Materials and Methods

Chemicals.

Labeled [tyrosyl-3,5-³H]deltorphin II (50–53 Ci/mmol) and [tyrosyl-2,6-³H(N)]DPDPE (46 Ci/mmol) were from DuPont-NEN (Boston, MA). Unlabeled deltorphin II, DPDPE, Tyr-Gly-Gly-Phe-Leu-OH (Leu-enkephalin), and Tyr-d-Ala-Gly-N-methyl-Phe-glycinol (DAMGO) were purchased from Bachem (Bubendorf, Switzerland). Estrone-3-sulfate, naloxone, and naltrindole were from Sigma (St. Louis, MO). All other chemicals were obtained in the highest degree of purification available from Merck (Dietikon, Switzerland), Sigma, or Fluka (Buchs, Switzerland).

Production and Purification of OATP-A Antiserum.

A polyclonal antiserum was raised in rabbits with a fusion protein containing the C-terminal end of OATP-A (last 44 amino acids) and the maltose-binding protein of Escherichia coli as described inKullak-Ublick et al. (1997). Affinity purification of the OATP-A antiserum was performed by dialysis of 16 mg of fusion protein against 10 mM HEPES/Tris, pH 7.0, and its coupling to Affigel 10 (Bio-Rad Laboratories, Hercules, CA) according to the manufacturer's instructions. The affinity column was pre-equilibrated with 10 mM Tris/HCl, pH 8.0; 150 mM NaCl; and 0.02% NaN₃. The OATP-A antiserum was diluted 1:3 in the pre-equilibration buffer and circulated through the column at a flow rate of 12 ml/h for several hours at 4°C. After extensive washing, bound OATP-A antibodies were eluted with 3.5 M MgCl₂. Elution of protein was monitored by measuring the optical density at 280 nm. Positive fractions were pooled, adjusted to a protein concentration of 1 mg/ml with BSA, and dialyzed against 10 mM Tris/HCl, pH 8.0; 150 mM NaCl; and 0.02% NaN₃.

Human Brain Tissue Collection.

Three human frontal cortex specimens were collected during routine autopsies performed by the Department of Neuropathology, University Hospital Zürich, according to a protocol approved by the Ethics Committee of the University Hospital Zürich. The specimens were from patients (male, age 69–75) without evidence of neurological or psychiatric disorders and with normal brain morphology. The post-mortem delay until tissue collection ranged between 8 and 17 h. The blocks of the frontal cortex were dissected upon autopsies and frozen immediately on dry ice.

Western Blotting.

One human frontal cortex sample was homogenized in 0.32 M sucrose and 5 mM Tris/HCl, pH 7.4, to give a protein concentration of ∼2 to 3 mg/ml. The homogenate was incubated for 15 min at 60°C with an equal volume of 125 mM Tris/HCl, pH 6.8; 20% glycerol; 0.002% bromphenol blue; 10% β-mercaptoethanol; and 4% SDS; and subjected to SDS-polyacrylamide gel electrophoresis with 10% mini-gels (Mini Protean II; Bio-Rad Laboratories). Proteins were transferred onto nitrocellulose membranes in a semidry electroblotting apparatus (Trans Blot; Bio-Rad Laboratories) at 15 V for 60 min with 39 mM glycine, 48 mM Tris, and 0.04% SDS as transfer buffer. For immunodetection, the blots were blocked for 2 h in TBST (10 mM Tris/HCl, pH 8; 0.15 M NaCl; and 0.05% Tween 20) containing 5% nonfat dry milk (=blocker) at room temperature, followed by incubation with affinity-purified OATP-A antiserum overnight at 4°C in TBST/5% blocker. The blots were washed one time for 10 min with 20 mM Tris/HCl, pH 7.5; 60 mM NaCl; 2 mM EDTA; 0.4% SDS; 0.4% Triton X-100; and 0.4% deoxycholate; and three times with TBST. Incubation with secondary antibodies (horseradish peroxidase-conjugated goat anti-rabbit IgG diluted 1:5000 in TBST/5% blocker) was carried out for 1 h at room temperature. After extensive washing as described above, immunoreactivity was detected by the chemiluminescence method (Western blot chemiluminescence reagent plus; DuPont-NEN).

Immunofluorescence Staining.

