Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

Many drugs reach high concentrations in the liver due to rapid uptake from the bloodstream by passive, receptor- or transporter-mediated processes. At the sinusoidal membrane of hepatocytes, Na⁺-dependent taurocholate-cotransporting polypeptide (NTCP; Ntcp in rodent), organic anion transporting polypeptide (OATP; oatp in rodent), and organic cation transporters (OCT; Oct in rodent) are expressed (Lecureur et al., 2000). Among them, members of the OATP family can mediate the hepatic uptake of various drugs and xenobiotics, such as the organic anion bromosulfophthalein (Cui et al., 2001; Kullak-Ublick et al., 2001), benzylpenicillin (Tamai et al., 2000a), and the 3-hydroxy-3-methylglutaryl-CoA-reductase inhibitor pravastatin (Hsiang et al., 1999).

Oligopeptide transporters PEPT1 and PEPT2 accept not only dipeptides and tripeptides as substrates but also peptide-mimetic drugs, such as -lactam antibiotics (Saito et al., 1995; Miyamoto et al., 1996; Sai et al., 1996; Tamai et al., 1997), angiotensin-converting enzyme inhibitors (Hu and Amidon, 1998), the antiviral drug valacyclovir (Balimane et al., 1998), and the anticancer drug bestatin (Saito and Inui, 1993). It is well known that transport activity of oligopeptide transporters is pH-dependent (Fei et al., 1994; Liang et al., 1995). The physiological role of both transporters lies in the (re)absorption of peptides from the intestinal and renal tubular lumen (Daniel, 2000). In the liver, however, it has been reported that the ability to take up small peptides from the circulation was negligible (Lombardo et al., 1988). From these reports, the liver seems to lack expression of PEPT1 or PEPT2 at the sinusoidal membrane of hepatocytes.

There have been several experimental trials aimed at using the endogenously expressed oligopeptide transport activity for improving oral bioavailability (Tamai et al., 1998) or using cultured cells, such as the human fibrosarcoma cell line HT-1080 (Nakanishi et al., 1997) and the human pancreatic cell lines AsPc-1 and Capan-2 (Gonzalez et al., 1998) that express oligopeptide transport activity for tumor targeting. From this point of view, we previously examined the feasibility of tumor-selective delivery of dipeptides or peptide-mimetic drugs by using the oligopeptide transport activity (Nakanishi et al., 2000). To our knowledge, however, there has been no experimental trial on drug delivery using the activity of oligopeptide transporter in the liver since these transporters are unlikely to be expressed in the liver.

Recombinant adenovirus is an attractive method for gene transfer in vitro and for in vivo gene therapy (He et al., 1998). It is believed that high levels of transgene expression can generally be obtained in comparison with other viral or nonviral vectors, such as retrovirus or lipofectamine, respectively (Sato et al., 2000). In addition, since it has the advantage of having little integrated into the genome, it is possible to minimize insertional mutagenesis (Jaffe et al., 1992). Therefore, we attempted to explore a new concept of drug delivery of oligopeptide drugs into the liver by the use of adenovirus-mediated heterologous expression of oligopeptide transporter.

In the present study, to deliver small peptides to the liver selectively, we constructed a recombinant adenovirus encoding a fusion gene of human PEPT1 and enhanced yellow fluorescent protein (AdhPEPT1-EYFP) fusion gene. Then, oligopeptide transport activity was assessed in the polarized hepatoma cell lines HepG2 and WIFB9 transduced with or without AdhPEPT1-EYFP in vitro. After transduction of AdhPEPT1-EYFP into mice, the distribution of L-[³H]carnosine, a substrate of the oligopeptide transporter hPEPT1, into the liver was evaluated in vivo.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Materials. HepG2 cells were purchased from Dainippon Pharmaceutical Co. Ltd. (Osaka, Japan). HeLa cells stably expressing of hPEPT1 were previously prepared in our laboratory (Nakanishi et al., 2000). WIFB9 cells were used as described previously (Sai et al., 1999). Dulbecco's modified Eagle's medium, fetal calf serum (FCS), and nonessential amino acids were obtained from Invitrogen (Carlsbad, CA). Modified Eagle's medium was obtained from Nissui (Tokyo, Japan). Modified F-12 medium was purchased from Sigma-Aldrich (St. Louis, MO). Hypoxanthine, aminopterin, and thymidine were obtained from Sigma-Aldrich. pcDNA3 vector was purchased from Invitrogen. L-[³H]Carnosine (370 GBq/mmol) and [³H]glycylsarcosine (GlySar) (629 GBq/mmol) were purchased from Moravek Biochemical, Inc. (Brea, CA). [¹⁴C]Inulin (161 MBq/g) was purchased from ICN Biomedicals, Inc. (Costa Mesa, CA). Unlabeled L-carnosine and glycylsarcosine were purchased from Sigma-Aldrich. The protein assay kit was purchased from Bio-Rad (Hercules, CA). All other chemicals were commercial products of reagent grade. The pBluescript II SK() vector containing human PEPT1 2.2-kilobase cDNA was a gift from Prof. F. H. Leibach (Medical College of Georgia, Augusta, GA).

