Báo cáo khoa học: Transport of taurocholate by mutants of negatively charged amino acids, cysteines, and threonines of the rat liver sodium-dependent taurocholate cotransporting polypeptide Ntcp

doi:10.1046/j.1432-1033.2003.03463.x

Eur. J. Biochem. 270, 1117–1127 (2003) (cid:2) FEBS 2003

Transport of taurocholate by mutants of negatively charged amino acids, cysteines, and threonines of the rat liver sodium-dependent taurocholate cotransporting polypeptide Ntcp

Daniel Zahner, Uta Eckhardt and Ernst Petzinger

Institute of Pharmacology and Toxicology, Justus-Liebig-University Giessen, Germany

extracellular loop VI. Four mutations of threonines from the C-terminus of the Ntcp by alanines or tyrosines showed no effects on sodium-dependent taurocholate transport. Intro- duction of the FLAG(cid:1) motif into several transport negative point mutations demonstrated that all mutated proteins besides one were present within the cell membrane of the oocytes and provided proof that an insertion defect has not caused transport deficiency by these Ntcp mutants. The latter was observed only with the transport negative mutant Asp24Asn. In conclusion, loop amino acids are required for sodium-dependent substrate translocation by the Ntcp.

Keywords: bile acids; P-loop; glutamate; aspartate; mem- brane protein.

The relevance of functional amino acids for taurocholate transport by the sodium-dependent taurocholate cotrans- porting polypeptide Ntcp was determined by site-directed mutagenesis. cRNA from 28 single-points mutants of the rat liver Ntcp clone was expressed in Xenopus laevis oocytes. Mutations were generated in five conserved negatively charged amino acids (aspartates and glutamates) which were present in nine members of the SBAT-family, in two non- conserved negatively charged amino acids, in all eight Ntcp- cysteines, and in two threonines from a protein kinase C consensus region of the Ntcp C-terminus. Functional amino acids were Asp115, Glu257, and Cys266, which were found to be essential for the maintenance of taurocholic acid transport. Asp115 is located in the large intracellular loop III, whereas Glu257 and Cys266 are located in the large

predicted structure which is derived from hydrophobicity analysis contains either seven or nine transmembrane domains; all SBATs are glycosylated at the extracellular N-terminus and contain a cytoplasmic C-terminus.

The sodium-dependent taurocholate cotransporting poly- peptide Ntcp from rat liver is the major basolateral bile acid transporter of rat hepatocytes. It was the first sodium- dependent bile acid cotransporter (SBAT), that was obtained by expression cloning in Xenopus laevis oocytes [1]. It exhibits 77.4% identity and 88.8% similarity on amino acid level with the human transporter NTCP [2]. The proteins are coded by the Slc/SLC10 gene family in animals and man. SBATs are involved in the maintenance of the enterohepatic circulation of bile acids and therefore also participate in the homoeostatis of cholesterol. Members of SBATs are located either in apical membranes of ileum enterocytes, kidney tubule cells and bile duct cells where they perform bile acid reabsorption, or in the basolateral membrane of hepatocytes where they initiate bile acid secretion [1–9]. SBATs constitute a subgroup of the sodium-dependent cotransporters with superfamily of about 35% homology among the clones from different species, e.g. from rat, mouse, rabbit, hamster, and human [10]. Their molecular mass is about 50 kDa and the

All carriers transport sodium ions together with an organic substrate, e.g. a bile acid or an anionic sulfated or glucuroni- dated estrogen conjugate. The stoichiometry of this process is electrogenic; two sodium ions are supposed to be trans- located with one taurocholate molecule by the rat Ntcp [11,43,44]. Previous reports from Na+/H+ proton exchanger [12], Na+/Ca2+ exchanger [13], proton pumps [14], sodium- sensitive receptors [15] and sodium-coupled cotransporters [16–18] indicated that negatively charged amino acids in integral membrane channels or carrier proteins are binding sites for sodium ions or other cationic electrolytes. Therefore, an alignment of nine members of the SBAT-family for negatively charged amino acids was made which revealed five conserved glutamates and aspartates. The construction of point mutations of all five conserved and two nonconserved negatively charged amino acids into their noncharged counterparts asparagine and glutamine revealed the func- tional importance of two of them for taurocholate transport. In previous studies, we had reported that SH-group reagents with wide varying lipid–water partition values reversibly blocked taurocholate uptake into isolated rat hepatocytes [19,20]. We postulated that cysteines from intra- and extramembrane domains of the Ntcp are essential for the transport function of the sodium-dependent bile acid cotransporter. Very recently a report on the human NTCP indicated that Cys266, which is located in the final extracellular loop (loop VI as predicted by the seven

Correspondence to E. Petzinger, Institute of Pharmacology and Toxicology, Justus-Liebig-University Giessen, Frankfurter Str. 107, D-35392 Giessen, Germany. Fax: + 49 641 99 38409, Tel.: + 49 641 99 38400, E-mail: ernst.petzinger@vetmed.uni-giessen.de Abbreviations: SBAT, sodium-dependent bile acid cotransporter; TM, transmembrane. (Received 19 December 2001, revised 19 December 2002, accepted 14 January 2003)

