological function of this protein is not known in detail, but it seems to play an important role as a carrier protein in intracellular lipid metabolism because it is capable of binding fatty acids and possibly even transporting them through cytosol. Although the existence of this protein also has been reported previously in kidney (l), no definitive study of its precise char’ Supported by Grant-in-Aid 59570346 from the Ministry Culture of Japan. ‘To whom correspondence 6693-9861/87 Copyright All rights
acter and function has been carried out. Because of our interest in the fatty acid metabolism in kidney, we set out to isolate renal fatty acid-binding protein (FABP)3 and to evaluate its binding characteristics. Here, the purification and partial characterization of FABP from the rat kidney cytosolic fraction is reported. The possible relationship of this protein to Z protein is discussed also. MATERIALS
AND
METHODS
Materials. The following chemicals were obtained from commercial sources. [1-i4C]Palmitic acid (58 mCi/ mmol), [I-“Cloleic acid (57 mCi/mmol), and [l“Clarachidonic acid (55 mCi/mmol) were purchased from Amersham (UK); nonradiolabeled palmitic acid (99%) and bovine serum albumin were from Sigma (St. Louis, MO); Sephadex G-75 was from Pharmacia (Uppsala, Sweden); DEAE-cellulose (DE-52) was from Whatman (Springfield, UK); and charcoal and dextran were from Wako Pure Chemical Industries (Tokyo, Japan). All other chemicals were of analytical grade. Rabbit anti-rat liver FABP DE II IgG (7) was kindly donated by Professor T. Ono (Niigata, Japan).
for Scientific Research of Education, Science and should
$3.66
0 1987 hy Academic Press, Inc. of reproduction in any form resewed.
‘Abbreviations used: FABP, fatty protein; SDS, sodium dodecyl sulfate.
be addressed.
552
acid-binding
FATTY
ACID-BINDING
PROTEIN
Preparation of rat kidney c@aaol Kidneys from male albino Wistar rats were excised and perfused immediately with cold 0.25 M sucrose. The kidneys were minced with scissors and homogenized with a Polytron homogenizer (PT-10, Kinematica, Switzerland) at setting ‘7 for 60 s in 10 ~0110 mM Tris-HCl (pH 7.4), 10 mM KCl, 1 mM EDTA, 1 mM dithiothreitol (buffer A) and then with a Potter-Elvehjem homogenizer filled with a Teflon pestle. The total homogenate was centrifuged at 3000g for 10 min at 4°C. The supernatant was centrifuged again at 105,OOOg for 90 min at 4°C. The resultant supernatant, exclusive of floating fat, was used as the cytosolic fraction of the FABP. Purlficatim of fatty acid-binding protein from rat kidney cytosol The 105,000~ supernatant of rat kidney was concentrated by ultrafiltration (UK-lo, Toyoroshi, Tokyo, Japan) and 30-35 mg of the protein was applied to a Sephadex G-75 column (2.6 X 70 cm) equilibrated with buffer A. The fractions with high binding activity for palmitate were designated as FABP. They were then combined, dialyzed against 30 mM Tris-HCl (pH 8.5), and applied to a DEAE-cellulose column (2.1 X 15 cm) equilibrated with 30 mM Tris-HCl (pH 8.5). The column was initially eluted with the equilibrating buffer until unbound proteins were completely eluted. Then the retained protein was eluted with a linear gradient prepared from 100 ml each of the equilibrating buffer and 0.3 M NaCl in the same buffer. Preparation of antZnn+ to fatty acid-bindiw protein Outbred New Zealand white rabbits were immunized with the purified FABP. Purified FABP (200 ag) was dissolved in 0.5 ml phosphate-buffered sodium chloride, mixed with an equal volume of complete Freund’s adjuvant and administered. After two booster injections over a period of 2 months, anti-FABP antisera were obtained. IgG fractions of the antisera were prepared by standard techniques (11). The presence and purity of the antibody were examined against purified FABP by radial double-immunodiffusion on agar plates by the method of Oucbterlony (12). Assay of fatty acid binding. [l-‘%]Fatty acid binding to the protein was measured by a slight modification of the method of Lee and Wiggert (13). Unless otherwise stated, incubation mixtures contained [l‘*C]palmitic acid and unlabeled palmitic acid (total 1 pM), 30 mM Tris-HCI (pH 7.4; buffer B), and cytosolic protein in a final volume of 500 pl. [1-‘%]Palmitic acid dissolved in ethanol in a 2.0 ml polyethylene vial was evaporated prior to the addition of other assay components. After incubation with shaking at 37’C for 10 min, the reaction was stopped by adding 50 ~1 of a charcoal-dextran mixture (5% charcoal, 1% dextran (w/v), 10 mM Tris, 10 rnrc KCl, 1 mM EDTA). Our preliminary study revealed that the recovery of protein was 95-98s and that unbound fatty acids were effectively adsorbed in this assay system. Following vig-
FROM
RAT
KIDNEY
553
orous mixing and centrifugation at 15,000~ for 5 min at 4°C 400-pl aliquots of the supernatant were withdrawn and the radioactivity was measured in a liquid scintillation counter. LMipidation and amino acid ana&ia Some samples of the purified protein were delipidated by the method of Glatz (14). For determination of the amino acid compositions, proteins were hydrolyzed with 6 M HCl at 110°C for 24 h in evacuated sealed tubes. Amino acids were analyzed with a Hitachi KLA 3B amino acid analyzer, using norleucine as the internal standard. Other methods. Sodium dodecyl sulfate (SDS)-polyacrylamide slab gel electrophoresis was carried out according to the procedure of Maize1 (15). Each lane was loaded with lo-25 pg of protein. Protein concentration was determined (i) by the method of Lowry et al. (16) with bovine serum albumin as the standard, or (ii) from amino acid analysis (pure preparations only).
