Luminal Leptin Enhances CD147/MCT-1-mediated Uptake of Butyrate in the Human Intestinal Cell Line Caco2-BBE*

In the intestine, butyrate constitutes the major energy fuel for colonocytes. However, little is known about the transport of butyrate and its regulation in the intestine. In this study we demonstrate that the monocarboxylate transporter (MCT-1) is apically polarized in model human intestinal epithelia and is involved in butyrate uptake by Caco2-BBE cell monolayers. The butyrate uptake by Caco2-BBE cell monolayers displayed conventional Michaelis-Menten kinetics and was found to be pH-dependent, Na+-independent, 4,4′-diisothiocyanatostilbene-2,2′-disulfonic acid-insensitive, and inhibited by the monocarboxylate transporter inhibitor α-cyano-4-hydroxycinnamate and by an excess of unlabeled butyrate. We show that MCT-1 associates with CD147 at the apical plasma membrane in Caco2-BBE cell monolayers. Using antisense CD147, we demonstrate that the association of CD147 with MCT-1 is critical for the butyrate transport activity. Interestingly, we show for the first time hormonal regulation of CD147/MCT-1 mediated butyrate uptake. Specifically, luminal leptin significantly up-regulates MCT-1-mediated butyrate uptake by increasing its maximal velocity (V max) without any modification in the apparent Michaelis-Menten constant (K m ). Finally, we show that luminal leptin up-regulates butyrate uptake in Caco2-BBE monolayers by two distinct actions: (i) increase of the intracellular pool of MCT-1 protein without affecting CD147 expression and (ii) translocation of CD147/MCT-1 to the apical plasma membrane of Caco2-BBE cell monolayers.

Bacterial fermentation is high in the proximal large bowel, as is the production of short chain fatty acids (SCFA) 1 that constitute the major end product of the microbial digestion of carbohydrates and dietary fibers (1). At least 60% of SCFA uptake occurs by simple diffusion of the unionized form across the cell membrane; the remainder occurs by active cellular uptake of ionized SCFA involving an acid microclimate on the surface of the intestinal epithelium. SCFA are metabolized rapidly by colonocytes and are the major respiratory fuels in the intestine; indeed, oxidation of SCFA supplies 60 -70% of the energy needs in isolated colonocytes (2). Of the three major SCFA (acetate, propionate, and butyrate), butyrate is the major intestinal fuel even when competing substrates such as glucose and glutamine are available (3)(4)(5). Apart from its function as the dominant energy source for the colonocytes, butyrate also affects cellular proliferation, differentiation, and apoptosis (6 -10).
Recently, it has been suggested that the proton-linked monocarboxylate transporter 1 (MCT-1) may play a major role in the uptake of butyrate by the intestinal epithelial cells in vivo (11,12) as well as in vitro in Caco2 cells (13,14). MCT-1 belongs to the monocarboxylate transporter family including nine MCTrelated sequences that have been so far identified in mammals, each having a different tissue distribution (15). Hydropathy plots predict the number of transmembrane domains to be 12 for MCT-1 with the N and C termini located within the cytoplasm (16). MCT-1 can transport a wide range of short chain monocarboxylates, the K m values (5-10 mM) decreasing as the chain length increases from two to four carbon atoms. Monocarboxylates with longer branched aliphatic or aromatic side chains also bind to the transporter, but are not released following translocation and may act as potent inhibitors. One of these is the classical inhibitor, ␣-cyano-3-hydroxycinnamate (CHC) (17).  has been shown to interact specially with CD147, a member of the immunoglobulin superfamily. This interaction appears to assist MCT expression at the cell surface in heart cells and transfected cells; thus, CD147 acts as a chaperone to increase MCT-1 translocation from the endoplasmic reticulum to the Golgi and plasma membrane (18 -20). Although the regulation of MCT-1 expression has been extensively studied in skeletal muscle (21)(22)(23)(24)(25), little is known about the regulation of MCT-1 or its association with CD147 in the intestine.