1118 D. Zahner et al. (Eur. J. Biochem. 270)

(cid:2) FEBS 2003

transmembrane domains model), is involved in taurocholate transport of the human isoform [21]. We therefore looked for the role of each of the eight cysteines of the rat Ntcp for taurocholate transport.

the cDNA was extended at the 3¢ end by the sequence GATTACAAGGATGACGACGATAAG coding for the FLAG(cid:1) peptide. Insertion of the sequence was carried out by site-directed mutagenesis using the QuikChangeTM kit from Stratagene, La Jolla, USA. The primers used for the PCR are depicted in Table 1. Here, 18 PCR-cycles were applied. Location of the insertion was verified by SeqLab Laboratories, Go¨ ttingen, Germany.

Immunofluorescence microscopy

Finally, threonines, within a protein kinase C consensus region located in the C-terminus of the Ntcp protein were analyzed with regard to their role in taurocholate transport. Such threonines might be prone to phosphorylation/ dephosphorylation reactions as it was shown that the Ntcp is a serine/threonine phosphorylated phosphoprotein which is dephosphorylated by cAMP [22]. Upon phosphorylation of serine/threonine by a protein kinase A, taurocholate transport is increased but upon phosphorylation of the Ntcp by protein kinase C, taurocholate uptake is reduced [23]. Threonines from the C-terminal consensus region were therefore converted to either tyrosines or alanines to abrogate any phosphorylation signal by PKC.

Materials and methods

Site-directed mutagenesis, cloning procedures, and DNA sequencing

X. laevis oocytes, prepared and maintained in culture as described [28], were injected with 2.5 ng cRNA coding for the wild-type and mutant Ntcp-protein, both elongated by the FLAG(cid:1) sequence. After 2 days of expression, the vitelline membrane was removed by hand and the oocytes were fixed in a solution of 80% methanol/20% dimethyl- sulfoxide. Oocytes were washed in decreasing concentra- tions of methanol in phosphate-buffered saline (NaCl/Pi, 0.9%, pH 7.4) and were incubated with the mAb M2-anti- FLAG(cid:1) (Sigma-Aldrich, Taufkirchen, Germany). After a second washing step with NaCl/Pi buffer the oocytes were fixed with 3.7% formaldehyde in NaCl/Pi and incubated with Alexa Fluor(cid:1) 488 goat anti-mouse IgG conjugate (Molecular Probes, Leiden, Netherlands). They were again washed with NaCl/Pi and embedded in Technovit 7100 (Heraeus Kulzer, Wehrheim, Germany). Sections, 5-lm thick, were cut and proteins were detected by reflective fluorescence microscopy at 488 nm (Leitz Diaplan UV Microscope, Wetzlar, Germany).

Heterologous expression of Ntcp-cRNA in X.laevis oocytes

The cDNA of the Ntcp was a kind gift of B. Hagenbuch, University Hospital, Dept Clinical Pharmacology, Zu¨ rich. Point and deletion mutants of the rat liver Ntcp cDNA clone prLNaBA [1] were generated by site-directed mutagenesis by the use of the QuikChangeTM kit from Stratagene, La Jolla, USA. The primers were selected for each mutation according to the manufacturer’s manual and were purchased from MWG, Biotech AG, Ebersberg, Germany. They are shown in Table 1. Mutants were generated by PCR using 16 cycles according to the manufacturer’s protocol in a Perkin-Elmer GenAmp cycler 2400 (Perkin Elmer, U¨ berlingen, Germany). The template DNA prLNaBA was digested with DpnI. Each mutated plasmid was transformed into Epicurian Coli XL1- Blue Supercompetent cells by heat pulse. Bacterial cells were transferred to LB-ampicillin agar plates and single colonies were isolated and further cultivated to subconfluency in LB medium. Plasmid DNA was isolated according to the Qiagen Midi kit instructions (Qiagen, Hilden, Germany). The insert of 1663 bp length of each clone was upstream and down- stream sequenced by a dye terminated method using the ABI-Prism Dye Terminator Cycle Sequencing Ready Reac- tion kit from Applied Biosystems Inc., Weiterstadt, Germany in the DNA sequencer 373A from the same company.

Alignments

An alignment of nine members of the SBAT-family, five basolateral (Ntcp rat [1], Ntcp mouse1 and 2 [9], Ntcp rabbit [24], and NTCP human [2]) and four apical (Isbt rat [6], Isbt mouse [25], Isbt rabbit [26] and ISBT human [27]) was performed, using the CLUSTAL W 1.6 program from the Baylor College of Medicine Search Launcher (Houston, USA) to identify conserved negatively charged amino acids and cysteines.