RESULTS
Purification of Fatty Acid-Binding Protein from Rat Kidney In the first step of purification by Sephadex G-75 gel filtration, fatty acid-binding activities were present mainly in fractions corresponding to the molecular weights of 12,000-17,000 (Fig. 1A). Fatty acid-binding activity was also present in the high molecular weight region (>50,000) due to residual albumin. Some fatty acid-binding activity was also present in the region of molecular weights 1500-2000. These fractions, which were considered to contain a fatty acid-binding peptide (17-l@, were not studied further. The activity peaks of the residual albumin and the fatty acid-binding peptide were completely separated from the FABP fractions in this step. For further purification anion-exchange chromatography was used. The FABP fraction was then further purified. Part of the fraction was adsorbed weakly to the DEAEcellulose and eluted with the equilibrating buffer (30 mM Tris-HCl, pH 8.5), but the remainder was eluted by applying a linear gradient of NaCl (Fig. 1B). The fraction with high-binding activity gave rise to a single band with a molecular weight of 15,500 on 15% polyacrylamide gel electrophoresis in the presence of 0.1% SDS (Fig. 2). The FABP fractions were dialyzed
554
FUJII,
KAWAGUCHI,
AND
YASUDA
FIG. 1. Purification of FABP from rat kidney. (A) Gel filtration of rat kidney 105,OOOg supernatant on a Sephadex G-75 column (2.6 X 70 cm). The supernatant (32.4 mg protein) was chromatographed on a Sephadex G-75 column equilibrated with 10 mM Tris-HCl (pH 7.4), 10 mM KCl, 1 mM EDTA, 1 mM dithiothreitol (buffer A). Fraction volumes of 7 ml each were collected at a flow rate of 21 ml/ h. (B) Chromatography on DEAE-cellulose of FABP. The FABP fractions of the Sephadex G-75 column (13.2 mg protein) were chromatographed on a DEAE-cellulose column with 30 mM TrisHCl (pH 8.5), and subsequently with a linear gradient prepared from 100 ml each of the equilibrating buffer and 0.3 M NaCl in the same buffer. Fraction volumes of 5 ml each were collected at a flow rate of 15 ml/h. In A and B, fractions were assayed for protein by measuring& ( * * *) and palmitate binding (0 - 0). Binding activity was measured as described under Materials and Methods. Results of representative experiments are shown.
against 30 mM Tris-HCl (pH 7.4) and the binding activity of each fraction of the DEAE-cellulose eluate was measured. The fractions with the highest binding activity were pooled and used as the final rat kidney FABP. FABP (10 pg) was labeled by a tracer amount of [14C]palmitate and resolved in the sample buffer (it was not treated by a heat-denaturation procedure for the electrophoresis) and subjected to 0.1% SDS-polyacrylamide gel electrophoresis. Our preliminary experiments revealed that the presence of up to 0.3% SDS did not interfere with the binding of fatty acid to FABP. Approximately 60% of the total recovered radioactivity was found at the position corresponding to the band of molecular weight 15,500 (results not shown). The rest of the radioactivity was found to have comigrated with the tracking dye, suggesting that part of the [‘“Clpalmitate was released from the protein during the electrophoresis procedure. A summary of typical purification procedures, as routinely carried out, is given in Table I. To calculate the specific binding precisely, protein concentration was determined by using amino acid analysis of pure preparations. The rat kidney FABP
was finally purified approximately with a 50% yield.