Leptin, the ob gene cloned in 1994 by Zhang et al. (26), is a hormone mainly secreted by adipocytes and is involved in central regulation of body weight homeostasis (27)(28)(29) via its specific receptors in the hypothalamus (30). Subsequent studies have established that nonadipose tissues, such skeletal muscle (31), pituitary gland (32), and stomach (33) also produce luminal leptin in the nanomolar range as concentration (33,34). Moreover under secretin, pentagastrin, or vagal stimulation, the gastric luminal leptin output increased by ϳ50 times (34,35). In addition, it has been recently demonstrated that some of the stomach-derived leptin secreted in the gastric juice is not fully degraded by proteolysis, suggesting that it reaches the intestine in an active form, and thus can initiate biological processes involved in controlling functions of the intestinal tract, such as absorption and secretion (34 colon is in the low nanomolar range. We suggest that this leptin is coming from the gastric gland because no leptin staining was detected from the epithelial cells along normal small and large intestine. 2 Interestingly, under inflammatory states, we have detected a strong leptin staining from colonic epithelial cells and the luminal leptin concentration increased significantly (ϳ10 times greater compared with noninflamed tissues). 2 During inflammation the luminal colonic leptin concentration is likely to be the addition of the leptin produced by the gastric gland and the leptin produced by the colonic epithelial cells. These results suggest that luminal leptin could have an important physiological and/or pathological role in the colon. Indeed, the different leptin receptor isoforms including the functional long isoform (Ob-Rb) have been detected in the rat intestine from duodenum to colon and in the model intestinal cell line Caco2 (36 -40). The demonstration of leptin receptor in intestinal tract has initiated several investigations on the possible role of leptin in the digestive physiology as absorption and secretion. Evidence has been provided that leptin can regulate intestinal triglyceride transport by inhibiting apolipoprotein AIV expression via activation of a jejunal leptin receptor in mice (38). Similarly, in rat, intravenous leptin infusion attenuates the increase in synthesis and secretion of apoAIV induced by intraduodenal infusion of lipids (39). In addition, leptin administered to the basolateral side of Caco2 cells inhibits the triglyceride secretion, the biosynthesis of apoB-100 and apoB-48, as well as the output of chylomicron and low density lipoproteins (40). More recently, we have reported that luminal 2 S. V. Sitaraman and D. Merlin, unpublished observations. FIG. 1. A, concentration dependence of butyrate uptake by Caco2-BBE cell monolayers. Caco2-BBE monolayers were incubated at 37°C for 1 h with varying concentrations of [ 14 C]butyrate (specific activity, 16 mCi/mmol) added to the apical reservoir. The pH of the medium was 6.4 (apical reservoir) and 7.5 (basolateral reservoir). Thereafter, medium was aspirated, cells were rapidly washed twice with ice-cold HBSS-HEPES medium, and the amount of butyrate accumulated by Caco2-BBE cell monolayers was determined by liquid scintillation counting. Values represent means Ϯ S.E. of three experiments in triplicate. Inset, Eadie-Hofstee plots (V versus V/S; V is the rate of uptake (pmol/cm 2 /h and S is the substrate concentration (mM)) transformation of the data for carrier-mediated uptake. B, substrate specificity of the butyrate transporter expressed in Caco2-BBE cell monolayers. Caco2-BBE cell monolayers were incubated for 1 h at 37°C with 20 M [ 14 C]butyrate (specific activity, 16 mCi/mmol) alone (CTRL) or with 1 mM CHC, 50 mM cold butyrate (Butyrate 50 mM), 100 M DIDS (DIDS) added to the apical reservoir or butyrate uptake was performed in an uptake solution without Na ϩ (Choline chloride). Thereafter, medium was aspirated, cells were rapidly washed twice with ice-cold HBSS-HEPES medium, and the amount of butyrate accumulated by Caco2-BBE cell monolayer radioactivity was determined by liquid scintillation counting. Each bar represents means Ϯ S.E. of four experiments in triplicate. **, p Ͻ 0.001; ***, p Ͻ 0.001 versus control. C, the pH dependence of butyrate accumulation by Caco2-BBE cell monolayers. Caco2-BBE monolayers were incubated for 5 min at 37°C with 20 M [ 14 C]butyrate (specific activity, 16 mCi/mmol) added to the apical reservoir at various pH values (pH 7.5, 6.5, and 5.5). Thereafter, medium was aspirated, cells were rapidly washed twice with ice-cold HBSS-HEPES medium, and the amount of butyrate accumulated by Caco2-BBE cell monolayer radioactivity was determined by liquid scintillation counting. The pH of the unlabeled incubation medium (basolateral reservoir) was maintained at 7.4. Each bar represents mean Ϯ S.E. of three experiments in triplicate. *, p Ͻ 0.005; **, p Ͻ 0.01 versus control; ##, p Ͻ 0.01 versus pH 6.5.
leptin improves the transport of oligopeptides across the intestinal epithelium through the H ϩ -dependent, di-and tripeptide transporter PepT-1 in vitro and in vivo (36). Together, the results clearly demonstrate that leptin is a key hormone of the intestinal tract. This study aims to investigate the regulation of butyrate uptake by luminal leptin using the model human intestinal epithelial cell line Caco2-BBE.