Two brain cortex samples were cut at 10 μm on a cryostat, mounted onto glass slides coated with 3-aminopropyltriethoxysilane (Sigma), and stored at −80°C until use. Just before further processing, sections were fixed with 2% paraformaldehyde in PBS for 20 min and quenched for endogenous peroxidase activity with 1.5% H₂O₂ in PBS for 10 min. Autofluorescence was eliminated by the tyramide signal amplification method (NEN Life Science Products, Brüssels, Belgium; Loup et al., 1998). The sections were incubated overnight at 4°C with the affinity-purified OATP-A antiserum diluted to 0.5 μg/ml in 50 mM Tris/saline, pH 7.4; 0.05% Triton X-100; and 4% normal goat serum. Sections were washed three times for 10 min with Tris/saline and incubated for 1 h at room temperature with a biotinylated goat secondary antibody (Jackson Immunoresearch, West Grove, PA) diluted 1:500 in the same buffer as for the primary antibody. Thereafter, sections were processed for tyramide signal amplification-indirect staining according to the manufacturer's instruction. They were incubated sequentially in streptavidin-horseradish peroxidase diluted 1:1000 in TNB (0.1 M Tris/HCl, pH 7.5; 0.15 M NaCl; and 0.5% blocking agent) for 30 min, in biotinylated tyramide diluted 1:75 in amplification diluent for 10 min, and finally with streptavidin conjugated to Cy3 (Jackson Immunoresearch) diluted to 1:1000 in TNB for 30 min. Then sections were washed with PBS and coverslipped with Immu-mount (Shandon, Pittsburgh, PA). The stained sections were analyzed by confocal laser microscopy (MRC 600; Bio-Rad Laboratories). The specificity of the immunoreactivity against OATP-A was controlled for by preincubating the antiserum with increasing concentrations (5–20 μg/ml) of the fusion protein used for immunization.

Transport Assays in X. laevis Oocytes.

MatureX. laevis females were purchased from the African Xenopus facility at Noordusek R., South Africa, and kept under standard conditions (Hagenbuch et al., 1996). In vitro synthesis of OATP-A-, Oatp1-, and Oapt2-cRNA was performed from the respective cDNAs as described in Kullak-Ublick et al. (1994). Oocytes were prepared and incubated overnight at 18°C. Healthy oocytes were microinjected with either H₂O (controls) or 5 ng of cRNAs encoding OATP-A, Oatp1, or Oatp2. After injection, the oocytes were cultured for 3 days to allow the expression of the carrier proteins in the plasma membrane. Tracer uptake experiments were carried out in Na⁺-containing (100 mM NaCl) and Na⁺-free (100 mM choline chloride instead of NaCl) media containing in addition 2 mM KCl; 1 mM CaCl₂; 1 mM MgCl₂; and 10 mM HEPES/Tris, pH 7.5. Ten oocytes were prewashed in the uptake medium and then incubated at 25°C in 100 μl of uptake medium supplemented with the indicated amounts of radiolabeled and unlabeled peptides (see figure legends). Subsequently, the oocytes were washed with 3 × 6 ml of ice-cold incubation buffer and each oocyte was dissolved in 10% SDS. After addition of 5 ml of scintillation fluid (Ultima Gold; Canberra Packard, Zürich, Switzerland), the oocyte-associated radioactivity was measured in a Packard Tri-Carb 2200 CA liquid scintillation analyzer (Canberra Packard).

Results

To determine the in situ localization of OATP-A in human brain, the specificity of the affinity-purified OATP-A antiserum was first tested by Western blot analysis of a human cortex homogenate. As illustrated in Fig. 1, the antiserum reacted in a concentration-dependent manner with a single band of ∼60 kDa. This immunoreactivity was specific for OATP-A because the immunopositive band disappeared completely after preadsorption of the antiserum with the fusion protein used for immunization (Fig. 1). The mass of ∼60 kDa corresponds to the deglycosylated form of OATP-A as previously demonstrated in in vitro translation experiments (Kullak-Ublick et al., 1995). The same band also was detected after specific expression of OATP-A with a baculovirus vector in Sf9 cells (Fig. 1). However, in the latter system a portion of partly or even fully glycosylated OATP-A with an apparent molecular mass between 80 and 90 kDa (Fig. 1) also was present. Hence, similar to Oatp2 at the rat BBB (Gao et al., 1999), human brain OATP-A appears to be considerably less glycosylated compared with liver OATP-A, which exhibits an apparent molecular mass of 85 to 90 kDa in immunoblots of normal human liver specimens (Kullak-Ublick et al., 1997).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Immunochemical identification of OATP-A. Proteins from homogenate of human frontal cortex (20 μg/lane, lanes 1–6) and of OATP-A-expressing Sf9 cells (lanes 7 and 8) were subjected to SDS-polyacrylamide gel electrophoresis and Western blotting with affinity-purified OATP-A antibody. Lanes 1 and 2 were incubated with the OATP-A antibody at a concentration of 0.66 μg/ml, lanes 3 and 4 at 0.33 μg/ml, lanes 5 and 6 at 0.16 μg/ml, and lanes 7 and 8 at 0.016 μg/ml. Specificity of the OATP-A antibody was assessed by coincubation with 20 μg/ml fusion protein used for immunization (lanes 2, 4, 6, and 8). The position of molecular mass markers is indicated on the left.