Construction of Recombinant Adenovirus Coding hPEPT1-EYFP Fusion Gene. Two primers were designed to perform polymerase chain reaction (PCR) for amplification of the 3'-end 524-bp fragment to mutate the stop codon and to introduce HpaI/SmaI-cloning sites. The two primer sequences are as follows: the sense primer F3M (5'-TATGGATCCAACCTAATTTCAATACTTTCTACC-3') containing the sequence nucleotides 1674 to 1698 of hPEPT1 and an additional BamHI site and the antisense primer R4M (5'-TATCCCGGGACATCTGTTTCTGTGAATTGGC-3'), which contains the last 18 nucleotides of the hPEPT1 coding sequence, excluding the stop codon, and a SmaI site. PCR was performed in a final volume of 100 µl of PCR buffer containing 0.2 µM sense and antisense primers, 90 µl of Platinum PCR SuperMix (Invitrogen), and 1 µl of hPEPT1-pBluescript II SK(). The PCR fragment was digested with HpaI/SmaI and ligated into HpaI/SmaI-digested pBluescript containing hPEPT1 cDNA. The hPEPT1-EYFP vector was constructed by subcloning the full-length hPEPT1, which lacked the stop codon, into the KpnI/SmaI sites of the multiple cloning site of pEYFP-N1 (BD Biosciences Clontech, Palo Alto, CA). The vector containing the hPEPT1-EYFP fusion gene was used to generate the recombinant adenovirus AdhPEPT1-EYFP. To generate pAdhPEPT1-EYFP, the 2.2-kilobase hPEPT1-EYFP insert digested with KpnI/NotI was ligated into the multiple cloning site of pShuttle-CMV vector (kindly provided by T.-C. He, Howard Hughes Medical Institute, Chevy Chase, MD). The constructs were restriction enzyme-mapped and sequenced to determine insert orientation. The resulting plasmid was linearized by digestion with PmeI and cotransformed into Escherichia coli BJ5183 with an adenoviral backbone plasmid, pAdEasy-1 (He et al., 1998). Finally, the recombinant plasmid was linearized with PacI and transfected into HEK293 cells. For amplification, the recombinant adenovirus was propagated in HEK293 cells and purified by CsCl banding. AdhPEPT1-EYFP was stored at the concentration of 1.0 × 10¹¹ plaque-forming units (PFU)/ml.

Cell Culture. HeLa cells transfected with hPEPT1 or vector DNA (pcDNA3) alone were grown in Dulbecco's modified Eagle's medium supplemented with 10% FCS, 2 mM glutamine, and 0.6 mg/ml G418.

In Vitro Gene transfer to Cells. Cells were plated onto a four-well plate (Nalge Nunc, Naperville, IL) for 3 days (for HeLa cells) or 7 days (for HepG2 and WIFB9 cells) before transduction. Transduction of adenovirus was performed after cells reached 70 to 90% confluence, with AdhPEPT1-EYFP at the multiplicity of infection (MOI) of 40. The uptake experiment was performed 1 day after transduction for HeLa cells or 2 days after transduction for HepG2 and WIFB9 cells. Phase contrast and fluorescence images of the cells were obtained using a Zeiss Axiovert S100 microscope (Carl Zeiss, Inc., Thornwood, NY) equipped with an appropriate filter unit.

