Mechanism of cellular fatty acid uptake

Biochemical Society Transactions

Mechanism of cellular fatty acid uptake W. Stremmel,* H. Kleinert, B. A. Fitscher, 1. Gunawan, C. Klaassen-Schluter, K. Moller and M. Wegener Division of Gastroenterology, Department of Internal Medicine, University Hospital Dusseldorf, Germany 814

Fatty acids are tightly bound to albumin within the circulation and represent an important, readily available energy source for certain organs like liver and heart. Controversy exists over how they are translocated across the plasma membrane [ 13. Since albumin is not taken up into the cells, only the dissociated unbound fatty acids are available for uptake. For evaluation of whether this uptake occurs by simple diffusion or facilitated transport, it was first examined whether fatty acids interact specifically with the plasma membrane. Therefore the binding characteristics of a representative long chain fatty acid, ['Hloleate, to isolated rat liver plasma membranes were analysed [2]. For these experiments oleate was complexed to albumin in various molar ratios which served to keep albumin in solution and to modulate the unbound oleate concentration in the medium. Saturable membrane binding characteristics were obtained with a K,, of 20 m i , indicating the presence of high affinity binding sites [2]. Furthermore, binding was inhibited by heat denaturation and trypsin pretreatment of the plasma membranes [2, 31. All of these observations suggested the presence of a membrane fatty acid 'receptor' protein. This suggestion was pursued by isolation of such a membrane fatty acid binding protein from rat liver plasma membranes. A single step affinity chromatography technique of detergent solubilized plasma membrane proteins over an oleate agarose gel revealed a single protein with a molecular weight of 40 kDa and a pI of 8.5-9.0 [2] (Fig. 1). This membrane fatty acid binding protein (MFABP) had no carbohydrate or lipid components, was poorly soluble in water, showed also affinity for long chain fatty acids in vitro, and was distinct from the 12 kDa cytosolic fatty acid binding protein (cFABP). It was isolated from plasma membranes of rat hepatocytes, cardiomyocytes and jejunal mucosal cells [2, 4, 51. The protein was clearly distinct from rat mitochondria1 glutamic oxaloacetic transaminase (mGOT) [6] which was recently suggested by others to represent the responsible carrier protein [ 7 ] . This group isolated the protein by a different isolation procedure using Abbreviations used: MFABP, membrane fatty acid binding protein; cFABP, cytosolic fatty acid binding protein *Address for correspondence: Prof. Dr. W. Stremmel, Dept. of Gastroenterology, Heinrich-Heine-University Dusseldorf, Moorenstr. 5,4000 Dusseldorf, Germany.

Volume 20

preparative isoelectric focusing and subsequent gel filtration of salt-extracted plasma membrane fractions as initial steps [8]. T o the MFARP polyclonal and monoclonal antibodies were raised.. The monoclonal antibody identified in 2-D immunoblots a 40 kDa, PI 8.5-9.0 protein pattern identical to the authentic MFABP [2, 91. Immunofluorescence studies and Western blot analyses revealed the presence of this protein in plasma membranes of various cell types with high energy requirement [2]. Moreover, the monoFig. I

SDS-PAGE of the membrane fatty acid binding protein (MFABP) isolated by oleate agarose affinity chromatography (lane A ) For comparison, Triton-X 100 solubilized total rat liver plasma membrane proteins (lane B), and molecular weight standards (lane C) are shown. The protein amounts applied t o the gel were 20 pg of MFABP and 500 pg of total solubilized plasma membranes. A silver staining procedure was employed. (Reproduced with permission from [9].) Lane A

Lane B

Lane C

Dalton

97.400 66.200 42.69:

31.000 '

21.500

14.40C

MFABP

PLASMA MEMBRANES

MW-STANDARDS

Lipid-BindingProteins

specific antibodies to MFAHP inhibited binding of long chain fatty acids to plasma membranes. The next question was whether this MFABP also has transport function and mediates the translocation of fatty acids across the plasma membrane. First it was analysed whether cellular uptake of fatty acids reveals criteria of a facilitated transport process. These studies were performed in isolated and short term cultured hepatocytes, cardiomyocytes, adipocytes, alveolar type I1 cells, and jejunal mucosal cells [4, 10-161. For these studies it was required that only the cellular influx component was measured, representing the translocation process across plasma membranes and that this component was the rate determining step, independent of intracellular metabolism including oxidation and esterification [ 171. Under these conditions oleate influx into the various cell types was determined as a function of the unbound oleate concentration in the medium. Modulation of the unbound oleate concentration was achieved by variation of the molar ratio of fatty acids to albumin from 0.25 :1 to 2: 1, keeping either the albumin or oleate concentration constant [2]. With increasing unbound oleate concentrations in the medium, uptake followed saturation kinetics in all cell types examined [4, 10-161. In a representative study with short term cultured hepatocytes, revealing a K , value of 90 nM and a V,,,, of 835 pmol min-' per mg of cell protein was obtained [ 101. This observation of saturable fatty acid uptake kinetics was considered to be a criterion of a carrier-mediated transport mechanism. Other criteria were inhibition of uptake by pretreatment of cells with trypsin or phloretin (inhibitor of membrane translocation mechanisms) as well as energy and sodium dependency of influx [ 11, 17, 181. For evaluation of whether this facilitated transport process is mediated by the identified MFABP, the effect of the monospecific antibody to this protein on cellular influx of oleate was determined. Fatty acid influx kinetics in the presence of the IgGfraction of the antiserum to MFABP were significantly inhibited compared with control preparations incubated with the IgG-fraction of the pre-immune serum (Fig. 2). This indicated the significance of MFABP as the responsible carrier protein. For evaluation of the specificity of this uptake process the effect of the antibody to MFABP on initial uptake of various long chain fatty acids (173 ,LAM, molar ratio of fatty acid:albumin = 1: 1) and representatives of other classes of organic anions (200 p ~ was ) examined. In short-term cultured hepatocytes pre-treated with the IgG fraction of the anti-