Tagging of Ntcp mutants by the FLAG(cid:1) motif

Mutated and nonmutated plasmids were linearized by PvuI (MBI Fermentas, Vilnius, Lithuania). Capped mRNA was transcribed in vitro using T7 RNA polymerase (Promega, in the presence of capping analog Madison, USA) m7G(5¢)ppp(5¢)G from Pharmacia, Freiburg, Germany. Unincorporated nucleotides were removed with a Sephadex G-50 spin column (Boehringer, Mannheim, Germany). cRNAs were recovered by ethanol precipitation and resuspended in double distilled water for oocyte injection. X. laevis oocytes were prepared and maintained in culture as described [28]. They were microinjected with 2.5 ng of Ntcp/mutant cRNA per oocyte in standard experiments. In a series of saturation experiments, 0.46–6.9 ng cRNA per oocyte were injected. For expression, oocytes were incubated for 2 days at 18 (cid:4)C in modified Barth solution. For uptake measurements, 10–15 oocytes were incubated at 25 (cid:4)C in a medium containing 5 lM [3H]taurocholate (NEN Life Science Products, Boston, MA, USA; specific activity 2–3.47 CiÆmmol)1), 10 mM Hepes/Tris pH 7.5, 2 mM KCl, 1 mM CaCl2, 1 mM MgCl2 and either 100 mM NaCl or 100 mM choline chloride in order to calculate the Ntcp- mediated sodium-dependent taurocholate uptake. Hill coefficient analysis of the sodium-coupled taurocholate uptake by wild-type Ntcp and two Ntcp mutants with mutated negatively charged amino acids (Asp115Asn and Glu257Gln) was deduced from [3H]taurocholate uptake experiments in the same Hepes/Tris buffer, however, with sodium chloride concentrations of zero, 30, 50, 100, 150 and

To determine whether the wild-type and the mutant proteins are expressed and located on the surface of the oocytes,

Site-directed mutagenesis of Ntcp (Eur. J. Biochem. 270) 1119

(cid:2) FEBS 2003

Table 1. Primers used for generating the indicated Ntcp-mutations by QuikChangeTM.

Desired mutation Primer name Sequence

Ggccaccgggccacaaacaaggcgcttagcatc

Asp24Asn

Gatgctaagcgccttgtttgtggcccggtggcc Gctctcactgggcgccaccatggaattcagc

Cys44Ala

Gctgaattccatggtggcgcccagtgagagc Catgctctcactgggctggaccatggaattcagc

Cys44Trp

gctgaattccatggtccagcccagtgagagcatg ctgggctgcaccatgcaattcagcaagatcaag

Glu47Gln

cttgatcttgctgaattgcatggtgcagcccag cacctgagcaacattcaagctctggccatcctc

Glu89Gln

gaggatggccagagcttgaatgttgctcaggtg ctggccatcctcatcgctggctgctctcccggg

Cys96Ala

cccgggagagcagccagcgatgaggatggccag ctggccatcctcatctggggctgctctcccggg

Cys96Trp

cccgggagagcagccccagatgaggatggccag catcctcatctgtggcgcctctcccggggggaac

gttccccccgggagaggcgccacagatgaggatg

Cys98Ala

catcctcatctgtggctggtctcccggggggaac gttccccccgggagaccagccacagatgaggatg

Cys98Trp

ctggccatgaaggggaacatgaacctcagcatc gatgctgaggttcatgttccccttcatggccag

Asp115Asn

catcgtgatgaccaccgcctccagcttcagtgcc ggcactgaagctggaggcggtggtcatcacgatg

Cys125Ala

catcgtgatgaccacctccagcttcagtgcc ggcactgaagctggaggtggtcatcacgatg

Cys125Del

gcaaaggcatctacaatggagaccttaaggacaagg ccttgtccttaaggtctccattgtagatgcctttgc

Asp147Asn

gttctcattcctgccaccatagggatcgtcc ggacgatccctatggtggcaggaatgagaac

Cys170Ala

catagttctcattccttggaccatagggatcgtc gacgatccctatggtccaaggaatgagaactatg

Cys170Trp

caactcaatccaagcgccagacgcaccatcagc gctgatggtgcgtctggcgcttggattgagttg

Cys250Ala

ccaactcaatccaagcagacgcaccatcagc gctgatggtgcgtctgcttggattgagttgg

Cys250Del

gctgcagacgcaccatcagcatgcaaacaggattcc ggaatcctgtttgcatgctgatggtgcgtctgcagc

Asp257Asn

ccaaaacattcaactcgcttctaccatcctcaatgtg cacattgaggatggtagaagcgagttgaatgttttgg

Cys266Ala

ggattccaaaacattcaactctctaccatcctcaatgtgacc ggtcacattgaggatggtagagagttgaatgttttggaatcc

Cys266Del

cctcaatgtgaccttcccccctcaagtcattgggcc ggcccaatgacttgaggggggaaggtcacattgagg

Asp277Asn

catcattatcttccgggcctatgagaaaatcaagcctcc ggaggcttgattttctcataggcccggaagataatgatg

Cys306Ala

catcattatcttccggtggtatgagaaaatcaagcctcc ggaggcttgattttctcataccaccggaagataatgatg

Cys306Trp

catcattatcttccggtatgagaaaatcaagcctc gaggcttgattttctcataccggaagataatgatg

Cys306Del

gcctccaaaggaccaagcaaaaattacctacaaagc

Thr317Ala

gctttgtaggtaatttttgcttggtcctttggaggc atcaagcctccaaaggaccaatacaaaattacctacaaagctgctg