Uptake Experiment. Uptake of [³H]GlySar (58.8 nM) and L-[³H]carnosine (100 nM) by cultured cells was examined at 37°C by the method described previously (Nakanishi et al., 2000). The flux was measured in Hanks' balanced salt solution (HBSS) (0.952 mM CaCl₂, 5.36 mM KCl, 0.441 mM KH₂PO₄, 0.812 mM MgSO₄, 136.7 mM NaCl, 0.385 mM Na₂HPO₄, 25 mM D-glucose, and 10 mM MES or HEPES) adjusted to pH 6.0 with NaOH. Osmolality of the HBSS was adjusted to 310 mOs/kg. To quantify [³H]GlySar or L-[³H]carnosine in the cells, the washed cells were solubilized by the addition of 5 N NaOH (0.25 ml), followed by shaking for 2 h. The resultant lysates were neutralized with 5 N HCl and mixed with 4 ml of the liquid scintillation cocktail Clearsol-I (Nacalai Tesque, Kyoto, Japan). Radioactivity was determined using a liquid scintillation counter (LSC-1000; Aloka Co. Ltd., Tokyo, Japan). Protein determinations were done using the protein-dye binding method with bovine serum albumin as a standard (Bradfold, 1976).

In Vivo Gene Transfer to Mice. All the animal experiments were performed according to the Guidelines for the Care and Use of Laboratory Animals of Takara-machi Campus of Kanazawa University. Male ddY mice (6-8 weeks old; Nippon SLC, Hamamatsu, Japan) were used in this study. All animals received standard mouse chow and water ad libitum. Mice were anesthetized with intramuscular administration of ketamine/xylazine (140:8 mg/kg). AdhPEPT1-EYFP and AdGFP were injected through the right jugular vein using a 0.5-ml syringe with a 29.5-gauge needle. The total amount of adenovirus injected per mouse was 1 × 10⁹ or 1 × 10¹⁰ PFU for AdhPEPT1-EYFP. AdGFP was given at a dose of 2.4 × 10⁸ PFU. The dose of AdGFP was selected to let fluorescence expression level of GFP comparable to that of hPEPT1-EYFP in mouse liver received 1 × 10¹⁰ PFU AdhPEPT1-EYFP. GFP expression level was confirmed by fluorescent microscopy.

Tissue Preparation for Analysis of hPEPT1-EYFP Transgene Expression. After sacrifice of the mice, tissues were immediately dissected and washed with PBS, then quickly frozen in liquid nitrogen and stored at 80°C. For fluorescence analysis of hPEPT1-EYFP expression, resected tissues were embedded in optimal cutting temperature compound (OCT; Sakura Finetechnical Co., Ltd., Tokyo, Japan). Serial 10-µm tissue sections were prepared with a Cryostat HM505E (Carl Zeiss, Inc.) at 20°C. Fluorescence images of sections were obtained using a Zeiss Axiovert S100 microscope (Carl Zeiss, Inc.).

Reverse Transcriptase-Polymerase Chain Reaction. Total RNA was extracted from frozen tissue using RNeasy Mini-Kit (QIAGEN, Chatsworth, CA) according to the manufacturer's instructions. Total RNA was reverse-transcribed and amplified by PCR in the presence of specific primers for hPEPT1-EYFP. The primer pair derived from the hPEPT1-EYFP cDNA sequence was as follows. The upstream primer was 3'-CCAACTGTAACACCGCCTTAG-5', and the downstream primer was 5'-CCTCTACAAATGTGGTATGGCTG-3', yielding a 1379-bp fragment. cDNA synthesis and predenaturation were performed at 48°C/30 min and 94°C/2 min. Amplifications consisted of 1 cycle at 94°C for 2 min followed by 40 cycles at 94°C for 30 s, 58°C for 30 s, and 72°C for 90 s, with an extension step at 72°C for 10 min. To demonstrate the integrity of the RNA, we measured glyceraldehyde-3-phosphate dehydrogenase (GAPDH) expression using primers 5'-TGAAGGTCGGTGTGAACGGATTGGC-3' and 5'-CATGTAGGCCATGAGGTCCACCAC-3', yielding a 983-bp fragment. cDNA synthesis and predenaturation were performed at 48°C/30 min and 94°C/2 min. PCR was performed in a total volume of 50 µl at 94°C/30 s, 60°C/30 s, 72°C/90 s (40 cycles), and 72°C/10 min for extension.