Fig. 2 Inhibition of influx rates of [3H]oleate into shortterm cultured hepatocytes by a monospecific antibody to the MFABP The cultures were pretreated with the IgG-fraction of the antiserum to MFABP or the IgG-fraction of the pre-immune serum. Illustrated are the initial uptake rates of [3H]oleate as a function of the unbound oleate concentration in the medium. (Reproduced with permission from [lo].) Oleate: albumin molar ratio falbumin constant 173 ,UM)

Anti fatty acid binding membrane protein

200 300 I00 Unbound oleate concentration (nM)

400

serum to MFABP, similar marked inhibition of cellular influx of oleic acid, arachidonic acid, palmitic acid and stearic acid by 5 8 4 9 % was observed. This provided evidence that the membrane fatty acid binding protein functions as a general carrier protein for long chain fatty acids. In contrast, uptake of sulphobromophthalein (HSP), cholic acid and taurocholic acid was not affected by the antibody. This suggests that they are taken up via a different transport mechanism. According to these results it is concluded that the translocation of long chain fatty acids across the sinusoidal membrane of hepatocytes involves:

1. dissociation of the albumidfatty acid complex at or near the cell surface; 2. binding to a specific membrane protein, which functions as a transmembrane carrier protein. At the cytoplasmic site of the plasma membrane fatty acids are bound to the cFABP serving as intracellular storage protein or directing fatty acids to the sites of further metabolism. The open question now was: What are the driving forces for the highly efficient cellular influx of fatty acid anions against the unfavorable electrical gradient inside the cell? For determination of the responsible driving forces for fatty acid uptake, basolateral rat liver plasma membrane vesicles were used as the experi-

I992

815

Biochemical Society Transactions

816

mental model [ 181. The advantage of vesicle studies is that they allow a direct evaluation of the membrane translocation process independently of intracellular metabolism. First, the Na+-dependency of uptake was analysed. Oleate uptake was examined in the presence of an inwardly directed Na+ gradient with 100 mM-NaC1 in the medium (Fig. 3). Under those conditions a rapid influx phase was observed with maximal accumulation of fatty acids after 20 s [ 181. This was followed by a slow decline as oleate effluxed from the vesicles until equilibrium was reached after 20 min. This typical overshoot phenomenon is characteristic of an active transport Fig. 3 Time course of Na+-dependent and Na+-independent [3H]oleate uptake by basolateral rat liver plasma membrane vesicles under cation gradient conditions (Reproduced with permission from reference 18.)

-

h

I

.-C0

en

U

-F 0

'i

E

v

al

U

-0 al

Y

c

mechanism. The stimulation of oleate influx in presence of Na+ was not observed in the presence of K + , Li + or choline +. The observation of a Na+-stimulated uptake mechanism may be explained by operation of a Na+-fatty acid co-transport system similar to the uptake system of glucose and amino acids. Alternatively it is conceivable that during the collision of the Na+-fatty acid complex with the carrier, Na+ is released from the complex and stimulates uptake of fatty acids into cells, without being internalized itself. Next it was analysed whether Na+-stimulated influx also reveals an electrical potential dependency. A transmembrane electrical potential gradient in the presence of an inside directed Na+ gradient was generated by variation of the accompanying anions with different membrane permeabilities [ 181. Anions which diffuse into vesicles more rapidly that Na+ (e.g. SCN-) transiently produce a more negative intravesicular membrane potential than anions which permeate the plasma membrane more slowly than Na+ (e.g. gluconate- ). It was shown that the initial uptake of oleate is accelerated in the presence of more permeable accompanying anions (Fig. 4, left). It suggested that the translocation of fatty acids across the plasma membrane is stimulated by a relatively more negative intravesicular charge. This was confirmed by other experiments, in which the vesicles were preloaded with 100 mM potassium gluconate and exposed to the K + ionophore valinomycin. In the absence of extravesicular K +, valinomycin permits

