Uptake of long chain fatty acids is regulated by dynamic interaction of FAT/CD36 with cholesterol/sphingolipid enriched microdomains (lipid rafts)
- 8.4k Downloads
Mechanisms of long chain fatty acid uptake across the plasma membrane are important targets in treatment of many human diseases like obesity or hepatic steatosis. Long chain fatty acid translocation is achieved by a concert of co-existing mechanisms. These lipids can passively diffuse, but certain membrane proteins can also accelerate the transport. However, we now can provide further evidence that not only proteins but also lipid microdomains play an important part in the regulation of the facilitated uptake process.
Dynamic association of FAT/CD36 a candidate fatty acid transporter with lipid rafts was analysed by isolation of detergent resistant membranes (DRMs) and by clustering of lipid rafts with antibodies on living cells. Lipid raft integrity was modulated by cholesterol depletion using methyl-β-cyclodextrin and sphingolipid depletion using myriocin and sphingomyelinase. Functional analyses were performed using an [3H]-oleate uptake assay.
Overexpression of FAT/CD36 and FATP4 increased long chain fatty acid uptake. The uptake of long chain fatty acids was cholesterol and sphingolipid dependent. Floating experiments showed that there are two pools of FAT/CD36, one found in DRMs and another outside of these domains. FAT/CD36 co-localized with the lipid raft marker PLAP in antibody-clustered domains at the plasma membrane and segregated away from the non-raft marker GFP-TMD. Antibody cross-linking increased DRM association of FAT/CD36 and accelerated the overall fatty acid uptake in a cholesterol dependent manner. Another candidate transporter, FATP4, was neither present in DRMs nor co-localized with FAT/CD36 at the plasma membrane.
Our observations suggest the existence of two pools of FAT/CD36 within cellular membranes. As increased raft association of FAT/CD36 leads to an increased fatty acid uptake, dynamic association of FAT/CD36 with lipid rafts might regulate the process. There is no direct interaction of FATP4 with lipid rafts or raft associated FAT/CD36. Thus, lipid rafts have to be considered as targets for the treatment of lipid disorders.
KeywordsLipid Raft Fatty Acid Uptake Cholesterol Depletion Placental Alkaline Phosphatase Plasma Membrane Cholesterol
long chain fatty acids
detergent resistant membrane.
Uptake of long chain fatty acids (LCFAs) is important for many cellular functions and the understanding of the uptake mechanisms is an important target for treatment of lipid disorders [1, 2, 3, 4]. The molecular mechanisms of fatty acid transport across the plasma membrane are still a matter of debate and the predominating mechanism likely differs from cell to cell (for reviews see [5, 6, 7, 8]). In general, two possible groups of mechanisms are discussed: simple diffusion and saturable transport processes. Whereas the uptake based on the passive diffusion process depends on the activity of intracellular metabolism creating a transmembrane downhill concentration gradient; the saturable process is regulated by expression of certain proteins and lipids at the plasma membrane. Much effort has been spent to identify candidate proteins that are directly involved in the facilitated fatty acid uptake mechanism. So far four candidates (FABPpm, FAT/CD36, FATP family proteins, caveolin-1) have been discussed [9, 10, 11]. Whereas FAT/CD36 and FABPpm are membrane associated fatty acid binding proteins that are thought to mediate their dissociation from albumin and accumulation at the outer leaflet of the plasma membrane , followed by flip-flop across the phospholipid bilayer to the cytosolic site; it has been suggested that FATPs are real transporters directly involved in the uptake process of LCFAs across the membrane bilayer . However, recent data from our laboratory suggest that this group of proteins are rather enzymes that indirectly facilitate the translocation process by encompassing acyl-coA activity .
The composition of membrane lipids also modulates fatty acid uptake. In particular, there is increasing evidence that cholesterol is of crucial importance. Cholesterol depletion in 3T3 adipocytes, HMEC, HEK293 or HepG2 cells decreased LCFA uptake and this effect was reversible after re-addition of cholesterol [14, 15, 16, 17]. Caveolin-1 has been suggested to regulate the cholesterol content of the plasma membrane [8, 18] and LCFA uptake. Caveolin-1 can bind LCFAs . Caveolin-1 knock out mice showed a reduced mass of adipocytes and increased serum free fatty acids, indicating that LCFA uptake into adipocytes might be impaired . LCFA uptake is increased by caveolin-1 overexpression and inhibited in caveolin-1 knockout mouse fibroblasts  or by expression of the dominant negative caveolin-1 mutant CAVDGV . Interestingly, the inhibitory effects of CAVDGV can be reversed by replenishing the cell membranes with cholesterol and can be mimicked by methyl-β cyclodextrin treatment .