Thr317Tyr

cagcagctttgtaggtaattttgtattggtcctttggaggcttgat ggaccaaacaaaaattgcctacaaagctgctgcaac

Thr320Ala

gttgcagcagctttgtaggcaatttttgtttggtcc ccaaaggaccaaacaaaaatttactacaaagctgctgcaactgagg

Thr320Tyr

cctcagttgcagcagctttgtagtaaatttttgtttggtcctttgg Ggtcagatggcaaatgattacaaggatgacgacgataagtagaatgtgaaacttcgaagc

Gcttcgaagtttcacattctacttatcgtcgtcatccttgtaatcatttgccatctgacc

FLAG(R)-insert – F – R – F – R – F – R – F – R – F – R – F – R – F – R – F – R – F – R – F – R – F – R – F – R – F – R – F – R – F – R – F – R – F – R – F – R – F – R – F – R – F – R – F – R – F – R – F – R – F – R – F – R – F – R – F – R – F – R

1120 D. Zahner et al. (Eur. J. Biochem. 270)

(cid:2) FEBS 2003

200 mM. The buffers of zero, 25, 50, and 100 mM NaCl were substituted with the corresponding choline chloride concen- tration (0/100, 25/75, 50/50, 100/0 NaCl/choline chloride). The oocyte-associated radioactivity was determined in a liquid scintillation counter (Wallac 1407, Wallac Inc., Turku, Finland).

Results

Search for conserved negatively charged amino acids, cysteines and C-terminal threonines by sequence identity

An alignment of the amino acid sequence of nine SBAT proteins, namely five basolateral Ntcp-proteins together with four apical Isbt-proteins, revealed that the following glutamates, aspartates, and cysteines in the rat Ntcp are conserved in all of the nine family members: Cys44, Cys98, Cys125, and Cys266 as well as Glu47, Asp115, Asp147, Glu257, and Glu277. The threonines Thr317 and Thr320 are only found in the rat liver Ntcp (Fig. 1).

Mutations of negatively charged amino acids residues

The predicted seven transmembrane (TM) structure of rat liver Ntcp according to [1] and all introduced mutations are depicted in Fig. 2. The organic anion transporting SBATs are cotransporters with sodium ions as the driving ion gradient. Therefore, in addition to substrate binding sites, regions for cation binding are also required. Earlier reports have indicated the importance of negatively charged amino acids for sodium-coupled substrate cotrans- port or exchange [12,13,16–18]. Mutations of all conserved and two nonconserved negatively charged amino acids to the noncharged counterparts, i.e. Asp to Asn and Glu to Gln, revealed that the aspartates Asp24 and Asp115 as well as Glu257 are required for taurocholate transport (Fig. 3).

As taurocholate transport via the carrier mutants Glu257Gln and Asp115Asn was almost nil (2 and 15% of wild-type Ntcp, respectively), tests were carried out to determine whether insufficient expression of the injected cRNA caused this lack of transport. Therefore, up to three times the amount of the cRNA compared to the standard amount was injected into oocytes, i.e. 6.9 ng instead of 2.3 ng cRNA. No improvement of taurocholate transport was observed (Fig. 5).

Tests were then performed to determine whether the absence of transport was caused by a sorting defect of these mutant proteins. For this reason, the FLAG(cid:1) motif was

The negatively charged Glu257 is exposed in an extracellular loop of the Ntcp and could represent the sodium ion sensor of sodium-coupled taurocholate trans- port via Ntcp. In order to find out whether and to what extent this amino acid affects sodium ion dependency of taurocholate uptake, transport studies were performed in the presence of varying amounts of extracellular sodium chloride and Hill analysis was applied (Fig. 4). For comparison the transport-negative Asp115Asn mutant was investigated in the same manner. As a result, the negative charge in Glu257 is an essential prerequisite for sodium-dependent taurocholate uptake. The Hill coeffi- cient of this cotransport by wild-type Ntcp is about 2–2.59 [43,44] but dropped to 0.32 (measured at 25–200 mM NaCl) if Glu257 was converted to Gln (Fig. 4). Increase of the sodium gradient by applying concentrations up to 200 mM NaCl to the outside did not alter the abolished transport of taurocholate significantly, although at 200 mM NaCl taurocholate transport slightly increased. In contrast, significant sodium cooperativity was found, however, at a much lower level, in Ntcp mutant Asp115Asn. The Hill number was 1.15 for the Asp115Asn mutant (Fig. 4) which corresponds to a sodium stoichiometry of one sodium ion per taurocholate molecule.

Fig. 1. Alignment of nine members of the SBAT family. Ntcp mouse1, mml1; Ntcp mouse2, mml2; Ntcp rat, rnl; Ntcp rabbit, oclm; NTCP human, hsl; Isbt rat, rni; Isbt mouse, mmi; Isbt rabbit, oci; Isbt human, hsi; conserved cysteines, c; conserved acidic amino acids, a.

