INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

THE RETINAL PIGMENT epithelium (RPE) forms the outer blood-retinal barrier and mediates the transport of metabolites, ions, and fluid between the choroidal blood supply and the neural retina (4, 5). The basolateral surface of the RPE is in contact with the blood plasma, which filters through the porous capillaries in the choroid. The apical surface of the RPE is in intimate contact with the neural retina and extends processes that protrude between the photoreceptor cell outer segments. Tight junctions at their apical-lateral borders link the RPE cells. These junctions impede the movement between cells of even small water-soluble molecules.

There are no anatomic junctions between the RPE and the neural retina. Although the two tissues are closely apposed, they are separated by the subretinal space (SRS). This space is the embryonic remnant of the optic vesicle and forms when the optic vesicle invaginates to form the optic cup. The RPE actively regulates the volume and chemical composition of the SRS in much the same way that the choroid plexus maintains the composition of the cerebrospinal fluid (5). This is accomplished through specific proteins in the apical and basolateral membranes (4, 5).

The maintenance of visual cell function depends on glycolysis (1-3, 19, 24). The neural retina, with its high rate of metabolism, uses glucose and produces substantial quantities of lactate, both in the light and in the dark (3, 19, 24). In the retina, lactate is not simply an end product of glycolysis but is an important metabolic intermediate. Glucose is transported across the RPE by GLUT-1 transporters that are found in both the apical and basolateral membranes (14, 21). Glucose is used primarily by Müller cells that have few mitochondria and rely on glycolysis for ATP production. Lactate produced in the Müller cells by glycolysis is transported out of the cells and is used by the photoreceptor cells to fuel oxidative phosphorylation (19). In studies with isolated photoreceptor cells, lactate was a better substrate for mitochondrial oxidative metabolism than glucose (19). The production of lactate by one tissue and use by another is reported in muscle (11) and brain (23) and is referred to as the "lactate shuttle."

Excess lactate not used by the photoreceptor cells is transported out of the subretinal space to the choroidal circulation by the RPE (1, 2, 22). Physiological studies on RPE explants and cultured RPE cells have documented the presence of H⁺-lactate cotransport mechanisms (13, 15, 16). Recently a specific H⁺-lactate transporter protein has been identified in RPE. MCT3 was originally cloned from an embryonic chick library screened with an RPE-specific monoclonal antibody and is the third member of the monocarboxylate transporter family to be cloned (17, 25).

Studies presented in this paper clearly establish that RPE expresses two members of the monocarboxylate transporter family, MCT1 and MCT3. We demonstrate that these two transporters are polarized to distinct membrane domains in RPE cells: MCT1 in the apical membrane and MCT3 in the basolateral membrane. We further correlate the presence of these transporters with the recently described proton-lactate-water symporter activity in the apical membrane of RPE (26).

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Animals. Male Sprague-Dawley rats weighing 125-150 g were purchased from Charles River Laboratories (Wilmington, MA) and housed with a 12:12-h light-dark cycle for 1 wk. Rats were euthanized during the light cycle using pentobarbital sodium (150 mg/kg body wt) before tissue collection. All procedures followed a protocol approved by the Pennsylvania College of Optometry Animal Care Committee.

Microdissection of eyes. Enucleated eyes were separated into anterior and posterior segments with a razor blade. The posterior eye cup was placed in PBS containing 10 mM EDTA and 3% sucrose, the neural retina was removed, and the RPE was peeled off the choroid with fine forceps under a dissecting microscope (18). RPE and neural retina were used to prepare total RNA and membrane extracts described below.

Preparation of probes. The MCT1 coding sequence was RT-PCR amplified from total rat RPE RNA using 5'aag atg cca tcc tgc gat tgg3' as a forward primer and 5'aga cag ggc tct cct cct ct3' as a reverse primer. The PCR product (1.4 kb in length) was cloned into pCR2.1 vector (Invitrogen, Carlsbad, CA), and an EcoR I fragment containing MCT1 cDNA was recloned in pBluescript SK() plasmid (Stratagene, La Jolla, CA). pBSSK()MCT1 was linearized with EcoR V enzyme, and digoxigenin-UTP-labeled riboprobe was prepared using T3 RNA polymerase following the manufacturer's instruction (Boehringer Mannheim, Indianapolis, IN).

The MCT3 probe was synthesized using a 600-bp fragment of the 3' end of mouse MCT3 (GenBank Accession Number AF019111) that was cloned into the pCR2.1 vector (Invitrogen). The pCR2.1 MCT3 was linearized with BamH I, and digoxigenin-UTP-labeled riboprope was prepared using T7 RNA polymerase. The 3' end of the rat MCT3 sequence was obtained by 3' rapid amplification of cDNA ends (RACE) and varied only in one nucleotide from the mouse sequence.