Next, we used the specific antiserum to immunolocalize OATP-A in human brain samples obtained from normal frontal cerebral cortex. As illustrated in Fig. 2, strong OATP-A reactivity was observed in brain microvessels and capillaries, whereas astrocytes and neurons were immunonegative (Fig. 2A). The positive immunoreactivity was lost after preadsorption of the antiserum with the fusion protein used for immunization, demonstrating its specificity for OATP-A (Fig. 2B). At higher magnification, predominant immunoreactivity was observed along the border of the brain microvessels (Fig. 2C) and capillary endothelial cells (Fig. 2D). These data demonstrate expression of OATP-A at the BBB of the human cerebral cortex and complement the similar localization of Oatp2 in rat brain (Gao et al., 1999). Unfortunately, possible additional expression of OATP-A at the choroid plexus could not yet be investigated because no human choroid plexus tissue samples could be made available so far.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Localization of OATP-A in human brain microvessels and brain capillary endothelial cells, visualized by immunofluorescence staining. A, human frontal cortex section was incubated with the antibody against OATP-A. Immunoreactivity is present only in the brain microvessels and capillaries. B, no signal was detected in an adjacent section incubated with OATP-A antibody preabsorbed with 20 μg/ml fusion protein used for immunization. C and D, high-power photomicrographs showing intense OATP-A staining for brain microvessel (C) and capillary endothelial cells (D). Scale bar, 30 μm (A and B); 8 μm (C); and 10 μm (D).

Functional transport studies with the δ-opioid agonists DPDPE and deltorphin II were performed in the X. laevis oocyte expression system. As illustrated in Fig.3, oocyte expression of OATP-A induced ∼3- to 4-fold increases of DPDPE and deltorphin II uptakes compared with water-injected oocytes. These OATP-A-mediated uptake activities were independent of sodium (Fig. 3). The same was true for the rat Oatp1 and Oatp2, which stimulated sodium-independent DPDPE transport activities ∼100- and 10-fold, respectively, when expressed inX. laevis oocytes (Fig. 3). However, although Oatp1 stimulated deltorphin II uptake to an approximately similar extent as human OATP-A, Oatp2 exerted no significant effect on deltorphin II uptake (Fig. 3). Kinetic transport analysis demonstrated saturability for all OATP-A-, Oatp1-, and Oatp2-mediated opioid peptide transport activities (Fig. 4). DPDPE was transported with highest affinity by Oatp2 followed by Oatp1 and OATP-A. The rat Oatp1 also exerted an ∼2-fold higher affinity for deltorphin II compared with OATP-A (Fig. 4). The reasons for these differences in opioid peptide transport affinities between OATP-A, Oatp1, and Oatp2 are not clear at present, but they may reflect the fact that the three organic anion-transporting polypeptides represent different gene products within the same transporter family. Thus, the data indicate that members of the Oatp gene family of membrane transporters play an important role in the transport of DPDPE and deltorphin II across the BBB of mammalian brain. This conclusion also may relate to other opioid peptides because OATP-A-mediated deltorphin II transport was inhibited not only by the OATP-A substrates estrone-3-sulfate and DPDPE but also by the μ-opioid receptor agonist DAMGO, the endogenous peptide Leu-enkephalin, and the opioid receptor antagonists naloxone and naltrindole (Fig.5).

• Download figure
• Open in new tab
• Download powerpoint

Figure 3

Uptake of DPDPE (A) and deltorphin II (B) in the presence (▪) and absence (□) of sodium mediated by OATP-A, Oatp1, and Oatp2 in X. laevis oocytes. Oocytes were injected with 5 ng of the respective cRNAs or with an equal volume of water. Uptakes of DPDPE (10 μM) and deltorphin II (50 μM) were measured at 30 min. Values are presented as mean ± S.D. of 7 to 10 oocyte uptake measurements. *P < .01 (unpairedt test; Systat 6.0.1, SPSS, Chicago, IL; significantly different from water-injected oocytes). **P < .001.