Site-directed mutagenesis of Ntcp (Eur. J. Biochem. 270) 1121

(cid:2) FEBS 2003

one at the beginning of the C-terminal tail was altered by site-directed mutagenesis. The three cysteines from the nontransmembrane domains and the one from the C-terminus were substituted by alanine or were omitted to attain deletion mutants. All deletion mutants, namely Cys125Del, Cys250Del, Cys266Del, and Cys306Del were transport-negative (Fig. 7), indicating that each cysteine per se is required. If their alanine counterparts were expressed in X. laevis oocytes, all except one showed restored transport activity. Only the Cys266Ala mutant remained transport-negative. We conclude that Cys266 is the only cysteine of the rat liver Ntcp which appears to be directly involved in taurocholate uptake into the oocytes.

Fig. 2. Topology model of rat Ntcp based on hydropathy analysis of the amino acid sequence (according to [1]). Transmembrane domains are symbolized as blocks of amino acids. Mutated amino acids are highlighted in gray. The resulting mutants are shown in boxes, with deletions indicated (del).

To show whether or not mutant Cys266Ala was present in the cell surface of X. laevis oocytes, the corres- ponding cDNA clone was also tagged by the FLAG(cid:1) motif and cRNA from this construct was again injected into oocytes. Immunofluorescence pictures confirm that the mutant Cys266Ala protein is present in the cell membrane in a similar amount as the wild-type Ntcp protein (Fig. 6).

insertion of

cloned into each transport-negative mutant clone. With this technique, the mutants Asp115Asn and Glu257Gln within the cell membrane of X. laevis oocytes was observed by use of antibodies raised against the FLAG(cid:1) peptide, and applied to permeabilized oocytes (Fig. 6). An exception was observed, the transport-negative mutant Asp24Asn, which did not appear in the membrane (Fig. 6), indicating that Asp24 from the extracellular N-terminus is not essential for transport but for appropriate cell sorting of the Ntcp protein.

Cysteine mutants

the eight cysteines,

Each of four in transmembrane domains, three in cytoplasmic or extracellular loops, and

Eight further cysteine mutations were generated regard- ing the four intramembrane cysteines (Fig. 7). Exchanges by alanine or tryptophane were generated. With the exception of Cys306 each tryptophane mutant was either transport negative (Cys98Trp, Cys170Trp, Cys96Trp) or showed decreased uptake (Cys44Trp). However, if these intramem- brane cysteines were substituted by alanines, taurocholate transport was fully regained. This indicates that none of the transmembrane cysteines appears to be directly involved in the transport process. An exception was the tryptophane substitution of Cys306. This Cys306Trp mutant transported taurocholate more effectively (more than 1.5-fold) than wild-type Ntcp. Cys306, however, is located at the beginning of the C-terminal tail of Ntcp (Fig. 7).

Fig. 3. Mutations of negatively charged conserved amino acids alters taurocholate transport via Ntcp. Uptake of [3H]taurocholate by X. laevis oocytes two days after microinjection of 50 nL containing 2.5 ng cRNA which was transcribed from wild-type or mutant Ntcp clones. Uptake is given in percentage of wild-type uptake after 30 min of exposure to 5 lM [3H]taurocholate.

1122 D. Zahner et al. (Eur. J. Biochem. 270)

(cid:2) FEBS 2003

Fig. 4. Sodium dependency of taurocholate uptake by Ntcp mutants. Uptake of 5 lM [3H]taurocholate was measured during 30 min after injection of 2.5 ng cRNA subscribed from wild-type and mutant Ntcp-clones into oocytes. The oocytes were incubated in the presence of increasing sodium chloride con- centrations. The results obtained by the clones Asp115Asn and Glu257Gln are also given in a Hill plot.

Threonine mutants

The threonines Thr317 and Thr320 are located within the protein kinase C consensus regions LysXXThrLys and LysXThrXLys of the Ntcp [29]. Therefore, both threonines were substituted by either tyrosine or alanine. None of these mutations significantly altered taurocholate uptake. The transport rate of each mutant was between 80 and 100% of the wild-type Ntcp (Fig. 6).

Fig. 5. The relationship between the amount of injected cRNA and taurocholate uptake into X. laevis oocytes via Ntcp and Ntcp mutants. Uptake of [3H]taurocholate by cRNA-injected oocytes after exposure to 5 lM [3H]taurocho- late for 30 min. The amount of cRNA of transport-negative mutants was increased 14-fold; the standard amount of cRNA which was injected for comparison of transport by mutated vs. wild-type Ntcp was 2.5 ngÆ oocyte)1.

Discussion

Hepatobiliary transport of the major bile acid taurocholate in humans and rats begins by uptake across the baso- lateral membrane of hepatocytes via the high affinity,

Fig. 6. Detection of the presence of Ntcp proteins in the cell membrane of X. laevis oocytes by the reporter FLAG(cid:1) motif. The FLAG(cid:1) encoded amino acid sequence was detected by sandwich immuno- fluorescence labeling with monoclonal anti-FLAG(cid:1) Ig and subse- quent labeling with Alexa Fluor(cid:1) 488 goat anti-mouse IgG conjugate in permeabilized and fixed oocytes after two days of cRNA expression. With the exception of Asp24Asn mutated Ntcp, each mutated protein was detected in the cell membrane of oocytes. Negative control was oocytes that were injected with water. From top left to right: upper, Asp24Asn; Cys266Del; middle, Glu257Gln; Cys266Ala; lower, Asp115Asn; water-injected oocyte (negative control); large picture, wild-type Ntcp (positive control).