Preparation of total RNA and Northern blot analysis. Total RNA was prepared from various rat tissues using triZOL Reagent (Life Technologies, Gaithersburg, MD) following the manufacturer's instructions. RNA (5 µg) was denatured with 0.5 M glyoxal and 50% dimethylsulfoxide and separated on a 1% agarose gel in 10 mM sodium-phosphate buffer (5 mM Na₂HPO₄, 5 mM NaH₂PO₄, pH 6.5). RNA was transferred and crosslinked to Hybond N⁺ nylon membrane (Amersham). The membrane was prehybridized 4 h and hybridized overnight at 65°C with MCT1 or MCT3 riboprobe (5 ng/ml) in hybridization solution. Blots were washed twice for 5 min each in 2× SSC (1× SSC is 0.15 M NaCl and 0.015 M sodium citrate, pH 7.0)/0.1% SDS at room temperature, twice for 20 min each in 0.5× SSC/0.1% SDS at 65°C, and rinsed in maleate buffer (0.1 M maleic acid, pH 7.5, 150 mM NaCl). After 1 h in blocking buffer (Boehringer Mannheim), blots were incubated 1 h with an alkaline phosphatase conjugated anti-digoxigenin antibody (1:5,000 in blocking buffer). Unbound antibody was removed by washing twice for 20 min each in maleate buffer and 5 min in GB3 (0.1 M Tris, pH 9.5, 0.1 M NaCl, 50 mM MgCl₂). Lumi-Phos 530 was applied, and the signal was detected using BIOMAX MR X-ray film (Kodak, Rochester, NY).

MCT antibodies. A polyclonal antibody was produced against a 21-mer synthetic peptide. The peptide corresponds to the carboxy terminal amino acids of rat MCT3 deduced from the cloned cDNA and was synthesized with an amino-terminal cysteine conjugated to keyhole limpet hemocyanin (KLH) (CAVPELDHESIGGHEARGQKA). The KLH-peptide (0.1 mg/injection) was emulsified by mixing with an equal volume of Freund's adjuvant for injection into New Zealand White rabbits. Boosts were at weeks 2, 6, and 8, and animals were bled at 0, 4, and 8 wk. The 8-wk bleed was used in these studies.

The production of the rat MCT1 polyclonal antibody described by Gerhart and co-workers (8) was purchased from Chemicon (Temecula, CA).

Western analysis of protein levels. Detergent-soluble lysates were prepared from rat RPE and retina as previously described (17). Protein was measured using bicinchoninic acid reagent (Sigma), and samples were diluted with two times Laemmli sample buffer. Samples, adjusted for equal protein (15 µg/lane), were separated on 4-12% SDS-polyacrylamide gradient gels (Novex, San Diego, CA) and transferred to Immobilon-P membrane (Millipore, Bedford, MA). Membranes were incubated for 1 h at room temperature in blocking buffer (20 mM Tris, 137 mM NaCl, pH 7.5, 5% BSA) followed by 1 h incubation with primary antibodies, polyclonal MCT1 antibody diluted 1:5,000 (Chemicon), or polyclonal MCT3 antibody diluted 1:500. The secondary antibody diluted 1:5,000 was rabbit anti-chicken IgY or goat anti-rabbit IgG conjugated to horseradish peroxidase. Chemiluminescence (Amersham) was used for detection. To demonstrate the specificity of the MCT3 antibody, blots were also probed with MCT3 antibody that was preabsorbed for 30 min with 1 µg/ml MCT3 peptide.

Immunocytochemistry. Paraffin sections (8 µm) of adult rat eye and brain were purchased from Novagen (Madison, WI). Sections were dewaxed in xylene and rehydrated using a graded series of ethanol. Sections were labeled with antibodies following standard protocols as previously described (17, 18). Sections were incubated 1 h in blocking solution (5% BSA-0.1% Tween 20 in PBS), then incubated for 1 h in anti-MCT1 antibody diluted 1:200 or anti-MCT3 diluted 1:50 in 1% BSA-0.1% Tween 20 in PBS. After four washes in PBS-0.1% Tween 20, slices were incubated 1 h with rhodamine conjugated rabbit anti-chicken IgY (diluted 1:100) or goat anti-rabbit IgG (diluted 1:100). Sections were washed, and coverslips were mounted using Fluormount-G (Southern Biotechnology, Birmingham, AL). Sections were examined, and images were captured using a Nikon microscope equipped with Metamorph Imaging System (Universal Imaging, West Chester, PA).