MATERIALS AND METHODS
Cell Culture and Treatments-Caco2-BBE (41) cells (between passage 30 and 50) were grown in high glucose Dulbecco's Vogt modified Eagle's medium (DMEM, Invitrogen) supplemented with 14 mmol/liter NaHCO 3 and 10% newborn calf serum. Cells were kept at 37°C in 5% CO 2 and 90% humidity, and medium was changed every day. Monolayers were subcultured every 7 days by trypsinization with 0.1% trypsin and 0.9 mmol/liter EDTA in Ca 2ϩ /Mg 2ϩ -free phosphate-buffered saline (PBS). Uptake, confocal immunofluorescence were performed with confluent monolayers plated on collagen-coated permeable supports (area, 1 cm 2 ; pore size, 0.4 m; Transwell-Clear polyester membranes from Costar) and examined 15 days after plating. For protein, membrane or RNA extractions, cells were plated on six-well cluster trays at a density of 10 4 cells/cm 2 and examined 15 days after plating. For treatments, leptin was added to the luminal compartment for 1-24 h in medium without serum.
Transport Experiments-Cells were washed twice with Hanks' balanced salt solution (Sigma Aldrich) complemented with 4 mM NaHCO 3 (HBSS) and 10 mM HEPES, pH 7.5, and stabilized 30 min in the same buffer. Caco2-BBE cells were then incubated in HBSS plus 10 mM MES, pH 6.4, containing 20 M [ 14 C]butyrate (specific activity, 16 mCi/mmol; Sigma) in apical compartment for 1 h at 37°C. We choose 1-h butyrate uptake to visualize the steady state butyrate uptake. However to in-vestigate the pH-dependent activity of MCT-1, 5-min incubation was used to avoid the equilibration of H ϩ concentration across the monolayer. The supernatant was then removed, and cells were washed twice with ice-cold HBSS-HEPES, pH 7.5. Cell-associated radioactivity was determined by liquid scintillation counting in a ␤-counter.
Membrane Extractions-Caco2-BBE pellets was resuspended and homogenized in HEPES (5 mM) containing protease inhibitors. The pellet was then incubated for 30 min at 4°C and centrifuged at 13,000 ϫ g at 4°C for 30 min. The resulting pellet was suspended in PBS by repeated passage through an 18-gauge needle. The protein solution was then boiled 5 min at 100°C in Laemmli buffer supplemented with 0.5% ␤-mercaptoethanol.
Cross-linking and Immunoprecipitation-Cross-linking was carried out using the bifunctional stilbene disulfonate (DIDS; Sigma), as described previously (20). Cells seeded on six-well plates were washed twice with PBS and incubated with 100 M DIDS during 1 h at 37°C. The cells were then washed in ice-cold PBS and lysed with the lysis buffer (1% Triton X-100 in 20 mM Tris, pH 5.0, 50 mM NaCl, 5 mM EDTA, 0.2% bovine serum albumin (BSA), and protease inhibitors). The resulting supernatants were immunoprecipitated with the appropriate amount of specific antibody (0.05 g/ml mouse anti-human CD147 (Cymbus Biotechnology Ltd, Chandlers Ford, UK); 0.05 g/ml of the polyclonal rabbit anti-human MCT-1 (Alpha Diagnostic, San Antonio, TX) was added and gently rocked overnight at 4°C. Subsequently, 50 l of protein G suspension was added to the mixture and incubated overnight at 4°C. The complexes were collected by centrifugation at 12,000 ϫ g for 1 min by microcentrifuge. The beads were washed one time with buffer 1 (1% Triton X-100 in 20 mM Tris, pH 5.0, 50 mM NaCl, 5 mM EDTA, and 2% BSA) and two times with buffer 2 (20 mM Tris-HCl, pH 8.0). The protein solution was then boiled 5 min at 100°C in Laemmli buffer and subjected to SDS-PAGE and transferred at 4°C to nitrocellulose membranes. The blots were blocked 1 h with 5% nonfat  4). Immunoprecipitates or crude cell lysates were subject to 4 -20% SDS-polyacrylamide gel electrophoresis, followed by transfer to nitrocellulose membrane. The blot was immunostained with anti-MCT-1 (lanes 1-4) or with anti-CD147 (lanes 5-8). Ab, antibody. B, importance of CD147 in the butyrate uptake by Caco2-BBE monolayers. Butyrate uptake in Caco2-BBE monolayers transfected