Site-directed mutagenesis of Ntcp (Eur. J. Biochem. 270) 1123

(cid:2) FEBS 2003

bile salt export pump BSEP, an ATP-driven ABC-cassette protein related to mdr1 [30,31]. Disturbances of the hepatocellular part of the enterohepatic circulation of bile

sodium-dependent and liver-specific basolateral bile acid carriers NTCP (humans) and Ntcp (rats). Subsequent to uptake the bile acid is released into the bile canaliculus by the

1124 D. Zahner et al. (Eur. J. Biochem. 270)

(cid:2) FEBS 2003

is electrogenic [40,46]. This 2 : 1 stoichiometry was altered in the mutant Glu257Gln as revealed by Hill analysis. A Hill number of almost zero was calculated indicating that this mutant is unable to translocate sodium ions together with taurocholate. Therefore we assume that Glu257 is the extracellular sodium sensor for sodium taurocholate co- transport. As taurocholate uptake with this Ntcp protein was almost nil (residual 2% transport compared with wild- type Ntcp), the long established importance of the sodium ion for the translocation step of monoanionic bile acids was reaffirmed. The cationic sodium ions are likely to interact with negatively charged amino acid residues at the outer surface of the Ntcp protein, but then need to be translocated through pore-forming transmembrane helices to the cyto- plasmic regions of the protein. It has been shown that extracellular loops, containing charged amino acids, can slide between TM domains into the membrane, forming P-loops [40]. P-loops allow the introduction of charged molecules into inner parts of the cell membrane from where these can be overtaken by further binding sites originating from the cytoplasmic region of the protein. It is tempting to

acids causes intrahepatic cholestasis [32–35]. Whereas nat- urally occurring mutations in the BSEP have been described, causing the rare Byler syndrome in children [36], naturally occuring mutations of the NTCP-gene locus have not yet been observed. Our study with the rat Ntcp indicates, however, that several amino acids may be essential for hepatocellular taurocholate uptake because mutations in these amino acids caused lack of transport; the amino acids in question are Asp115, Glu257, and Cys266. All of these are conserved in SBATs and are found also in the human NTCP protein. Functional mutations of these amino acids in the human NTCP gene locus would cause hypercholanaemia but would result in low intrahepatic bile salt levels and therefore little if any hepatocellular injury. This syndrome has been already described in two children, however, without mutations of the NTCP gene and therefore remained unexplained [37]. The clinical picture of nonfunc- tional NTCP carriers would differ from patients with cholestasis where blockade of bile acid secretion at the canalicula pole of hepatocytes leads to elevated intracellular (and extracellular) bile acid concentration and therefore causes severe liver injury. It should be noted that in such cases of cholestasis, Ntcp expression as a protecting mechanism decreases dramatically [38], but in the cases of benign hypercholanemia, Ntcp expression was normal [37]. In the latter syndrome, taurocholate uptake and also bile acid-dependent bile formation is not expected to cease as bile acid uptake by liver-type organic anion transporting poly- peptides OATP 8 and OATP-C continues.

All mutations from transport-negative mutants were located in loop structures of the rat Ntcp; two of them, Glu257 and Cys266, were located in loop VI, the final extracellular loop (Fig. 2). This region also appears to have key properties for taurocholate transport in other SBATs, as it was already reported that a naturally occuring point mutation of Thr262 in the human intestinal Na+/bile acid cotransporter ISBT abolished reabsorption of bile acids and caused primary bile acid malabsorption in patients [38]. This conserved threonine is located in loop VI of ISBT and Ntcp and is next to Glu257 in the rat Ntcp, which we report here, is also required for hepatic taurocholate transport.

Fig. 7. Taurocholate uptake into X. laevis oocytes after injection of cRNA of wild-type and mutated Ntcp. Uptake of [3H]taurocholate in X. laevis oocytes which were injected with 2.5 ng wild-type or mutant cRNA and incu- bated for two days as described in Materials and methods. The diagram depicts relative uptake in percentage of uptake by wild-type Ntcp (100%) after 30 min incubation with 5 lM [3H]taurocholate.

The negatively charged Glu257 is probably a binding site for extracellular sodium ions. The driving force for substrate transport via all SBATs is the sodium gradient across the cell membrane. Two sodium ions are supposed to be translo- cated together with one bile acid molecule via ileal Na+-bile acid cotransporters such as human ASBT [46] or the rat liver Ntcp [11]. Therefore, sodium-driven taurocholate transport

Fig. 8. P-loop model of rat Ntcp depicting Glu257 and Asp115 as putative binding sites for sodium ions. The extracellular loop 6 between TM VI and VII contains a sodium substrate-binding region for tauro- cholate during sodium ion–taurocholate cotransport.