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Expression of MCT1 and MCT3 mRNA in rat RPE. To identify MCTs expressed by the RPE, total RNA was extracted from microdissected rat RPE and neural retina. Equal amounts of RNA (5 µg) were subjected to Northern analysis and hybridized using probes specific for MCT1 and MCT3 as detailed in MATERIALS AND METHODS. The MCT1 probe hybridized with a single 3.7-kb transcript in both the RPE and neural retina (Fig. 1). The amount of MCT1 transcript in RPE and neural retina preparations was similar. MCT1 was also detected in RNA from rat heart, liver, and skeletal muscle (data not shown), confirming published reports (6, 7).

View larger version (44K): [in this window] [in a new window]   Fig. 1.   Differential expression of MCT1 and MCT3 mRNA in the retinal pigment epithelium (RPE) and neural retina (NR). Total RNA (5 µg) was prepared from rat RPE and neural retina using triZOL reagent, separated on a 1% agarose gel, and transferred to a nylon membrane. Blot was hybridized with digoxigenin-UTP-labeled MCT1 (left) or MCT3 (right) riboprobes. Detection was with alkaline phosphatase conjugated anti-digoxigenin antibody and chemiluminescence.

A parallel blot was hybridized with an MCT3 riboprobe that was prepared as detailed in MATERIALS AND METHODS. A 2.2-kb transcript was detected only in total RNA prepared from the RPE but not from the neural retina (Fig. 1). The MCT3 probe did not hybridize with RNA prepared from rat heart, liver, kidney, or thyroid (not shown). The limited expression of MCT3 in RPE and not in other tissues was reported previously in the chick (17).

Differential expression of MCT1 and MCT3 proteins in the RPE and neural retina. Detergent-soluble extracts were prepared from microdissected rat RPE and neural retina as detailed in MATERIALS AND METHODS. Protein (15 µg) was separated using a 4-12% gradient SDS-polyacrylamide gel and transferred to PVDF membrane. The blot was incubated with affinity-purified polyclonal antibody to MCT1, and labeling was detected by chemiluminescence as shown in Fig. 2. A 45-kDa immunoreactive protein was found in both RPE and neural retina, but the amount in RPE exceeded that found in neural retina. The amount of RNA does not correlate with the amount of expressed protein (Fig. 1 compared with Fig. 2). Unlike RPE that is composed of one cell type, the neural retina lysate is made up of several cell types.

View larger version (10K): [in this window] [in a new window]   Fig. 2.   Expression of MCT1 in RPE and neural retina as determined by immunoblotting. Detergent extracts were prepared from microdissected RPE and neural retina as detailed in MATERIALS AND METHODS. Protein (15 µg/lane) was separated on 4-12% gradient gels and transferred to Immobilon-P membranes. Blot was probed with MCT1 antibody (Ab) and detected with a horseradish peroxidase conjugated secondary antibody with an enhanced chemiluminescence system.

A polyclonal antibody was made to the COOH-terminus of mouse MCT3. This region is not conserved between different members of the MCT family, so the antibody is specific for MCT3. As shown in Fig. 3, a single band of ~43-kDa was detected in RPE. When the immunizing peptide was present at a concentration of 1 µg/ml, antibody binding was inhibited (Fig. 3). MCT3 immunoreactivity was not detected in lysates from neural retina (Fig. 3) nor in tissue extracts of rat liver, heart, and skeletal muscle (not shown).

View larger version (46K): [in this window] [in a new window]   Fig. 3.   Expression of MCT3 in RPE and neural retina as determined by immunoblotting. Detergent extracts were prepared from microdissected RPE and neural retina as detailed in MATERIALS AND METHODS. Protein (15 µg/lane) was separated on 4-12% gradient gels and transferred to Immobilon-P membranes. Blots were probed with MCT3 antibody alone (left) or MCT3 antibody with 1 µg/ml synthetic peptide (right). Horseradish peroxidase conjugated secondary antibody and enhanced chemiluminescence were used for detection.

Immunolocalization of MCT1 to the apical membrane of rat RPE. Indirect immunofluorescence was used to determine the subcellular localization of MCT1 in rat RPE. Paraffin sections through the posterior region of rat eye were prepared from formaldehyde-fixed tissue as detailed in MATERIALS AND METHODS and in Ref. 17. Sections were labeled with a polyclonal antibody directed against the COOH-terminus peptide of MCT1 and a rhodamine conjugated secondary antibody. The phase contrast (green) and fluorescent (red) images were captured and superimposed to clarify localization. The immunostaining (orange) of the apical membrane, with its extensive processes, is clearly seen in Fig. 4. The large amount of MCT1 detected in RPE on Western blot (Fig. 2) is consistent with the intense labeling of the apical membrane.