with antisense CD147 construct. Subconfluent Caco2-BBE monolayers on filters were transfected with CD147 construct (Antisense CD147), vector alone (Vector) using the Lipofectin technique or nontransfected Caco2-BBE monolayers (CTRL). Butyrate uptake was determined 2 days after transfection. Caco2-BBE monolayers were incubated at 37°C for 5 min with varying concentrations of [ 14 C]butyrate (specific activity, 16 mCi/mmol) added to the apical reservoir. The pH of the medium was 6.4 (apical reservoir) and 7.5 (basolateral reservoir). Thereafter, medium was aspirated, cells were rapidly washed twice with of ice-cold HBSS-HEPES medium, and amount of [ 14 C] butyrate accumulated by Caco2-BBE monolayers was determined by liquid scintillation counting. Values represent means Ϯ S.E. of three experiments performed in triplicate. *, p Ͻ 0.05 versus control; ##, p Ͻ 0.01 versus vector. dry milk in blocking buffer. After washing with blocking buffer, the blots were incubated for 1 h at room temperature with 1:2000 dilution of a mouse anti-human CD147 or rabbit anti-human MCT-1. They were further incubated for 30 min at room temperature with anti-mouse or anti-rabbit horseradish peroxidase-conjugated antibody diluted 1:1000 and probed using chemiluminescence system (ECL, Amersham Biosciences).
Plasmid Constructs and Transfections-The expression of CD147 in Caco2-BBE cells was determined using a reverse transcription-polymerase chain reaction (PCR) method with oligonucleotide primers specific for CD147. Poly(A) T RNA was isolated from Caco2-BBE cells with a Micro Fast Track 2.0 kit (Invitrogen). The yield of RNA was determined by ultraviolet spectrophotometry. One microgram of Poly(A) T RNA was primed with oligo(dT) and reverse transcribed with avian myeloblastosis virus-reverse transcriptase (cDNA Cycle Kit; Invitrogen). A dilution of the reverse transcription reaction was used as a template for amplification by PCR. PCR conditions are determined according the primer characteristics. CD147 coding sequence was amplified using the following primers from Invitrogen: 5Ј-GGAATAGGAATCATGGCG-3Ј and 5Ј-CCACCTGCCTCAGGAAGA-3Ј. The product was visualized and puri-fied from a 1% agarose gel using a DNA extraction kit (Qiagen, Valencia, CA) in the antisense orientation into the mammalian expression vector, pTarget (Promega). The constructed plasmids were verified by DNA sequencing. Plasmids were purified using the Qiagen Maxiplasmid kit. Subconfluent Caco2-BBE cells were plated on permeable filters 48 h prior to transfection with vector alone or antisense CD147 into vector using Lipofectin (Invitrogen) in serum-free medium for 48 h, and uptake experiments were performed. Uptake experiments were performed as described above with the exception that 20 M [ 14 C]butyrate was added to the apical plasma membrane of Caco2-BBE for only 5 min.
Western Blot Analysis-Proteins were separated by SDS-polyacrylamide gel electrophoresis on a 4 -20% gradient gel (Bio-Rad) and then transferred to nitrocellulose membranes. The blots were blocked for 1 h with 5% nonfat milk in blocking buffer, and then the blots were incubated 1 h with 0.05 g/ml CD147 human monoclonal antibody (Cymbus Biotechnology) or with 0.05 g/ml MCT-1 antibody (Alpha Diagnostic). After washing three times for 15 min in blocking buffer, they were further incubated for 1 h with the corresponding secondary horseradish peroxidase-conjugated (anti-rabbit or anti-mouse) antibody diluted The pH of the medium was 6.4 (apical reservoir) and 7.5 (basolateral reservoir). Amount of butyrate accumulated by monolayer was then measured. Thereafter, medium was aspirated, cells were rapidly washed twice with 3 ml of ice-cold incubation medium, and radioactivity was determined by liquid scintillation counting. Each bar represents mean Ϯ S.E. of three experiments in triplicate. **, p Ͻ 0.01 versus control; ###, p Ͻ 0.001 versus leptin alone.