Western Blot Analysis. Mouse liver membranes were prepared as previously described (Tamai et al., 2000b). Briefly, 800 µl of the homogenate was mixed with 750 µl of 1.17 M KCl solution containing 58.3 mM tetrasodium pyrophosphate and centrifuged at 230,000g for 75 min. The resultant pellet was suspended in 10 mM Tris-HCl and 1 mM EDTA, pH 7.4, and centrifuged at 230,000g again. The obtained pellet was suspended in 600 µl of 10 mM Tris-HCl and 1 mM EDTA, pH 7.4, and dispersed ultrasonically. After the addition of 200 µl of 16% SDS solution, the solution was mixed and centrifuged at 15,000g for 2 min, and the resultant supernatant was used for Western blot analysis. Protein quantification of samples was performed using the Lowry assay. Each sample was separated on 12% SDS-polyacrylamide gel, proteins were transferred to polyvinylidene difluoride membranes (Immobilon, Millipore, Bedford, MA), and the membrane was incubated in buffer consisting of 20 mM Tris, 137 mM NaCl, and 0.1% Tween-20, pH 7.5, containing 10% skim milk. The membrane was incubated with primary anti-EGFP antibody (Living Colors-A. v. Peptide Antibody; BD Biosciences Clontech) for 1 to 2 h, rinsed with the above buffer without skim milk three times, and incubated with secondary antibody (donkey anti-rabbit IgG, horseradish peroxidase-linked whole antibody; Amersham Biosciences UK, Ltd., Little Chalfont, Buckinghamshire, UK). The membrane was washed with the above buffer without skim milk, and the proteins were detected by enhanced chemiluminescence detection method using an ECL Plus Western blotting detection system (Amersham Pharmacia Biotech UK, Ltd.).

Immunofluorescence Analysis. Serial 10-µm liver sections were prepared with a cryostat at 20°C and fixed in methanol for 1 min. Sections were treated with 0.3% Tween-20 in PBS for 20 min and blocked for 30 min with 3% blocking agent (Amersham Pharmacia Biotech UK, Ltd.) in PBS. Then their samples were incubated with primary C219 anti-mdr1 antibodies (Signet Pathology Systems, Inc., Dedham, MA) at a 1:10 dilution for 1 h at room temperature. The incubation was followed by three 5 min washes in PBS. Secondary antibody Alexa Fluor 594 goat anti-mouse IgG (Molecular Probes, Inc., Leiden, The Netherlands) diluted with 3% blocking agent in PBS at 1:200 was added, and incubation continued for 30 min. The slides were washed as described above. They were mounted in VECTASHIELD (Vector Laboratories, Inc., Burlingame, CA) and observed under a fluorescence microscope.

L-Carnosine Disposition Using Anesthetized Mice Transduced with Adenovirus. Mice were anesthetized with intramuscular injection of ketamine/xylazine (140/8 mg/kg) 2 days after virus infection. A 100-µl aliquot of saline solution containing test compounds (L-[³H]carnosine, 5 µCi; [¹⁴C]inulin, 0.5 µCi/mouse/25 g) was injected through the external jugular vein. Blood (30-40 µl) was collected at the indicated time with heparinized microhematocrit capillary tubes. At 30 min after administration, mice were sacrificed, and then tissues were quickly isolated and rinsed with ice-cold PBS. The tissues were solubilized by SOLUENE-350 (Packard BioScience B.V., Groningen, The Netherlands) for 3 h, and radioactivity was measured using a liquid scintillation counter.

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

Uptake of [³H]GlySar by HeLa Cells Transduced with AdhPEPT1-EYFP. To examine whether the dipeptide transport activity is induced by AdhPEPT1-EYFP transduction, we evaluated [³H]GlySar uptake in HeLa cells. When HeLa cells were transduced with AdhPEPT1-EYFP, uptake 20 to 30 times higher than that of nontransduced HeLa cells was observed at 15 min (Fig. 1). The uptake of dipeptide was higher than that by HeLa cells stably transfected with hPEPT1-encoding plasmid vector. When HeLa cells were infected with AdhPEPT1-EYFP at an increasing MOI from 40 to 400, 1.5-fold increase of [³H]GlySar uptake was observed. Expression level of hPEPT1-EYFP protein in HeLa cells was increased with MOI (data not shown).