All these studies point out that cholesterol is critically involved in LCFA uptake. However, little is known about the mechanisms by which cholesterol modulates this process. We have previously hypothesized that the association of FAT/CD36 with lipid rafts might determine this process . Rafts are lateral assemblies of sphingolipids and cholesterol within cellular membranes involved in compartmentalization of membrane processes . Biochemically, lipid raft constituents are characterized by their insolubility in low concentration of detergents such as Triton X-100 . In this regard, it has been demonstrated that a fraction of FAT/CD36 associates with detergent-resistant membranes (DRMs) in a cholesterol-dependent manner . The reduction of overall LCFAs uptake by cholesterol depletion was as effective as the specific inhibition of the FAT/CD36 function by sulfo-N-succinimidyl oleate (SSO). Simultaneous treatments had no additional effect, suggesting that both procedures target the same cellular compartment .
The most straightforward interpretation of these findings is that there are two pools of FAT/CD36 at the plasma membrane, one associated with lipid rafts, in which LCFAs are transported, and another outside of rafts, where no transport occurs. That means, FAT/CD36 has to reside in lipid rafts to facilitate FA uptake. Cholesterol depletion would shift FAT/CD36 into the surrounding lipid bilayer and make this protein non-functional. If FAT/CD36 facilitated uptake of fatty acids requires that the receptor is associated with lipid rafts, then regulation of this association may represent a mechanism by which cellular uptake of fatty acids can be regulated. In this paper we provide evidence demonstrating that association of FAT/CD36 with lipid rafts is critical for the uptake process to occur.
Reagents and antibodies
Methyl-β-cyclodextrin, sphingomyelinase and myriocin were from Sigma. Antibodies used were mouse anti-placenta alkaline phosphatase (PLAP) (Dako Cytomation), rabbit anti-PLAP ; rabbit anti-FATP4 (C4 anti FATP4 described in ); mouse anti-FLAG (Sigma), Cy5 and Cy3 donkey anti-mouse/rabbit (Jackson Immuno Research); mouse monoclonal antibody anti GFP 3E6 (Invitrogen); rabbit anti GFP KG77 ; mouse anti-FAT/CD36 (Biosource).
GFP-FATP4 have been described previously . CD36-FLAG was kindly provided by Douglas M. Lubin, Washington University School of Medicine, St. Louis, MO. GFP-TMD is a membrane anchored version of the green fluorescent protein. The signal sequence of human CD8 (MALPVTALLLPLALLLHAARP) is followed by an epitope tag (VSV-G; YTDIEMNRLGK). Next is EGFP, followed by a 16 amino acid glycosylation tag from human rhodopsin (NGTEGPNFYVPFSNAT) and the transmembrane domain of podocalyxin (EFEDRFSMPLIITIVCMASFLLLVAALYGCCHRK). This plasmid is identical to the construct GFP-tail described in  except that the cytoplasmic tail has been removed and behaves like a non-raft protein .
Cumulative uptake of oleate was based on Stremmel and Berk . Adherent cells were incubated for 5 min at 37°C with [3H]-oleate solution 170 μM (0.68 μCi/ml [3H]-oleate (Amersham)) BSA fatty acid free (Sigma) in PBS; After stopping and washing with ice-cold 0.5% BSA in PBS, cells were lysed with 1 M NaOH and aliquots analyzed for protein concentration (Biorad) and radioactivity by scintillation counting as before .