1:2000. The nitrocellulose was washed three times for 20 min in blocking buffer and then probed using a chemiluminescence system (ECL, Amersham Biosciences).
Confocal Immunofluorescence-Caco2-BBE cells grown on filters were washed twice in HBSS, pH 7.4, and then fixed with 3.7% paraformaldehyde in Hanks' balanced salt solution with calcium, pH 7.4 (HBSS ϩ ). Caco2-BBE cells were then permeabilized with 0.5% Triton for 30 min at 25°C. The cells were rinsed and incubated with rhodamine-phalloidin (Molecular Probes Inc., Eugene, OR) diluted 1:60 for 40 min. Caco2-BBE cells were then blocked 1 h in a blocking solution containing 0.2% gelatin and 0.08% saponin in HBSS ϩ . These monolayers were incubated 1 h with 0.05 g/ml mouse anti-human CD147 human antibody (Cymbus Biotechnology) or with 0.05 g/ml rabbit anti-human MCT-1. These monolayers were then stained with appropriate fluorescein isothiocyanate (anti-goat or anti-mouse) antibody diluted at 1:1000. Microscopy was performed using a Zeiss epifluorescence microscope equipped with a Bio-Rad MRC600 confocal unit, computer, and laser scanning microscope image analysis software (Carl Zeiss, Jena, Germany).
Northern Blot Analysis-MCT-1 cDNA was obtained from ATCC MGC-1187 (ATCC, Manassas, VA) and used as probe for Northern blot analysis. Total RNA was isolated from Caco2-BBE cells with Tri-Reagent (Molecular Research Center, Cincinnati, OH). Total RNA (20 g) was denatured by heating at 65°C in 20 mM HEPES pH 7.2, 1 mM EDTA, 50% formamide, and 6% formaldehyde for 15 min and subjected to electrophoresis on a 1% agarose gel containing 2% formaldehyde. Resolved RNA was transferred to a nylon membrane (PerkinElmer Life Sciences) and covalently cross-linked by exposure to UV light. Hybridization was performed in a solution that contains 7% SDS, 1% BSA, 10% polyethylene glycol 8000, 250 mM NaCl, 1.25 mM EDTA, 125 mM NaPO 4 and 1 mg of salmon sperm DNA (Ambion, Austin, TX). MCT-1 cDNA probe was labeled with [␣-32 P]CTP using the Rediprime II random prime labeling system (Amersham Biosciences). A mouse glyceraldehyde-3-phosphate dehydrogenase probe was used as control (Ambion, Austin, TX).
Statistics-Results are expressed as means Ϯ S.E. Statistical significance was determined using the paired t test.

MCT Mediates Butyrate Uptake in Caco2-BBE Monolayers-
The existence of a specific transporter system mediating butyrate flux across the apical membranes of Caco2-BBE cell monolayers was investigated. The butyrate transport across apical plasma membranes of Caco2-BBE monolayers was saturable, as was evident from uptake measurements done at varying concentrations of butyrate in the range of 0.5-10 mM (Fig. 1A). Kinetic parameters were calculated from the Michaelis-Menten equation. The apparent K m and V max values for uptake across the apical membrane of Caco2-BBE were 2.6 Ϯ 0.2 mM and 8.25 Ϯ 1.75 mol/cm 2 /h, respectively. The Eadie-Hofstee transformation of the data for the carrier-mediated uptake yielded a linear plot (r ϭ 0.82) (Fig. 1A, inset). The substituted aromatic monocarboxylates such as CHC, known as one of the now classical inhibitors of MCT, decreased the butyrate uptake by 40% in Caco2-BBE monolayer (without CHC: 210 Ϯ 18 pmol/ cm 2 /h; with CHC: 126 Ϯ 18 pmol/cm 2 /h) (Fig. 1B). The uptake of 20 M radiolabeled [ 14 C]butyrate in Caco2-BBE monolayers was reduced by 95% in the presence of 50 mM unlabeled butyrate (402 Ϯ 32 versus 12 Ϯ 1 pmol/cm 2 /h), indicating that the butyrate uptake was almost completely carrier-mediated (Fig.  1B). However, these results suggest that other transporter(s) in addition to MCT are involved in the butyrate transport. When Na ϩ was replaced by choline in the uptake solution, butyrate uptake