View larger version (26K): [in this window] [in a new window]   Fig. 1.   Confirmation of hPEPT1-EYFP expression on the plasma membrane in AdhPEPT1-EYFP-transduced HeLa cells. [3H]GlySar uptake by HeLa cells transfected with the pcDNA vector alone (open column), the plasmid vector containing cDNA for hPEPT1 (hatched column), and adenovirus containing hPEPT1-EYFP fusion gene (solid column). HeLa cells was transfected by AdhPEPT1-EYFP at MOI 40 or 400. Uptake of [3H]GlySar (58.8 nM) by transfected HeLa cells was measured at 37°C by incubating cells in HBSS at pH 6.0 for 15 min. Data are presented as means ± S.E.M. of four experiments.

Expression of hPEPT1-EYFP in HepG2 and WIFB9 Cells. The expression of hPEPT1-EYFP was examined by fluorescence microscopy in HepG2 and WIFB9 cells. Those cells were transduced by AdhPEPT1-EYFP at MOI 40. As shown in Fig. 2, fluorescence of hPEPT1-EYFP was observed in both HepG2 and WIFB9 cells. In addition, the expression was observed at both basolateral and apical membranes of HepG2 and WIFB9 cells.

View larger version (118K): [in this window] [in a new window]   Fig. 2.   Localization of hPEPT1-EYFP fusion protein in adenovirus -transduced HepG2 and WIFB9 cells. Fluorescence (A) and phase-contrast (B) micrographs of AdhPEPT1-EYFP-transduced HepG2 cells were obtained by conventional fluorescence microscopy. Fluorescence (C) and phase-contrast (D) micrographs of AdhPEPT1-EYFP-transduced WIFB9 cells were also obtained by similar methods. Cells were infected with AdhPEPT1-EYFP at MOI 40. Arrows indicate the sinusoidal membrane. Scale bar is 20 µm.

Uptake of L-[³H]Carnosine in HepG2 and WIFB9 Cells Transduced with AdhPEPT1-EYFP. To examine whether the dipeptide transport activity was induced in AdhPEPT1-EYFP-transduced HepG2 and WIFB9 cells, uptake of L-[³H]carnosine was measured for 15 min. As shown in Fig. 3, AdhPEPT1-EYFP-transduced HepG2 and WIFB9 cells (MOI 40) showed 10 and 2 times higher uptake, respectively, than nontransduced controls. When AdGFP was transduced in HepG2 cells as a control vector at MOI 40, uptake of L-[³H]carnosine was similar to nontransduced cells (data not shown).

View larger version (19K): [in this window] [in a new window]   Fig. 3.   Uptake of L-[3H]carnosine by HepG2 and WIFB9 cells transduced with AdhPEPT1-EYFP. Time courses of L-[3H]carnosine uptake by HepG2 (A) and WIFB9 (B) cells. Cells were nontransduced (open circle) or were transduced with AdhPEPT1-EYFP at MOI 40 (closed circle). Uptake of L-[3H]carnosine (100 nM) by transduced HepG2 and WIFB9 cells was measured at 37°C by incubating cells in HBSS at pH 6.0. Data are presented as means ± S.E.M. of four experiments.

View larger version (18K): [in this window] [in a new window]   Fig. 4.   Effect of inhibitor and pH on the uptake of L-[3H]carnosine by HepG2 cells transduced with AdhPEPT1-EYFP. A, competition for L-[3H]carnosine uptake by inhibitor in HepG2 cells with or without AdhPEPT1-EYFP transduction at MOI 40. Uptake of L-[3H]carnosine (100 nM) by HepG2 in the absence or presence of inhibitor (20 mM carnosine or 20 mM glycylsarcosine) was measured at 37°C by incubating cells in HBSS at pH 6.0 for 15 min. B, uptake of L-[3H]carnosine in HepG2 cells with (closed columns) or without (open columns) AdhPEPT1-EYFP transduction at pH 6.0 or 7.4 using a 15-min incubation period at 37°C. Data are presented as means ± S.E.M. of four experiments.