9.3-fold
Pmpertks of Pur$ed Rat Kidney Acid-Binding Protein
Fatty
Fatty acid binding to the freshly prepared delipidated binding protein was studied at 3’7°C. Under our standard assay conditions [14Clpalmitate binding increased depending on the amount of FABP (results not shown). Scatchard analysis revealed a single class of high-affinity binding sites with an apparent dissociation constant (Kd) of 1.0 PM. The maximal binding capacity (B,& was 48 nmol/mg of protein (Fig. 3). The Kd values for oleate and arachidonate, calculated in the same way, were 1.0 and 0.9 j&M, respectively. The maximal binding capacities for oleate and arachidonate were 54.4 and 44.5 nmol/mg protein, respectively. Binding of [14C]palmitate to purified FABP was a time-dependent process; the binding activity rose rapidly until 5 min, reached its maximum after 10 min, and gradually declined thereafter (results not shown). The optimum pH range of binding was between 7.2 and 7.6, with the maximum binding observed at pH
FATTY
ACID-BINDING
PROTEIN
Mr (x103)
RAT
555
KIDNEY
20 and 0°C markedly decreased binding activity (to 3.2 and 1.4 nmol/mg of protein, respectively). A decrease of binding activity of purified FABP occurred gradually with storage at -70°C (about 10% loss after 2 months). Under our standard assay conditions fatty acid binding by FABP was about 30% lower without delipidation when assayed with 0.3 PM palmitate, but not when assayed using 1 PM palmitate (results not shown). Competition from endogenous fatty acid was observed to be more marked at low palmitate concentrations. To evaluate the effect of delipidation on fatty acid binding, delipidation was repeated twice in some experiments. Fatty acid-binding activity was not altered further after the second delipidation.
92.5 66.2 45
31
21.5
14.4
FROM
1
1
2
3
Immunological Studies of Fatty AcidBinding Protein
4
FIG. 2. Polyacrylamide slab gel electrophoresis in 0.1% SDS of FABP-containing fractions from rat kidney at various stages of purification, Lane 1, proteins of known subunit molecular weights used for calibration in the gel system (with M,): lysozyme (14,400), soybean trypsin inhibitor (21,500), carbonic anhydrase (31,000), ovalbumin (45,000), bovine serum albumin (66,200), and phosphorylase b (92,500); lane 2, rat kidney cytosol (25 gg protein); lane 3, FABP fractions combined after Sephadex G-75 gel filtration (10 pg protein); lane 4, fraction from DEAE-cellulose column (10 pg protein).
7.4. Fatty acid binding was virtually unobserved at pH values below 6.8 or above 8.4. Lowering of the incubation temperature to TABLE PURIFICATION
Fraction 105,OOOg Supernatant Sephadex G-75 DEAE-cellulose
OF FATTY
ACID-BINDING
In a double-immunodiffusion test, rabbit anti-rat kidney FABP IgG formed a single precipitin line with purified rat kidney FABP and with rat kidney cytosol (Fig. 4), whereas no precipitin line was observed between the rabbit anti-rat liver Z protein IgG and purified rat kidney FABP, indicating that rat kidney FABP can be discriminated immunologically from rat liver Z protein. In studies employing various concentrations of rabbit anti-rat kidney FABP IgG with constant amounts of FABP and palmitate, antibody against the FABP inhibited binding of [14C]palmitate to FABP, I PROTEIN
FROM RAT KIDNEY
Total activity
Yield (%)
32.4'
71280
100
13.2” 1.8”
60720 36720
85 50
Protein (mg)
a Results are representative of four purifications. * Measured with 1 NM [1-‘“Cbalmitate at 37°C. “Determined by the method of Lowry. *Determined with part of the supernatant after dealbuminzation. e Determined from amino acid analysis.
CYTOSOL~
Specific activity* (pm011 P&d 2.2* 4.6 20.4
Purification (fold) 1 2.1 9.3
556
FUJII,
KAWAGUCHI,
AND
YASUDA
controls. In incubations containing the same amounts of rabbit anti-rat kidney FABP IgG and palmitate but various amounts of FABP, a linear relationship was observed between the logarithm of the FABP concentration and the percentage inhibition of palmitate binding, as determined by comparison with control incubations containing the control IgG fraction (results not shown). O0
20
10 BOUND
(pmol/pg
30
40
1
50
protein)
FIG. 3. Scatchard plot of the binding of palmitate by purified FABP from rat kidney. Freshly purified FABP (5 pg) was incubated with palmitate in a total volume of 500 ~1 buffer B solution at 37°C. After equilibration, protein-bound and unbound palmitate were separated by using charcoal-dextran. The ratio of bound to free (B/F) [“C]palmitate against the concentration of bound [14C]palmitate is shown. The concentration of free fatty acids was calculated from the initial concentration and the amount of protein-bound fatty acid. Protein was determined from amino acid analysis. Data shown are mean values from four incubations (SD < f5% in all cases).