by Caco2-BBE monolayers was not affected (with NaCl: 717 Ϯ 31 pmol/cm 2 /h; with choline chloride: 716 Ϯ 86 pmol/ cm 2 /h), demonstrating that the butyrate uptake was Na ϩ -independent (Fig. 1B). In addition, butyrate uptake was not affected in the presence of 100 M DIDS (without DIDS: 583 Ϯ 35 pmol/cm 2 /h; with DIDS: 529 Ϯ 27 pmol/cm 2 /h), excluding the involvement of an anion-exchange mechanism in the butyrate uptake by Caco2-BBE monolayers (Fig. 1B). Furthermore, uptake of butyrate by Caco2-BBE monolayers increased with H ϩ concentration (pH 7.5: 121 Ϯ 10; pH 6.5: 169 Ϯ 23; pH 5.6: 257 Ϯ 34 pmol/cm 2 /h) (Fig. 1C). These observations are consistent with those reported by others, showing that the transfer of butyrate across apical plasma membranes of Caco2-BBE involves cotransport of butyrate and a proton. Because MCT-1 is the most abundant MCT isoform expressed in Caco2 cells (13) and because our kinetic parameters are similar to those reported in a previous study using Caco2 cells (13), MCT-1 seems to be one of the transporter involved in the butyrate uptake by Caco2-BBE. Kirk et al. (18) demonstrated that CD147 is associated to MCT-1 in the plasma membrane, and this association was shown to be critical for the MCT-1 activity in heart cell line and in transfected cells. We next investigated whether there was an interaction between CD147 and MCT-1 in Caco2-BBE cells by using DIDS to crosslink the two proteins. After DIDS treatment of cells for 1 h, Western blot of the MCT-1 immunoprecipitated with mouse anti-CD147 antibody revealed the presence of 100-kDa immunoreactive band corresponding to the expected size of MCT-1/ CD147 complex ( Fig. 2A, lanes 7 and 8). Similarly, Western blotting of CD147 immunoprecipitated with a rabbit anti-MCT-1 revealed an identical 100-kDa immune complex ( Fig.  2A, lanes 3 and 4). When cell lysates were mot immunoprecipitated (lanes 1, 2, 5, and 6), the presence of ϳ40 -45 kDa (lanes 1 and 2) and ϳ50 -55 kDa (lanes 5 and 6) corresponding to MCT-1 and CD147, respectively, were detected. These results demonstrate that CD147 and MCT-1 are associated to the Caco2-BBE cell membranes. To study the functional role of CD147 in the butyrate uptake, Caco2-BBE monolayers were transfected with vector alone, antisense CD147 cDNA (fulllength) inserted into pTarget/CMC vector. The uptake of butyrate was determined 48 h after transfection. As shown in Fig.  2B, the antisense CD147 inhibited butyrate uptake by ϳ25% when compared with the vector alone or nontransfected cells. This inhibition is significant, given that the transfection efficiency in Caco2-BBE is not 100%. These results confirm the functional role of CD147 in the butyrate uptake by Caco2-BBE cell monolayers.  (Fig. 3A). The effect of luminal leptin on butyrate uptake is specific to leptin, because an unrelated peptide, interleukin-8, did not induce any change in butyrate uptake in Caco2-BBE monolayers (data not shown). In addition, the effect of luminal leptin on butyrate uptake was concentrationdependent and reached a plateau at 10 nM (without luminal leptin: 231 Ϯ 87; 10 nM luminal leptin: 367 Ϯ 21; 100 nM luminal leptin: 370 Ϯ 11 pmol/cm 2 /h) (Fig. 3B). Moreover, the MCT inhibitor, 1 mM CHC, reversed luminal leptin induced increase in butyrate uptake (507 Ϯ 22 versus 203 Ϯ 18 pmol/ cm 2 /h) (Fig. 3C). Furthermore, the addition of an excess of butyrate (50 mM) to the apical compartment completely suppressed (98%) the leptin-induced increase in [ 14 C]butyrate uptake (507 Ϯ 22 versus 10 Ϯ 1 pmol/cm 2 /h) (Fig. 3C). In addition, the presence of 100 M DIDS or the absence of Na ϩ in the uptake solution did not affect the butyrate uptake induced by luminal leptin in Caco2-BBE monolayers.