Expression of hPEPT1-EYFP in Mouse Liver after AdhPEPT1-EYFP Transduction. To investigate whether hPEPT1-EYFP transcript was expressed in vivo after intravenous administration of AdhPEPT1-EYFP into mice, RT-PCR of hPEPT1-EYFP was conducted. At the virus dose of 1 × 10⁹ PFU/mouse, mRNA of hPEPT1-EYFP was detected only in the liver (Fig. 5). At the virus dose of 1 × 10¹⁰ PFU, the hPEPT1-EYFP transgene was predominantly found in the liver, and low levels were present in the kidney and the spleen. To evaluate whether the hPEPT1-EYFP protein was induced in vivo, expression in the liver was examined by fluorescence microscopy and Western blot analysis. As shown in Fig. 6A, at the virus dose of 1 × 10⁹ PFU/mouse, hPEPT1-EYFP fluorescence was very low (Fig. 6A) despite the presence of mRNA signals (Fig. 5A). However, when the dose was increased to 1 × 10¹⁰ PFU/mouse, fluorescence of hPEPT1-EYFP markedly increased in the liver (Fig. 6A, c) but was not detected in the kidney or the spleen (data not shown). On the other hand, Western blot analysis (Fig. 6B) indicated the expression of the fusion protein with a molecular mass of about 130 to 140 kDa in the membrane of the liver after transduction with AdhPEPT1-EYFP. In addition, the amount of hPEPT1-EYFP protein increased with an increase of the viral dose from 1 × 10⁹ to 1 × 10¹⁰ PFU/mouse. No expression of hPEPT1-EYFP was observed in nontransduced liver.

View larger version (39K): [in this window] [in a new window]   Fig. 5.   RT-PCR analysis of mRNA levels of hPEPT1-EYFP in various organs after adenovirus-transduction of mice. Expression of hPEPT1-EYFP mRNA in mouse transduced with AdhPEPT1-EYFP at the dose of 1 × 109 PFU/mouse (A) or 1 × 1010 PFU/mouse (B) was detected by RT-PCR. Total RNA extracted from liver, kidney, and spleen of transduced mouse was subjected to RT-PCR using primer pairs specific for hPEPT1-EYFP or GAPDH. The RT-PCR products were analyzed by agarose gel electrophoresis and stained with ethidium bromide. GAPDH was used as a control for RNA integrity.

View larger version (19K): [in this window] [in a new window]   Fig. 6.   In vivo expression of hPEPT1-EYFP fusion protein in liver after adenovirus-transduction of mice. A, fluorescence microscopic images of cryosections of AdhPEPT1-EYFP-transduced mouse liver. Nontransduced liver of mice was used as a control (a). The liver tissues of transduced mice were obtained 2 days after i.v. injection of AdhPEPT1-EYFP at a dose of 1 × 109 PFU/mouse (b) or 1 × 1010 PFU/mouse (c) (original magnification, 200×). B, Western blotting analysis of the expression of hPEPT1-EYFP fusion protein in the liver. Nontransduced mouse liver was used as a control (a). Mice were transduced with AdhPEPT1-EYFP at a dose of 1 × 109 PFU/mouse (b) or 1 × 1010 PFU/mouse (c). Each lane was loaded with 20 µg of protein. Molecular masses are indicated.

View larger version (68K): [in this window] [in a new window]   Fig. 7.   Cellular localization of hPEPT1-EYFP fusion protein in mouse liver after adenovirus-transduction. A semithin cryosection (10 µm) of the liver of the mouse transduced with AdhPEPT1-EYFP at a dose of 1 × 1010 PFU/mouse was subjected to indirect immunofluorescence microscopy. A, expression of hPEPT1-EYFP (green) was directly visualized by its fluorescence under microscopy. B, C219 monoclonal antibody against P-glycoprotein was used to identify canalicular membrane domains (red). C, superimposition of A and B revealed only partial colocalization of the two molecules in the canalicular membranes (yellow), suggesting that hPEPT1-EYFP was expressed in both sinusoidal (green, arrowheads) and canalicular membranes (arrows) of hepatocytes. Scale bar is 20 µm.

