FAT/CD36-mediated long-chain fatty acid uptake in adipocytes requires plasma membrane rafts

Abstract

We previously reported that lipid rafts are involved in long-chain fatty acid (LCFA) uptake in 3T3-L1 adipocytes. The present data show that LCFA uptake does not depend on caveolae endocytosis because expression of a dominant negative mutant of dynamin had no effect on uptake of [3H]oleic acid, whereas it effectively prevented endocytosis of cholera toxin. Isolation of detergent-resistant membranes (DRMs) from 3T3-L1 cell homogenates revealed that FAT/CD36 was expressed in both DRMs and detergent-soluble membranes (DSMs), whereas FATP1 and FATP4 were present only in DSMs but not DRMs. Disruption of lipid rafts by cyclodextrin and specific inhibition of FAT/CD36 by sulfo-N-succinimidyl oleate (SSO) significantly decreased uptake of [3H]oleic acid, but simultaneous treatment had no additional or synergistic effects, suggesting that both treatments target the same mechanism. Indeed, subcellular fractionation demonstrated that plasma membrane fatty acid translocase (FAT/CD36) is exclusively located in lipid rafts, whereas intracellular FAT/CD36 cofractionated with DSMs. Binding assays confirmed that [3H]SSO predominantly binds to FAT/CD36 within plasma membrane DRMs. In conclusion, our data strongly suggest that FAT/CD36 mediates raft-dependent LCFA uptake. Plasma membrane lipid rafts might control LCFA uptake by regulating surface availability of FAT/CD36.

Figure 1.

Inhibition of cholera toxin uptake by the dominant negative dynamin II mutant K44A. (A) Rhodamine-labeled cholera toxin. (B) GFP-labeled K44A. (C) Merged image showing that cells transiently expressing K44A do not efficiently take up cholera toxin, which remains at the cell surface (40× magnification).

Figure 2.

Isolation of a lipid raft-enriched membrane fraction. CHAPS-insoluble lipid rafts (DRMs) were separated from soluble membranes (DSMs) on sucrose gradients and immunoblotted with antibodies to the transferrin receptor (TfR), caveolin-1, FAT/CD36, FATP1, and FATP4. Representative blots are shown. For quantification of the respective protein a given fraction (indicated on the y-axis) has been set to 100, and the other fractions were expressed relative to this fraction. Values are means ± SD of three independent experiments.

Figure 3.

Effects of cyclodextrin and SSO on [³H]oleate uptake. Pretreatment of cells with β cyclodextrin (10 mM for 30 min at 37°C) or SSO (400 μM for 30 min at 37°C) resulted in a significant reduction in the rate of [³H]oleate uptake over the course of 5 min. However, combining both treatments had no additional effect on [³H]oleate uptake. The asterisks indicate statistical significance (p < 0.05). Values are means ± SD of three independent experiments.

Figure 4.

Binding of [³H]SSO to lipid raft-enriched membrane fractions and FAT/CD36. After incubating live 3T3-L1 adipocytes (A and B) or total cell homogenates (C) with 400 μM [³H]SSO for 30 min, the reaction was stopped by an ice-cold solution containing 0.5% (wt/vol) albumin and 200 μM phloretin. Afterward, CHAPS-insoluble lipid rafts were separated from soluble membranes on sucrose gradients. (A) Binding of [³H]SSO to DRMs of live cells. When radioactivity of DRM and DSM fractions was determined the major peak of [³H]SSO cofractionated with DRMs. (B) Binding of [³H]SSO to FAT/CD36 in DRM and DSM fractions of live cells. One hundred microliters of each CHAPS fraction was separated by 12% SDS-PAGE, and the 88-kDa band representing FAT/CD36 (as confirmed by Western blotting) was excised from the gel and assessed for radioactivity. The figure shows [³H]SSO content associated with the FAT/CD36 band (closed squares) and the rest of the gel (open diamonds). (C) Binding of [³H]SSO to FAT/CD36 in DRM and DSM fractions of cell homogenates. As opposed to live cells, [³H]SSO incubated with total cell homogenates mainly associated with FAT/CD36 bands (closed squares) in DSM fractions. However, there was also modest radioactivity that was not associated with FAT/CD36 (open diamonds). Values in A, B, and C are means ± SD of five independent experiments.

Figure 5.

Membrane fractionation of 3T3-L1 cell homogenates. (A) Expression of marker proteins for plasma membrane (anti-Na,K-ATPase), Golgi network (gp 26/27), and ER (calreticulin). (B) Expression of proteins involved in LCFA uptake in membrane fractions. Fractionation was performed using a discontinuous OptiPrep step gradient, and equal amounts of protein were separated by SDS-PAGE, blotted, and probed with antibodies against the indicated proteins. Signals were quantified by densitometry, and protein content was expressed relative to a given fraction (indicated on the y-axis) that has been set to 100. Values are means ± SD of three independent experiments. (C) Binding of [³H]SSO to membrane fractions. Membrane fractionation was performed after incubation of 3T3-L1 adipocytes for 30 min with 400 μM [³H]SSO. Values are means ± SD of five independent experiments.

Figure 6.

Isolation of a lipid raft-enriched membrane fraction from plasma membrane fractions (A) and nonplasma membrane fractions (B). Representative blots are shown. For quantification of FAT/CD36 a given fraction (indicated on the y-axis) has been set to 100, and the other fractions were expressed relative to this fraction. Values are means ± SD of three independent experiments. Whereas FAT/CD36 at the plasma membrane level was exclusively located within DRMs (A), intracellular FAT/CD36 was expressed predominantly in DSMs with only minor amounts cofractioning with DRMs (B).