• Download figure
• Open in new tab
• Download powerpoint

Figure 4

Kinetics of OATP-A-, Oatp1-, and Oatp2-mediated DPDPE and deltorphin II uptake in cRNA (5 ng)-injected oocytes. Uptake measurements were performed in sodium-free (choline chloride) uptake medium at 30 min because separate experiments showed linear DPDPE and deltorphin II uptake up to 45 min. Unspecific uptake into water-injected oocytes was subtracted from all uptake measurements. Values are presented as mean ± S.E. of 7 to 10 oocyte uptake measurements. The curves were fitted by computerized nonlinear regression analysis (Systat).

• Download figure
• Open in new tab
• Download powerpoint

Figure 5

Cis-inhibition of OATP-A-mediated deltorphin II uptake in cRNA (5 ng)-injected oocytes. Sodium-independent (choline chloride) uptake of deltorphin II (50 μM) was measured at 30 min in the absence (control = 100%) and the presence of 0.5 mM DPDPE, 0.2 mM DAMGO and Leu-enkephalin, 1.5 mM naloxone and naltrindole, and 0.15 mM estrone-3-sulfate (E-3-S). Data are presented as mean ± S.D. of 7 to 10 oocyte uptake measurements.

Discussion

The present study has identified three members of theOatp gene family of membrane transporters as candidates for carrier-mediated translocation of the δ-opioid peptides DPDPE (human OATP-A, rat Oatp1, and Oatp2) and deltorphin II (OATP-A and Oatp1) across the BBB (OATP-A and Oatp2) and/or the blood-cerebrospinal fluid barrier (Oatp1 and Oatp2). Most importantly, the human OATP-A (previously called OATP; Kullak-Ublick et al., 1995, 1997, 1999; Meier et al., 1997; Noé et al., 1997) has been newly localized at the BBB (Fig. 2) and shown to transport the linear peptide deltorphin II and, albeit to a lesser extent, the cyclic peptide DPDPE (Figs. 3 and4). The latter also is transported by the rat Oatp2 (Fig. 3; Kakyo et al., 1999), which is highly expressed at the BBB and at the basolateral plasma membrane of choroid plexus epithelial cells (Gao et al., 1999). Finally, rat Oatp1, which has been localized only at the apical plasma membrane of choroid plexus epithelial cells (Angeletti et al., 1997;Gao et al., 1999), exhibits high-transport activities for both DPDPE and deltorphin II (Fig. 3). Although the present transport studies have been performed exclusively in cRNA-injected X. laevisoocytes, previous studies with Oatp1 and Oatp2 have shown excellent correlations between their transport properties in oocytes, in isolated hepatocytes, and in the intact liver (Meier et al., 1997;Kullak-Ublick, 1999), indicating that the newly identified opioid peptide transport activities of OATP-A, Oatp1, and Oatp2 are also functional at the BBB and/or at the blood-cerebrospinal fluid barrier in vivo. This assumption may even relate to deglycosylated Oatps/OATPs because mutation of N-linked glycosylation sites resulting in surface expression of deglycosylated protein did not affect the transport properties of Oatp1 (Wang et al., 1999).