Drug Disposition in AdhPEPT1-EYFP-Transduced Mice. To evaluate whether the disposition of dipeptides was improved after AdhPEPT1-EYFP transduction of mice, the tissue distribution of L-[³H]carnosine was measured at 30 min after administration, and K_p values, which were obtained by dividing the total concentration of L-[³H]carnosine in the tissue by that in plasma, were evaluated. [¹⁴C]Inulin was used to estimate the distribution in the extracellular fluid space because it hardly enters the cell (Tsuji et al., 1983). As shown in Fig. 8, in the nontransduced mice, the K_p value of L-[³H]carnosine was very high in kidney compared with other tissues. In particular, L-[³H]carnosine was distributed less to the liver than the kidney. Distribution of L-[³H]carnosine in the kidney may be ascribed to the reabsorption of it by the endogenous mouse oligopeptide transporter in the kidney (Rubio-Aliaga et al., 2000). In the liver of AdhPEPT1-EYFP-transduced mice at the dose of 1 × 10¹⁰ PFU/mouse, the K_p value of L-[³H]carnosine was about 7 times higher than that in nontransduced mice. There was no significant difference in the K_p values of [¹⁴C]inulin between nontransduced and AdhPEPT1-EYFP-transduced mouse liver, which indicated minimal pathological liver damage. When AdGFP was infected as a control vector, no significant increase in the K_p value of L-[³H]carnosine in the liver was observed, and the value was comparable to that of nontransduced mice. To investigate the correlation between the accumulation of L-[³H]carnosine and the expression of AdhPEPT1-EYFP, we evaluated K_p of L-[³H]carnosine in mice transduced with AdhPEPT1-EYFP at different doses (Table 1). When mice were transduced with AdhPEPT1-EYFP at a higher dose, the mean value of K_p of L-carnosine was increased, whereas that of inulin was unchanged. No significant difference, however, was observed between nontransduced mice and mice given the dose of 1 × 10⁹ PFU/mouse.

View larger version (24K): [in this window] [in a new window]   Fig. 8.   Kp value of L-[3H]carnosine and [14C]inulin (inset) in various organs of mice without transduction (open column) or with transduction of AdhPEPT1-EYFP (closed column) and AdGFP (striped column). Kp values were obtained 30 min after intravenous administration of these compounds in mice with or without transduction of adenovirus. Each value represents the mean ± S.E.M. of three mice. *, significantly increased at a dose of 1 × 1010 PFU/mouse compared with nontransduced mice by Student's t test (p < 0.05).

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

In the current study, we attempted to induce expression of the oligopeptide transporter hPEPT1 in mouse liver by using a recombinant adenovirus to examine the feasibility of this approach for drug delivery. A recombinant adenovirus encoding hPEPT1-EYFP was constructed. EYFP was used to monitor the expression level of hPEPT1. The molecular mass of hPEPT1-EYFP protein is 130 to 140 kDa by Western blot analysis (Fig. 6B). Basu et al. (1998) reported that immunoblotting of a Caco-2 membrane protein preparation using the anti-peptide antibody against hPEPT1 revealed a positive band with an apparent molecular mass of ~110 kDa. The anti-EGFP antibody recognized the EYFP tag in hPEPT1-EYFP protein and showed a 30-kDa band in the control GFP vector-transduced liver (data not shown). This result shows that hPEPT1-EYFP protein expressed in the liver is about 30 kDa larger than hPEPT1 alone. The function of this protein was confirmed by uptake studies in AdhPEPT1-EYFP-transduced cell lines (Figs. 1, 3, and 4). Uptake of [³H]carnosine by AdhPEPT1-EYFP-transduced HepG2 cells was time-and pH-dependent and inhibited by dipeptides. Thus, the functional expression of hPEPT1-EYFP protein as an oligopeptide transporter protein was successfully achieved by adenoviral transduction.

In the case of delivery of drugs from the systemic circulation to the liver under physiological conditions, the pH of the systemic circulation is 7.4. In our study, uptake of [³H]carnosine in AdhPEPT1-EYFP-transduced HepG2 cells was observed at pH 7.4 (Fig. 4B) and pH 6.0. Transport activity of hPEPT1 at neutral pH has also been reported by other investigators (Liang et al., 1995; Covitz et al., 1996). Transport of glycylsarcosine in Xenopus laevis oocytes injected with hPEPT1 cRNA was observed at pH 7.5 (Liang et al., 1995). Uptake of [³H]GlySar in Chinese hamster ovary cells stably expressing hPEPT1 was also observed at pH 7.5 (Covitz et al., 1996). These results are consistent with the present study. Therefore, we consider that hPEPT1-EYFP expression was functional at neutral pH in liver cells in this model after AdhPEPT1-EYFP transduction.