Cells, transient transfection, cholesterol/sphingolipid depletion
COSJ (ATCC CRL-1651) and Vero (CCL-81) cells were maintained under standard tissue culture conditions with the appropriate culture media (COSJ: D-MEM Invitrogen 4,5 g glucose/L 10% FBS 2 mM L-Glutamine and Vero: D-MEM Invitrogen 1 g glucose/L 5% FBS 2 mM L-Glutamine). Cells grown to near confluency (10 cm2) were transfected with 4 μg plasmid-DNA and 10 μl lipofectamine 2000 (Invitrogen). Analysis was performed 16–20 h hours after transfection. For cholesterol depletion the cells were treated for 30 min with 10 mM methyl-β-cyclodextrin (MβCD) in DMEM. Cholesterol determinations were done using the Amplex Red Cholesterol Assay kit (Molecular Probes). For inhibition of sphingolipid synthesis the cells were seeded at a density of 0.2 × 106 in 10 cm2 dishes and were grown for 3 d in complete medium in the presence of 5 μM myriocin. During this period the medium was changed once. Growing COS cells in the presence of myriocin for a longer time caused the cells to detach and undergo apoptosis. The extent of sphingolipid-depletion was estimated microscopically on cells incubated for 30 min with 25 μg/ml of rhodamine-conjugated cholera toxin subunit B (Rh-CTB) (Molecular Probes) at 4°C, which was followed by a wash and subsequent fixation in 4% paraformaldehyde in PBS (PFA). The expression of FATP4 and FAT/CD36 for the different conditions were in initial experiments by Western blotting and found not to differ significantly (data not shown).
Immunofluorescence and antibody-induced patching
For immunofluorescence microscopy, cells were fixed for 4 min with 4% PFA at 8°C followed by an incubation in methanol at -20°C for 4 min. Fixed cells were incubated for 1 h at room temperature with the appropriate dilution of antibodies in PBS/0.2% gelatine (see below). After three washes with PBS/0.2% gelatine they were incubated with the respective secondary antibodies in PBS/0.2% gelatine for 1 h at room temperature.
To aggregate raft proteins the respective antibodies were diluted in CO2 independent medium (GIBCO) containing 2 mg/ml BSA. The polyclonal antibodies against PLAP were diluted 1:35; the monoclonal anti-GFP (3E6) and anti-FAT/CD36 1:50. The cells were incubated for 45 min with the respective combination of antibodies at 10°C, briefly washed and further incubated for 45 min at 10°C with mixed fluorescently labelled secondary antibodies. Cy3-labelled secondary antibodies were diluted 1:500, and the Cy5-labelled ones 1:100. The cells were fixed as described above. Fluorescent images were acquired on an Olympus microscope and arranged with Adobe Photoshop.
Preparation of detergent resistant membranes (DRMs)
Detergent extraction with Triton X-100 was performed as described before for N2a cells . Cells were grown in 3.5 cm dishes, transfected and 10–12 h later washed once with PBS and scraped on ice into 1.5 ml homogenisation buffer (250 mM Sucrose, 10 mM Hepes, 2 mM EDTA) and after centrifugation (5 min 2000 rpm) cell pellets were homogenized in homogenisation buffer containing 20 μg/ml each of chymostatin, leupeptin, antipain and pepstatin A (Sigma) through a 26 G needle and centrifuged for 5 min at 3000 rpm. The postnuclear supernatant was subjected to extraction for 30 min at 4°C in 1% Triton X-100. The extracts were adjusted to 40% OptiPrep (Axis-Shield) and overlaid in a TLS 55 centrifugation tube with 30% OptiPrep/TNE, and TNE (25 mM Tris-HCl, pH 7.4, 150 mM NaCl, 2 mM EDTA old protocol/25 mM Tris-HCl, pH 10.8, 150 mM NaCl, 5 mM EDTA new protocol since 03.2007). The gradients were centrifuged at 400000 g in a Beckman SW41 rotor for 20 h at 4°C. Fractions were obtained and used for Western blotting as described  and protein levels quantified by densitometry.
All values are reported as mean and standard error of the mean (SEM). The Kruskal-Wallis test was used to test for statistical significance. Probability values of p < 0.05 were set as threshold for statistical significance.
Cholesterol depletion inhibits long chain fatty acid uptake in COS cells
Inhibition of sphingolipid synthesis decreases LCFA uptake
Rafts are cholesterol and sphingolipid-enriched microdomains. To find out whether sphingolipids are also important for LCFA uptake, we tested the effect of inhibition of sphingolipid biosynthesis.