whereas the control IgG fraction (obtained from nonimmune rabbit sera) did not affect palmitate binding to FABP (Fig. 5). The anti-FABP antibody inhibited binding by a maximum of 45% compared with the
Amino
Acid Anal&s
The amino acid composition of kidney FABP (Table II) has not been reported until now and appears to be different from that of liver protein and heart protein (7, 19). DISCUSSION
Although the presence of FABP in the kidney has been previously indicated (l), its precise nature has remained unstudied. This is especially true of the kinetic aspects of the interactions between renal FABP and fatty acids. The results of this study showed that there is FABP in rat kidney cytosol and that this FABP is closely related to, but essentially different from, rat liver Z protein. FABPs studied so far have similar molecular weights but show different isoelectric points. Rat liver FABP is
FIG. 4. Double-immunodiffusion test. The test was carried out at 4°C for 24 h in 1.2% agar (Difco Lab.) containing 0.02 M sodium phosphate buffer, pH 8.5, and 0.02% NaNs. Well A, rabbit anti-rat kidney FABP IgG (1 mg); well B, rabbit anti-rat liver FABP IgG (1 mg); wells 1 and 3, purified rat kidney FABP (50 fig); wells 2 and 4, rat kidney cytosol(250 pg).
FATTY
ACID-BINDING
Anti-FABP
PROTEIN
IgG
o0
25 IgG
50 added
75
100
Cpg,
FIG. 5. Effect of the IgG fraction of rabbit antibody on the FABP (anti FABP) (0) and a control IgG fraction (made from nonimmune rabbit sera) (0) on total [Wlpalmitate binding to FABP. FABP (5 pg) was incubated in a total volume of 500 pl buffer B solution for 30 min at 37°C with various amounts of IgG and then incubated for an additional 10 min with 1 pM [i4Cjpalmitate. Data shown are mean values from four incubations (SD < +5% in all cases).
reported to consist of three proteins with the same molecular weight but with different pIvalues (5, ‘7,20). The acidic forms may be denatured protein or may be due to various ligands bound to the protein (5, 21). Therefore, delipidation of FABP appears to be necessary for characterization. The Kd values of rat kidney FABP, to our knowledge, have not yet been reported. With freshly purified kidney FABP, our value of&for palmitate (1.0 PM) correlates well with the value reported for heart and liver FABP (19). Due to the loss of binding activity by partial denaturation during purification steps, it is probable that the maximal binding capacity of rat kidney FABP might be underestimated, and therefore is lower than the value of liver FABP and similar to the value of the heart FABP (19). The binding capacities for other fatty acids were also slightly variable, probably due to the difficulty of protein determination and to the small amounts of purified FABP available for binding assays. FABP from rat kidney has a molecular weight and palmitate binding property similar to those from heart and liver
FROM
RAT
557
KIDNEY
FABP. However, the amino acid composition of kidney FABP (Table II) seems different and more acidic than that from liver FABP, judging from the relatively large number of acidic residues. The non-identity of Z protein and rat brain FABP has also been reported (22). The utilization of our fatty acid-binding assay system made a sensitive and precise detection of the functional activity of FABP in column fractions possible. This is better than the methods by which FABP is detected by coelution of labeled fatty acids or sulphobromophthalein from a gel filtration column, since in these techniques fatty acids also bind to the gel itself. Furthermore, when our system is used, the relative activity of FABP is estimated more easily and the characterization of FABP is made more easily. Interference of charcoal against protein recovery has been reported (23); however, when small but sufficient quantities of charcoal-dextran were used, interference was negligible and protein recovery was not affected in our assay system. TABLE AMINO
Amino
ACID COMPOSITION ACID-BINDING acid
II OF RAT KIDNEY PROTEIN ’
Residues/1000
amino
acids
11 144 58 37 140 8 72 38 55 22 6’7 90 44 69 85 20 53 0
CYS Asx Thr Ser Glx Pro GUY Ala Val Met Ile Leu TY~ Phe LYS His Arg Tw a Values represent rations; SD i f5%
FATTY
means of four in all cases.