Luminal Leptin Increases Butyrate Uptake in Caco2-BBE
To determine whether leptin may affect monocarboxylate uptake by modifying the intrinsic activity of the transporter, the effect of apical leptin on the kinetics of butyrate uptake was studied. Kinetic analysis of the data (Fig. 4) indicates that apical leptin treatment significantly increased the V max (V max(apical leptin) ϭ 14.4 Ϯ 1.28 mol/cm 2 /h versus V max(control) ϭ 8.25 Ϯ 1.75 mol/cm 2 /h), but did not modify the K m of the transporter (K m(apical leptin) ϭ 3 mM Ϯ 0.21 versus K m(control) ϭ 2.65 Ϯ 0.2 mM). These data demonstrate that leptin does not change the affinity of MCT-1 localized on the apical membrane of the Caco2-BBE cell monolayers.
Luminal Leptin Increases MCT-1 Expression but Does Not Affect the Expression of CD147-Western blot analysis was performed using mouse anti-MCT-1 or rabbit anti-CD147 antibody on total protein lysate from Caco2-BBE monolayers treated with leptin for various periods of time. The total amount of MCT-1 immunoreactive protein (ϳ 40 -45 kDa), was not significant at 12 h, but was significant at 24 h after leptin treatment (Fig. 5A). In contrast, the total amount of CD147 immunoreactive protein (ϳ50 -55 kDa) was not increased after luminal leptin (Fig. 5A).
Luminal Leptin Increases MCT-1 mRNA Expression-mRNA levels of MCT-1 were analyzed by Northern blot in Caco2-BBE cells. Total RNA was obtained from untreated monolayers or monolayers treated with 10 nM luminal leptin for 4 h. Fig. 5B showed that MCT-1 mRNA (3.3 kb) levels were significantly increased (2-fold) in cells treated by leptin compared with controls. This increase was transient, as the mRNA levels returned to basal levels at 8 h (data not shown).
Luminal Leptin Induces the Translocation of MCT-1/CD147 to the Apical Plasma Membranes of Caco2-BBE Cell Monolayers-To visualize the outline of the monolayers, the actin was labeled with rhodamine phalloidin as described under "Materials and Methods." The membrane localization of MCT-1 and CD147 was assessed in confluent Caco2-BBE cell monolayers. We examined the effect of luminal leptin treatment (10 nM for 24 h) on the membrane expression of MCT-1 and CD147 by using confocal immunofluorescence microscopy. The staining for the immunoreactive CD147 protein in control cells was localized mainly in the tight junction associated with a staining below the apical membrane (Fig. 6). After luminal leptin stimulation (10 nM for 24 h), the major localization of CD147 was steadily below the apical membrane, suggesting a translocation of CD147 to the apical plasma membrane of the cells (Fig. 6). At the basal level, MCT-1 protein was primarily localized in the apical membrane and in the cytoplasm just below the apical plasma membrane and no staining in the basolateral membranes (Fig. 6). The degree of expression of MCT-1 varied between cells, as judged by both the vertical xy image and the horizontal xz image (Fig. 6). After apical leptin treatment, the level of immunostaining for MCT-1 increased with a strong signal both in the apical membrane and in the cytoplasm compartment below the membrane (Fig. 6). Moreover, the number of positive cells for MCT-1 was increased, consistent with an To confirm that CD147 and MCT-1 are translocated to the plasma membrane after luminal leptin treatment, Caco2-BBE cell membranes were isolated from Caco2-BBE cells treated with 10 nM leptin for 24 h or without treatment. Western blot analysis was performed using these membrane cell lysates and probed with anti MCT-1 (A) and CD147 (B) antibodies. As is apparent from Fig. 7, luminal leptin treatment for 24 h increased the amount of MCT-1 (a similar result was found after luminal leptin treatment for 12 h) and CD147 proteins in the plasma membrane of Caco2-BBE cell monolayers. Together, these results demonstrate a translocation of both CD147 and MCT-1 to the apical plasma membranes of Caco2-BBE monolayers after luminal leptin treatment. DISCUSSION In this study, we demonstrate that MCT-1 is apically polarized in model human intestinal epithelia and is involved in the butyrate uptake by Caco2-BBE cell monolayers. The butyrate uptake by Caco2-BBE cell monolayers is pH-dependent, Na ϩindependent, DIDS-insensitive, and inhibited by the MCT inhibitor CHC and by an excess of unlabeled butyrate. Together, these results indicate that apical butyrate uptake by Caco2-BBE cell monolayers are mainly the result of a single carrier MCT-1 and are in agreement with previous studies (13). Caco2-  BBE at 15 days after plating exhibited partially small intestinal phenotype. However, we found the same butyrate transport characteristics in HT29-Cl.19A (data not shown), which exhibited colonocyte features. Thus, with respect to butyrate uptake studies, Caco2-BBE cells represent an appropriate cellular model.