Site-directed mutagenesis of Ntcp (Eur. J. Biochem. 270) 1125

(cid:2) FEBS 2003

which fixes the protein to the plasma membrane. If this is true for the Ntcp protein, this could explain why the hydrophobic amino acid tryptophane fully substituted Cys306 only in that protein position and why in contrast to all other tryptophane substitutions, taurocholate trans- port was not abolished but was even enhanced by this mutation.

hypothesize that if such a translocation mechanism is active in the Na+/bile acid cotransporter Ntcp, extracellular Glu257 and cytoplasmic Asp115 may constitute an appro- priate pair of binding sites for sodium ions allowing ion translocation across the cell membrane. The cytoplasmic Asp115 may detract the two sodium ions delivered by extracellular Glu257 via P-loop formation (Fig. 8). Consis- tent with this suggested model is the observation that charge modification in the putative cytoplasmic sodium sensor Asp115 through its conversion into Asn decreased sodium- dependent taurocholate uptake to 15% of wild-type Ntcp, probably because the sodium stoichiometry of 2 : 1 declined to 1 : 1 (Fig. 4).

Cys306 marks the border to a 56 amino acid tail which stretches to the end of the final amino acid, Asn362, of the Ntcp C-terminus. It was already shown that this C-terminal tail is not required for transport properties [41]. A truncated rat liver Ntcp protein lacking all amino acids beyond Cys306 transported taurocholate with a Km identical to that of wild-type Ntcp. Similarly, a trans- port-positive Ntcp splicing variant which was shortened by 45 amino acids from the end of the C-terminus was cloned from mice [9]. Thus the C-terminal tail appears to be unnecessary for taurocholate uptake. However, it was required for appropriate basolateral sorting of the protein, because mutations of Tyr307 (following next to Cys306) and Tyr321 (following next to Thr320) accumulated within the cytosol but were absent from the cell membrane [42]. Apart from sorting signal motifs, other regulatory func- tions might be phosphorylation/dephosphorylation reac- tions. For this reason, two threonines, Thr317 and Thr320, within a protein kinase C-consensus region were mutated to alanines but also to tyrosines. None of these mutations showed any effects on taurocholate transport into X. laevis oocytes. Our results would not disprove such phosphory- lation reactions being present in mammalian cells, but the localization of receptive serine/threonine residues for a particular protein kinase are unlikely to be expected within that protein kinase C-consensus region, as even the alanine substitutions were without any effect on taurocholate transport.

Whereas negatively charged residues are not suspected to interact with the anionic organic substrates of SBATs cysteines are. It has already been shown that Cys266 is essential for taurocholate transport by the human NTCP protein [21]. Here we report that the same amino acid in the corresponding position is also required for taurocho- late transport by the rat Ntcp. This cysteine appears to be directly involved in taurocholate transport as it is the only one which remained transport-negative when substituted by alanine. However, this is in contrast to the report by Halle´ n et al. 2000 [21], showing that mutant Cys266Ala of the human NTCP still transported taurocholate without any marked change in Km and Vmax. These authors obtained evidence for a separate function of that cysteine by indirect means with SH-group reagents. The reason for this discrepancy is unclear, but might be due to the different expression systems used for detection (Xenopus oocytes in this report vs. HEK293 cells in Halle´ n’s report) or to different local interactions of this cysteine within loop VI of the different Ntcps. It should be noted that loop VI of the human NTCP contains four cysteines whereas in the rat Ntcp only two cysteines (Cys250, Cys266) are present.

Acknowledgments

The authors wish to acknowledge the receipt of the Ntcp-containing plasmid prLNaBA from Dr Bruno Hagenbuch, Zurich. Support of this project was given by Drs Frank and Marita Langewische who helped to initiate the study by constructing cysteine mutants. Technical help was provided by Mrs Elisabeth Ju¨ ngst-Carter and Steffi Weghenkel. Dr Bruce Boschek has provided critical liguistical advice in preparing the manuscript.

References

Other cysteines (seven out of eight) of the Ntcp may have indirect effects on taurocholate transport. Such effects were analyzed by cysteine deletion mutants and trypto- phane substitutions. Indirect effects could be space-holding properties of these cysteines tested by deletion mutants and lipophilic binding properties other than by SH-groups the tested by tryptophane. The deletion mutants of cysteines from loops III and VI (Cys125, Cys250 and Cys266) (Fig. 2) were transport-negative. Replacement by alanine, however, restored uptake in the case of Cys125 and Cys250. Therefore, these cysteines appear to have space holder functions for the loops. With respect to the Cys250Ala mutant of the rat Ntcp, our finding is in full agreement with the results observed with the human NTCP, where taurocholate transport by the Cys250Ala mutant was also not altered [21]. Because seven out of eight cysteine/alanine substitutions were transport-positive (with the exception of Cys266Ala), we conclude that no disulfide bonding between cysteines within a monomeric Ntcp protein has occurred.

the

Among

cysteine/tryptophane

1. Hagenbuch, B., Stieger, B., Foguet, M., Lu¨ bbert, H. & Meier, P.J. (1991) Functional expression cloning and characterisation of the hepatocyte Na+/bile acid cotransport system. Proc. Natl Acad. Sci. USA 88, 10629–10633.