different
prepa-
558
FUJII,
KAWAGUCHI,
Our present study shows that rat kidney contains a distinct type of FABP which is essentially different from rat liver Z protein. It remains to be established, however, whether other FABPs such as those purified from rat heart (19) and brain (22) are different proteins. The various FABPs may perform various functions in the cellular metabolism of fatty acids, though these functions have not yet been established. As to kidney, it has been reported that prostaglandin synthesis and phospholipase activity are increased in spontaneously hypertensive rats (24,25). These results suggest that a unique pattern of fatty acid metabolism may be ongoing in spontaneously hypertensive rats and that FABP in kidney might be related to altered fatty acid metabolism through arachidonic acid transport. A functional assay system for FABP will help to establish its physiological and pathophysiological role in kidney and other tissues. ACKNOWLEDGMENTS We are grateful to Professor T. Ono (Department of Biochemistry, Niigata University School of Medicine) for supplying the anti-rat liver Z protein IgG, and to Dr. K. Kameda (Department of Biochemistry, Hokkaido University School of Medicine; head, Dr. T. Ishibashi) for stimulating discussions. The technical assistance of Dr. H. Okamoto is appreciated. REFERENCES 1. OCKNER, R. K., MANNING, J. A., POPPENHAUSEN, R. B., AND Ho, W. K. L. (1972) Science (Washington DC) 177,56-58. 2. MISHKIN, S., AND TURCO~, R. (1974) Biochem Biophys Res. Commun 60,376-381. 3. OCKNER, R. K., AND MANNING, J. A. (1974) J. Clin Invest. 54,326-338. 4. SAID, B., AND SCHULZ, H. (1984) J. BioL &em. 259, 1155-1159.
AND
YASUDA
5. TRIJLZSCH, D., AND ARIAS, I. M. (1981) A&L. B&hem Biophys 209,433-440. 6. OCKNER, R. K., MANNING, J. A., AND KANE, J. P. (1982) J. BioL Ch.em. 257,7872-7878. 7. TAKAHASHI, K., ODANI, S., AND ONO, T. (1983) Eur. J B&hem 136.589-601. 8. TAKAHASHI, K., ODANI, S., AND ONO, T. (1982) Biochem Biophys. Res. Commun. 106, 10991105. 9. TAKAHASHI, K., ODANI, S., AND ONO, T. (1982) FEBS Leti 140,63-66. 10. LEVI, A., GATMAITAN, Z., AND ARIAS, I. M. (1969) J. Clin Invest. 48,2156-2167. 11. KOCHWA, S. (1961) J. CZin. Invest. 40,874-883. 12. OUCHTERLONY, 0. (1958) Prog. AUe~gg 5,1-78. 13. LEE, L., AND WIGGERT, B. (1984) J. Neurochxm. 42, 47-53. 14. GLATZ, J. F. C., BAERWALDT, C. C. F., VEERKAMP, J. H., AND KEMPEN, H. J. M. (1984) J. BioL Chem 259,4295-4300. 15. MAIZEL, J. V. (1971) lMetb3.s ViroL 5,179~246. 16. LOWRY, 0. H., ROSENBROUGH, N. J., FARR, A. L., AND RANDALL, R. J. (1951) J. BioL Chem 193, 265-275. 17. SUZUE, G., AND MARCEL, Y. L. (1975) Canoo! J. Biochem 53,804-809. 18. R~STOW, B., HODI, J., KUNZE, D., REICHMANN, G., AND EGGER, E. (1978) FEBS Lett. 95,225-228. 19. GLATZ, J. F. C., JANSSEN, A. M., BAERWALDT, C. C. F., AND VEERKAMP, J. H. (1985) Biochim Biophys. Acta 837,57-66. 20. KETTERER, B., TIPPING, E., HACKNEY, J. F., AND BEALE, D. (1976) Biochem J. 155,511-521. 21. DEMPSEY, M. E., MCCOY, K. E., BAKER, H. N.,DIMITRIADOU-VARFIADOU, A., LORSBACH, T., AND HOWARD, J. B. (1981) J. BioL Chem 256,18671873. 22. SENJO, M., ISHIBASHI, T., IMAI, Y., TAKAHASHI, K., AND ONO, T. (1985) Arch Biochem Biophys 236, 662-668. 23. GLATZ, J. F. C., AND VEERKAMP, J. H. (1983)Anal. Bioch,em 132,89-95. 24. LIMAS, C., AND LIMA% C. J. (1979) Amer. J. PhpioL 5, H65-H72. 25. KAWAGUCHI, H., ISHIBASHI, T., AND IMAI, Y. (1981) tipids 16,37-41.