FAT/CD36 co-patches with placental alkaline phosphatase and segregates from non-raft marker GFP-TMD
Antibody induced cross-linking increases the association of FAT/CD36 to detergent resistant membranes
Various proteins associate to lipid rafts with different kinetics and partition coefficients . It has been shown that antibody-induced patching may stabilize association of raft proteins with detergent-resistant membranes (DRMs) [22, 31]. We therefore analysed the association of FAT/CD36 to DRMs under cross-linking with antibodies.
Cross-linking of FAT/CD36 with antibodies increases oleate uptake
The data presented here strengthen the evidence that FAT/CD36 partition into lipid rafts and that LCFA uptake mediated by FAT/CD36 depends on this association.
First of all, we could show that LCFA uptake was critically dependent on the integrity of lipid rafts. Lipid rafts are cholesterol and sphingolipid enriched microdomains. Removal of raft lipids from cells leads to disruption of raft functions . Consistent with previous results on other cells [14, 15, 16, 17] we could demonstrate that by decreasing cellular levels of cholesterol, LCFA uptake in COS cells was also inhibited by > 50%. Additionally, we could show that by decreasing ceramide synthesis with myriocin or depleting membrane sphingolipids using external sphingomyelinase we were able to lower LCFA uptake. Sphingolipid depletion was less dramatic, probably because it is more difficult to lower cellular sphingolipid levels than cholesterol levels. The effects of sphingolipid synthesis inhibition may also be more pleiotropic because there are many different kinds of sphingolipids (sphingomyelins and glycosphingolipids). Nevertheless, we found that plasma membrane GM1 levels were reduced after myriocin treatment, demonstrating that not only cholesterol but also sphingolipids modulate LCFA uptake.
Another important finding supporting raft association was that after cross-linking with antibodies FAT/CD36 co-patched with placental alkaline phosphatase (PLAP), a GPI-anchored raft protein, in living cells and segregated from patches formed by cross-linking of GFP-TMD that served as a non-raft marker. This assay has previously been used to monitor if proteins associate with lipid rafts at the cell [25, 39]. Antibody cross-linking of surface proteins leads to the formation of plasma membrane clusters that can be easily observed during light microscopy. It has been shown that co-patching is dependent on plasma membrane cholesterol, as patching is inhibited by cholesterol removal .
Antibody cross-linking increased DRM association of FAT/CD36 as was previously demonstrated for other raft proteins [25, 31]. Because a considerable amount of FAT/CD36 was also found in the soluble fraction after detergent treatment, FAT/CD36 is probably found (in steady state) in two membrane pools, one raft associated and another localized in the surrounding bilayer. How partitioning between these two pools is regulated is an open question, but interestingly it has been shown that FAT/CD36 can dimerize . Oligomerization of raft components is known to lead to increased raft affinity [37, 38]. Therefore this mechanism might be important to regulate raft association of FAT/CD36.
It has been previously hypothesized that the detergent soluble fraction might represent intracellular FAT/CD36 and that upon cholesterol depletion FAT/CD36 might be stuck in the secretory pathway to the plasma membrane . It has been described, that FAT/CD36 can indeed translocate between the plasma membrane and intracellular compartments . However, recent data from Covey et al  using immunofluorescence and flow cytometry convincingly show that in 3T3-L1 adipocytes cholesterol depletion inhibits uptake of LCFAs without affecting FAT/CD36 or caveolin-1 distribution within the cells, indicating that cholesterol levels regulate LFCA uptake via a pathway that does not involve altered surface localization of FAT/CD36. Our cross-linking data, which were obtained on living cells at 10°C, can explain these findings because under these conditions intracellular antigens are not accessible to antibodies. Therefore increased DRM association of FAT/CD36 (Figure 5) after cross-linking with antibodies likely represent a horizontal recruitment of plasma membrane non-raft associated FAT/CD36 into raft domains. The enrichment of DRM