Fig. 4 Potential dependency of oleate uptake Left: transmembrane charge differences in the presence of an inwardly directed Na+ gradient were created by substitution of various accompanying anions (increasing negative intravesicular membrane potential with SCN- > CI-, NO,> SO:- > gluconate-). Right: creation of a negtive intravesicular charge by a valinomycin-induced K + diffusion potential. (Reproduced with permission from [ 181.)

c

e

Time

Volume 20

Time

Lipid-Binding Proteins

a rapid outward directed diffusion of K + , thereby transiently creating a negative charge within the vesicles. In those valinomycin pretreated vesicles, uptake of fatty acids is also accelerated, supporting the hypothesis that cellular influx is stimulated by a negative intracellular potential (Fig. 4, right) [ 181. For evaluation of whether stimulation of N a + dependent fatty acid influx by a negative intracellular potential represents an electrogenic transport mechanism, experiments in the isolated perfused rat liver were performed by Weisiger and co-workers [ 191. Microelectrodes were used to continuously monitor the electrical potential differences across the plasma membrane while simultaneously monitoring the rate of oleate uptake. In these experiments the liver was perfused in presence and absence of Na+. In the presence of N a + , infusion of oleate resulted in depolarization of the membrane potential [ 191. This indicates influx of positive charge. In contrast, in a Na+-free medium no depolarization was detected. These data confirm that depolarization was caused by a specific N a + dependent process. Therefore, it was suggested that the carrier mediated uptake of fatty acids is driven by an electrogenic transport mechanism in which fatty acids enter the cell as positively charged complexes. The above described studies provide strong evidence for the hypothesis that hepatocellular uptake of fatty acids represents a carrier mediated transport process. The identification of a membrane fatty acid transport protein is of physiologic significance since such a carrier mediated uptake process might represent a site of metabolic and hormonal control of fatty acid metabolism.

3. Stremmel, W.. Kochwa, S. & Berk, P. 1). (1983) Biochem. Biophys. Res. Commun. 1,88-95 4. Stremmel, W. (1988)J. Clin. Invest. 81, 844-852 5. Stremmel,W., Lotz, G., Strohmeyer, G. & Herk. P. D. (1985)J. Clin. Invest. 75, 1068-1078 6. Stremmel, W.. Diede, H. E., Rodilla-Sala, E., Vyska. K., Schrader, M., Fitscher, B. & Passarella, S. (1990) Mol. Cell. Biochem. 98, 191- 199 7. Berk, P. D., Wada, W., Horio, Y., Potter, B. J., Sorrentino, D.. Zhou, S.-L., Isola, I,. M., Stump, D., Kiang, C.-I,. & Thung, S. (1990) Proc. Natl. Acad. Sci. U.S.A. 87,3484-3488 8. Potter, B. J., Stump, D., Schwieterman, W., Sorrentino, D., Jacobs, L. N., Kiang, C. I,., Rand, J. & Berk, P. D. (1987) Biochem. Biophys. Res. Commun. 148,1370-1376 9. Diede, H. E., Rodilla-Sala,E., Gunawan,J., Manns, M. & Stremmel, W. (1992) Biochim. Biophys. Acta 1125, 13-20 10. Stremmel, W. & Theilmann. I,. (1986) Biochim. Riophys. Acta 877, 191- 197 11. Stremmel. W., Strohmeyer, G. & Berk, P. D. (1986) Proc. Natl. Acad. Sci. U.S.A. 83, 3584-3588 12. Stremmel, W. (1988)J. Clin. Invest. 82, 2001-2010 13. Sorrentino, D., Stump, D., Potter, H., Robison, R. B., White, R.. Kiang, C. L. & Berk, P. D. (1988) J. Clin. Invest. 82,928-935 14. Schwietermann, W.. Sorrentino, D.. Potter, R. J., Kiang, C. L., Stump, D. & Rerk, P. D. (1988) Natl. Acad. Sci. U.S.A. 85,359-363 15. Abumrad. N. A.. I’erkins, R. C. & Park, J. H. (1981) J. Biol. Chem. 256,9183-9191 16. Maniscalco, W. M., Stremmel, W. & Heeney, M. (1990) Am. J. Physiol. 259,1,206-2 12 17. Stremmel, W. & Herk. P. D. (1986) Proc. Natl. Acad. Sci. U.S.A. 83.3086-3090 18. Stremmel,W. (1987) Biol. Chem. 262,6284-6289 19. Weisiger, R. A., Fitz, J. G. & Scharschmidt, B. F. (1988)J. Clin. Invest. 83,411-420

1. Stremmel,W. (1989)J. Hepatol. 9,374-382

2. Stremmel, W., Strohmeyer, G., Borchard, F., Kochwa, S. & Herk, P. D. (1985) Proc. Natl. Acad. Sci. U.S.A. 82, 4-8

Received 16 July 1992

I992

817