In the present study, we bring important information to the field of MCT by providing the first evidence for its stimulation by a hormone. Indeed, luminal leptin significantly increased the maximal velocity (V max ) for butyrate uptake, whereas the apparent Michaelis-Menten constant (K m ) did not change. Moreover, the addition of an excess of butyrate or the use of the MCT inhibitor, CHC, suppressed the luminal leptin-induced increase in butyrate uptake. These results demonstrate that the increased of butyrate uptake induced by luminal leptin is mediated by the same transporter (MCT-1) involved in baseline conditions.
We show that MCT-1 and CD147 are localized to the apical plasma membrane in Caco2-BBE monolayers. Furthermore, we demonstrate that the CD147/MCT-1 association is critical for the butyrate transport activity. The inhibition of butyrate uptake with an antisense construct to CD147 transiently transfected into Caco2-BBE cell monolayers supports the requirement of CD147 in butyrate transport. CD147 protein is probably specifically interacting with MCT-1, because previous studies using MCT-1 antisense show similar inhibition (ϳ35%) of butyrate uptake by Caco2 monolayers (13). These data are consistent with similar findings in murine heart plasma membrane and in cells co-transfected with CD147 and MCT-1 in which CD147 was reported to facilitate proper expression of MCT-1 at the cell surface, where they remained tightly associated (18 -20). This interaction has potential significance for the regulation of MCT-1 activity. Proteins acting as a protein chaperone for other transporter have been described. For example, glycophorin associates with the anion exchanger AE1 (42,43), CD98 that associates with amino acid transporters (44,45), or CD36 that associates with the long chain fatty acid transporter (46). All these associations have been shown to be essential for the appropriate function of these transporters.
Interestingly, we show that luminal leptin increases the total amount of MCT-1 proteins, attributable to a relative increase in expression of MCT-1 mRNA, but did not change the total mRNA or protein level of CD147. Moreover, luminal leptin treatment induced an increase in the number of positive cell for MCT-1 in accordance with the enhancement of protein synthesis. Furthermore we demonstrate, using confocal imaging and by isolating membrane from Caco2-BBE cells, that after luminal treatment CD147 and MCT-1 membrane proteins increase, indicating that CD147 and MCT-1 are translocated to the apical membrane. Based on our observations that the association of MCT-1 and CD147 is crucial for the MCT-1 transport function, the leptin-induced translocation of CD147 to the apical membrane could play a role in the increase of butyrate uptake by leptin.
Little is known about the regulation of MCT-1 expression in various tissues, but, in skeletal muscles, endurance and high intensity training have been shown to increase the expression of MCT-1 and transport activity (25,47). However, to date there is no published promoter analysis for any of the MCT isoforms, and we cannot speculate whether luminal leptin effect involved direct or indirect regulation of MCT-1 gene promoter activity.
The increase in butyrate uptake in the enterocyte by luminal leptin represents an important issue because butyrate has been shown to present physiological and therapeutic interest. The presence of luminal leptin in the intestinal lumen (35) could contribute to maintain MCT-1/CD147 membrane expression and thereby regulate intestinal inflammation via butyrate-dependent system in response to changes in the intestinal milieu. The ability of butyrate to induce cancer cell apoptosis may contribute to the cancer preventive activity of SFCA. In the diseased colon, expression of MCT-1 or CD147 protein may be impaired; this in turn would reduce the availability of SCFA required to maintain the intracellular events regulating normal differentiation and proliferation in the colonic mucosa.
In summary we demonstrate that (i) luminal leptin up-regulates butyrate uptake mediated by MCT-1, (ii) MCT-1 associates with CD147 at the apical membrane and the expression of CD147 is crucial for MCT-1 transport activity, (iii) luminal leptin increases the intracellular pool of MCT-1 protein, and (iv) luminal leptin induces a translocation of CD147/MCT-1 to the apical plasma membrane.