associated FAT/CD36 could only be detected in presence of co-expressed FATP4. Whether the FATP4 acitivity as acyl-CoA synthetase drives cellular fatty acid influx and, thus, the requirement for raft associated FAT/CD36 or whether FATP4 is actively involved in the raft constitution process remains to be elucidated. FATP4 has been originally described to by a major fatty acid transporter at the apical membrane of enterocytes  and it has been speculated that both proteins might cooperate at the plasma membrane in fatty acid uptake [5, 41]. However recent data could not support this view . FATP4 seems rather to be an enzyme for acyl-CoA activity at the ER level and therefore indirectly increases fatty acid uptake. Deletion of FATP4 resulted in a perinatal lethality with a phenotype reminiscent of lethal restrictive dermopathy . By histology the skin was characterized by a hyperproliferative hyperkeratosis with a disturbed epidermal barrier. Lipid analysis of the skin revealed an increased proportion of ceramides and cholesterol. Within the ceramide fraction very long chain fatty acid substitutes were significantly reduced . These finding suggests that FATP4 is involved in ceramide biosynthesis, possibly by providing CoA activity for very long chain fatty acids. It is intriguing to speculate that this protein might therefore have also impact on lipid raft integrity. Rafts are enriched in sphingolipids (that are made from ceramides) and their long chain fatty acid substitutes have been implicated to be important to connect the outer and the inner layer of the microdomain. FATP4 would therefore be indirectly involved in the fatty acid uptake process by providing the molecules for a functional lipid raft assembly. This could explain our results of an enhanced raft association of FAT/CD36 after overexpression of FATP4.
In general clustering of lipid rafts can be achieved by different mechanisms . Besides antibodies also other ligands (e.g. insulin), lectins or linker proteins have been discussed. Which would be a reasonable mechanism for FAT/CD36 clustering? It is conceivable that under physiological conditions the clustering of FAT/CD36 is mediated by binding of long chain fatty acids. It was indeed shown that the artificial, long chain fatty acid [3H]-sulfo-N-succinimidyl oleate (SSO) – when incubated with adipocytes – binds to FAT/CD36 with high affinity and is almost exclusively found in DRMs . Thus the concentration of fatty acids presented to the plasma membrane may regulate the rate of fatty acid uptake by providing a sufficient functional platform in the form of FAT/CD36-raft complexes.
Our data support a crucial role for lipid rafts in LCFA uptake. Compartmentalization of FAT/CD36 at the cell surface by lipid rafts seems to be important in regulating its involvement in LCFA uptake. Taken all data together it seems possible that fatty acid binding increases raft affinity of FAT/CD36. The protein is shifted from the surrounding bilayer to these microdomains were LCFA uptake is likely to occur.
Work was supported by Stiftung Nephrologie and the Dietmar-Hopp-Stiftung
- 16.Pohl J, Ring A, Korkmaz U, Ehehalt R, Stremmel W: FAT/CD36 Mediated Long-Chain Fatty Acid Uptake in Adipocytes Requires Plasma Membrane Rafts. Mol Biol Cell. 2004Google Scholar
- 19.Razani B, Combs TP, Wang XB, Frank PG, Park DS, Russell RG, Li M, Tang B, Jelicks LA, Scherer PE, Lisanti MP: Caveolin-1-deficient mice are lean, resistant to diet-induced obesity, and show hypertriglyceridemia with adipocyte abnormalities. J Biol Chem. 2002, 277: 8635-8647. 10.1074/jbc.M110970200.CrossRefPubMedGoogle Scholar
- 40.Luiken JJ, Koonen DP, Willems J, Zorzano A, Becker C, Fischer Y, Tandon NN, Van Der Vusse GJ, Bonen A, Glatz JF: Insulin stimulates long-chain fatty acid utilization by rat cardiac myocytes through cellular redistribution of FAT/CD36. Diabetes. 2002, 51: 3113-3119. 10.2337/diabetes.51.10.3113.CrossRefPubMedGoogle Scholar
- 42.Herrmann T, van der Hoeven F, Grone HJ, Stewart AF, Langbein L, Kaiser I, Liebisch G, Gosch I, Buchkremer F, Drobnik W, Schmitz G, Stremmel W: Mice with targeted disruption of the fatty acid transport protein 4 (Fatp 4, Slc27a4) gene show features of lethal restrictive dermopathy. J Cell Biol. 2003, 161: 1105-1115. 10.1083/jcb.200207080.PubMedCentralCrossRefPubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.