DPDPE is a cyclic and enzymatically stable enkephalin analog that has been reported to enter brain, at least in part, by a saturable, probably carrier-mediated mechanism (Chen and Pollack, 1997a; Thomas et al., 1997b). Furthermore, the efficiency of its translocation across the BBB is an important rate-limiting step for the induction of antinociception (Chen and Pollack, 1997a). The half-life of DPDPE in vivo is short (∼14 min in rats; Chen and Pollack, 1996) due to its rapid and extensive hepatic elimination into bile (Weber et al., 1991;Chen and Pollack, 1997b). Although its efflux from brain across the BBB and its concentrative biliary excretion can be both explained by P-glycoprotein (Mdr1)-mediated transport (Chen and Pollack, 1998), this study indicates that DPDPE uptake across the BBB as well as into hepatocytes is mediated by OATP-A in humans and by Oatp2 in rats. This conclusion is supported by the expression of both organic anion-transporting polypeptides at the BBB (Fig. 2; Gao et al., 1999) and at the basolateral membrane of hepatocytes (Kullak-Ublick et al., 1995, 1997; Reichel et al., 1999) and by the similarity ofK_m values for DPDPE transport in Oatp2-expressing oocytes (∼19 μM; Fig. 2) and at the rat BBB (∼46 μM; Thomas et al., 1997b). The higherK_m value for human OATP-A compared with rat Oatp1- and Oatp2-mediated DPDPE transport probably reflects a species difference between humans and rats. Although the high-transport activity of the rat Oatp1 (Fig. 3) may help in the final elimination of DPDPE by rat liver (Eckhardt et al., 1999; Reichel et al., 1999), the exact role of Oatp1-mediated peptide transport at the apical domain of the choroid plexus remains unknown. An interesting possibility is that Oatp1 might mediate uptake of inflammatory cytokines from the cerebrospinal fluid in exchange for reduced glutathione (Gao et al., 1999). Whether and to what extent the polar localization of Oatp1 (apical) and Oatp2 (basolateral) at the choroid plexus (Gao et al., 1999) indicates their synergistic involvement in the suggested translocation of N-tyrosinated peptides from cerebrospinal fluid to blood (Banks and Kastin 1984) requires further investigation.

Deltorphin II is a linear heptapeptide with marked resistance to enzymatic degradation in biological fluids. Similar to DPDPE, the degree of deltorphin II-induced analgesic effects depends largely on its transfer across the BBB (Thomas et al., 1997a). In microvessels isolated from bovine brain, deltorphin II has been shown to exhibit saturable transport with an apparentK_m value of ∼150 μM (Fiori et al., 1997). We found a similar K_m value for Oatp1-mediated deltorphin II transport, which, however, cannot be adequately interpreted with respect to BBB transport because of the selective brain localization of Oatp1 in the choroid plexus (Angeletti et al., 1997; Gao et al., 1999). The ∼2-fold higherK_m value for OATP-A-mediated deltorphin II transport (Fig. 4) compared with bovine brain microvessels (Fiori et al., 1997) may be due to species differences between bovine and human organic anion-transporting polypeptides. The same also might be true for the differences in the inhibitory effects of DAMGO, Leu-enkephalin, and naltrindole on deltorphin II transport in OATP-A-expressing oocytes (Fig. 5) and in bovine brain microvessels (Fiori et al., 1997). However, similar to deltorphin II transport in bovine brain microvessels, OATP-A-mediated deltorphin II transport also was inhibited by the nonselective opiate antagonist naloxone (Fig. 5). Because deltorphin II transport was stimulated by preloading of bovine microvessels with glutamine (Fiori et al., 1997), and glutamine did not trans-stimulate deltorphin II uptake in OATP-A-expressing oocytes (data not shown), it is possible that deltorphin II transport across the BBB may be mediated by additional carriers, which might well include new human OATPs that are currently under investigation in various laboratories.

In conclusion, this study has identified human and rat members of the Oatp gene family of membrane transporters as candidates for the translocation of δ-opioid peptides across the BBB and the blood-cerebrospinal fluid barrier in the mammalian brain. Although the exact structural requirements for qualification as a substrate for a given Oatp/OATP isoform remain to be determined, it is interesting to note that both DPDPE and deltorphin II have terminal tyrosine residues and both Oatp1 and Oatp2 have been shown to actively transport thyroxine and triiodothyronine (Reichel et al., 1999). Because of their multiple localization, the same or closely related organic anion-transporting polypeptides are probably involved in the brain disposition and the hepatic elimination of opioid peptides. In this regard, it is interesting to note that elevated opioid agonists in plasma and increased opioidergic tone in brain may be responsible for centrally induced pruritis in patients with cholestatic liver disease (Jones and Bergasa, 1999). Thus, cholestatic liver injury might be associated with increased Oatp/OATP-mediated transport of opioid peptides across the BBB, an assumption that requires further investigation, including the analysis of expression and exact functional states of OATP-A and Oatp1/2 in BBB endothelial and choroid plexus epithelial cells in patients and animals with decreased liver function. Finally, it is highly probable that additional human OATPs (e.g., OATP-B and OATP-C) that are currently being characterized in different laboratories also may have transport affinities for centrally active peptides. Hence, the identification of OATP-A, Oatp1, and Oatp2 as opioid peptide transporters and their localization at the BBB and/or the choroid plexus (Gao et al., 1999) should facilitate future research regarding the development, brain targeting, and cerebral disposition of therapeutically important new drugs acting on the central nervous system.