The localization of hPEPT1-EYFP protein is critical for delivery of peptide drugs from the systemic circulation to the liver. Our results demonstrate that hPEPT1-EYFP protein was expressed at apical and basolateral membranes in HepG2 or WIFB9 cells (Fig. 2) and mouse liver (Fig. 7). Sun et al. (2001) detected hPEPT1-GFP at apical and basolateral membranes in Caco-2 cells. It was reported that hPEPT1 was present in the plasma membrane and intracellular vesicular structures of AsPC-1 and Capan-2 cells by Gonzalez et al. (1998). hPEPT1 is localized in nuclei of vascular smooth muscle cells and lysosomes of the exocrine pancreas (Bockman et al., 1997). Thus, the distribution of hPEPT1 in various cells is diverse. Our studies demonstrate that the K_p value of [³H]carnosine was increased in AdhPEPT1-EYFP-transduced mouse liver (Fig. 8). Therefore, the hPEPT1-EYFP fusion protein is functional at the sinusoidal membrane of the liver. However, significant cytoplasmic staining was also observed in HepG2 or WIFB9 cells and mouse liver, probably due to the overexpression of the transporter. Further investigation is required, including colocalization of hPEPT1-EYFP with various subcellular markers.

For drug delivery to the liver, it is important to control the drug concentration. Kurata et al. (1999) reported that expression of luciferase depended on a viral dose between 10⁷ and 10⁹ PFU/mouse. In our case, [³H]GlySar uptake by HeLa cells depended on the dose of AdhPEPT1-EYFP (Fig. 1). These results are consistent with the observation by Hsu et al. (1998) that [³H]GlySar uptake increased depending on the MOI in Caco-2 cells. The expression levels of mRNA and protein of hPEPT1-EYFP in mouse are dependent on the dose of AdhPEPT1-EYFP (Figs. 5 and 6). At 10⁹ PFU, a small increase of the average value of [³H]carnosine K_p in the liver, although not statistically significant, was observed, and we detected significant expression of hPEPT1-EYFP mRNA and protein. A plausible explanation is that the relative expression of hPEPT1-EYFP protein in the apical and the basolateral membranes differs with viral dosage. In the liver of mice given the dose of 10⁹ PFU, hPEPT1-EYFP may localize mainly at the apical membrane, as observed in intestinal epithelial cells (Sai et al., 1996; Walker et al., 1998), and consequently, the uptake of carnosine by hPEPT1-EYFP at the basolateral membrane is low. When a high dose of 10¹⁰ PFU was injected, hPEPT1-EYFP protein was expressed both in the apical and basolateral membranes (Fig. 7), resulting in a large increase of K_p value in the liver. Therefore, to control drug disposition in the liver, the dose of adenovirus must be optimized according to the required drug concentration.

Duration of transgene expression has been examined by many investigators. Jaffe et al. (1992) reported that intraportal infusion of Ad-1AT (human 1-antitrypsin cDNA) produced a detectable serum level of human 1AT for 4 weeks. Recombinant adenovirus expression of human pancreatic lipase gene was sustained for 7 days (Kuhel et al., 2000). In our experiment, the duration of expression of AdGFP was 4 weeks in mouse liver (data not shown), whereas that of hPEPT1-EYFP is unknown. Recombinant adenovirus has advantages as a vector system in minimizing insertional mutagenesis and affording transient expression of the transgene. Therefore, duration of gene expression must be considered to optimize the concentration in the liver.

PEPT1 is a low-affinity oligopeptide transporter (e.g., apparent K_m for carnosine is 12.9 mM), whereas PEPT2 is the high-affinity one (Ramamoorthy et al., 1995), so PEPT1 is expected not to be saturated by substrates in the circulation. Thus, we preferred PEPT1 as a candidate for drug delivery from the systemic circulation to the liver. Because PEPT1 accepts a wide range of peptide and peptide-like substances as substrates (Saito and Inui, 1993; Saito et al., 1995; Miyamoto et al., 1996; Sai et al., 1996; Tamai et al., 1997; Hu and Amidon, 1998; Balimane et al., 1998), other peptide drugs may be transported. This system is also applicable to chemically modified substrates that are recognized by PEPT1. Candidates include L-valyl ester prodrugs of zidovudine and acyclovir. Indeed, the absorption of acyclovir in the intestine was improved by peptide-mimetic derivation (Soul-Lawton et al., 1995).

In conclusion, we have demonstrated a novel strategy for drug delivery to the liver by means of adenovirus-mediated heterologous expression of an oligopeptide transporter gene. Further studies on regulation of hPEPT1-EYFP in the liver may allow accurately controlled delivery of small peptides.