View larger version (99K): [in this window] [in a new window]   Fig. 4.   Immunolocalization of MCT1 and MCT3 in rat RPE. Paraffin sections of rat eye were prepared from formaldehyde-fixed tissue as described in MATERIALS AND METHODS. Sections were incubated with an MCT1 antipeptide antibody followed by a rhodamine conjugated rabbit anti-chicken secondary antibody or an MCT3 anti-peptide antibody followed by a rhodamine conjugated goat anti-rabbit antibody. Phase image (green) and immunofluorescence (red) were superimposed using the Metamorph imaging system. OS: outer segments.

Immunolocalization of MCT3 to the basal membrane of rat RPE. Indirect immunofluorescence was used to localize MCT3 in rat tissues. Sagittal sections through adult rat eye showed RPE was the only tissue that stained with MCT3 antibody. When viewed at low magnification, labeling was only observed in the RPE, but not in other ocular tissue or in the extraocular muscles. Higher magnification revealed that the labeling was restricted to the basolateral membrane of the RPE (Fig. 4). Additionally, MCT3 immunostaining was not detected in sagittal sections through the brain that contained choroid plexus, a tissue with several barrier functions in common with RPE.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

In vivo and in vitro studies have shown that both lactate and water are transported from the retina to the choroid by the RPE (2, 13, 15, 16, 22). Transepithelial movement of lactate and water requires transport proteins on the apical and basolateral membranes of the RPE. Monocarboxylate transporters through the cotransport of proton, lactate, and H₂O could regulate these transport activities of the RPE (26). The data presented in these studies show that the RPE cells express two monocarboxylate transporters, MCT1 and MCT3.

The monocarboxylate transporter MCT1 is found throughout the apical membrane of the RPE (Figs. 2 and 4). The apical membrane and the tight junctions of the RPE form the outer boundary of the subretinal space. The transport proteins in the apical membrane regulate the composition of the "extracellular" fluid of the subretinal space (5). One of the constituents of the subretinal space is lactate, which is both a substrate and product of retinal metabolism. Lactate concentrations in the subretinal space are estimated to be between 7 and 13 mM (1, 10, 15) and may fluctuate during visual activity (19). The cytosolic concentration of lactate is unknown, but blood plasma is usually ~1 mM. A chemical gradient for lactate would therefore exist from the subretinal space to the choroidal blood plasma. Monocarboxylate transporters mediate the electroneutral symport of a proton and lactate. The finding of MCT1 on the apical membrane (Fig. 4) and MCT3 on the basolateral membrane (Fig. 4) identifies two proteins, acting in concert, that can mediate the transepithelial transfer of H⁺-lactate from the subretinal space to the systemic blood system.

MCT1 also may have an important role in water homeostasis. Studies by Zeuthen and co-workers (26) elegantly demonstrate that water is cotransported with H⁺-lactate across the apical membrane of frog RPE. Whether MCT1 is responsible for this symport of H⁺-lactate and water has not been shown, but MCT1 is abundant and correctly located in the apical membrane. The Michaelis constant (K_m) of MCT1 in rat RPE is unknown, but the reported K_m of MCT1 in other tissues is 2-11 mM (11) and in agreement with the K_m (~4-7 mM) reported for H⁺-lactate-water symport across the apical membrane in frog RPE (26).

For water to continue to enter across the apical membrane with the downhill influx of lactate and against an osmotic gradient, the cytosolic concentration of lactate in the RPE cannot exceed that in the subretinal space. MCT3 on the basolateral membrane provides a mechanism for lactate efflux and thereby maintains intracellular lactate levels below 7 mM (26). Whether proton-lactate transport by MCT3 is also accompanied by water has not been demonstrated, but aquaporins (9, 20) and ion transport proteins, for example the HCO₃/ Cl exchanger (12), may contribute to the transepithelial movement of water.

In addition to water homeostasis, MCT1 and MCT3 in RPE may have a role in maintaining intracellular pH. Intracellular pH of RPE is estimated to be 7.38, a value more alkaline than most mammalian cells and well above the calculated value for equilibrium pH. As proposed by Kenyon et al. (12), H⁺-lactate transport is only one of many overlapping mechanisms available to RPE for maintaining pH.

Similar to the role of the choroid plexus in regulating the chemical composition of the fluid available to brain cells, the RPE plays a critical role in maintaining and regulating the environment required for retinal function. The demonstration of MCT1 on the apical membrane and MCT3 in the basal membrane, identifies specific proteins involved in the discriminate and critical regulation of water and lactate transport from the retina to the choroid.