2. Hagenbuch, B. & Meier, P.J. (1994) Molecular cloning, chromo- somal localization and functional characterisation of a human liver Na+/bile acid cotransporter. J. Clin. Invest. 93, 1326–1331. 3. Hagenbuch, B. & Meier, P.J. (1996) Sinusoidal (basolateral) bile salt uptake systems of hepatocytes. Semin. Liver Dis. 16, 129–136. 4. Wong, M.H., Oelkers, P., Craddock, A.L. & Dawson, P.A. (1994) Expression cloning and characterization of the hamster ileal sodium-dependent bile acid transporter. J. Biol. Chem. 269, 1340– 1347.

substitutions, Cys306Trp was exceptional in that taurocholate transport was not abolished but was even enhanced to 150% of wild- type transport. Cys306 is located at the border of TM7 to the cytoplasmic C-terminus. Cysteines in that position might serve as an anchor for a palmitoyl/isoprenyl residue

5. Wong, M.H., Oelkers, P. & Dawson, P.A. (1995) Identification of a mutation in the ileal sodium-dependent bile acid transporter

1126 D. Zahner et al. (Eur. J. Biochem. 270)

(cid:2) FEBS 2003

gene that abolishes transport activity. J. Biol. Chem. 270, 27228– 27234. hepatic Na+/taurocholate cotransport by cyclic AMP. J. Biol. Chem. 268, 17743–17741.

24. Stengelin, S., Becker, W., Maier, M., Noll, R. & Kramer, W. (1998) Rabbit cDNA encoding hepatic sodium-dependent bile acid transporter. GeneBank accession number AJ131361. 6. Shneider, W.L., Dawson, P.A., Christie, D.-M., Hardikar, W., Wong, M.H. & Suchy, F.J. (1995) Cloning and molecular char- acterisation of the ontogeny of rat ileal sodium dependent bile acid transporter. J. Clin. Invest. 95, 745–754.

25. Saeki, T., Motoba, K., Furukawa, H., Kirifiji, K., Kanamoto, R. & Iwami, K. (1999) Characterisation, cDNA cloning, and func- tional expression of mouse ileal sodium-dependent bile acid transporter. J. Biochem. (Tokyo) 125, 846–851. 7. Lazarides, K.N., Pham, L., Tietz, P., Marinelli, P.C., Levine, S., Dawson, P.A. & LaRusso, N.F. (1997) Rat cholangiocytes absorb bile acids at their apical domain via the ileal sodium-dependent bile acid transporter. J. Clin. Invest. 100, 2714–2721. (1997) Ileal

26. Stengelin, S., Apel, S., Becker, W., Maier, M., Rosenberger, J., sodium/bile acid Wess, G. & Kramer, W. cotransporter; direct submission, GeneBank accession number Q28727. 8. Lazarides, K.N., Tietz, P., Wu, T., Kip, S., Dawson, P.A. & LaRusso, N.F. (2000) Alternatic splicing of the rat sodium/bile acid transporter changes ist cellular localization and transport properties. Proc. Natl Acad. Sci. USA 97, 11092–11097.

9. Cattori, V., Eckhardt, U. & Hagenbuch, B. (1999) Molecular two alternatively cloning and functional characterization of spliced Ntcp isoforms from mouse liver. Biochim. Biophys. Acta 1445, 154–159.

10. Hagenbuch, B. (1997) Molecular properties of hepatic uptake systems for bile acids and organic anions. J. Membrane Biol. 160, 1–8. 27. Craddock, A.L., Love, M.W., Daniel, R.W., Kirby, L.C., Walters, H.C., Wong, M.H. & Dawson, P.A. (1998) Expression and transport properties of the human ileal and renal sodium-depen- dent bile acid transporter. Am. J. Physiol. 274, G157–G169. 28. Eckhardt, U., Horz, J.A., Petzinger, E., Stu¨ ber, W., Reers, M., Dickneite, G., Daniel, H., Wagener, M., Hagenbuch, B., Stieger, B. & Meier, P.J. (1996) The peptide-based thrombin inhibitor CRC 220 is a new substrate of the basolateral rat liver organic anion-transporting polypeptide. Hepatology 24, 380–384.

11. Weinman, S.A. & Weeks, R.P. (1993) Electrogenicity of Na+- coupled bile salt transport in isolated rat hepatocytes. Am. J. Physiol. 265, G73–G80. 29. Pearson, R.B. & Kemp, B.E. (1991) Proteinkinase phosphoryla- tion site sequences and consensus specificity motifs. Tabulations. Methods Enzymol. 200, 62–82.

30. Gerloff, T., Stieger, B., Hagenbuch, B., Madon, J., Landmann, L., Roth, J., Hofmann, A.F. & Meier, P.J. (1998) The sister of P-glycoprotein represents the canalicular bile salt export pump of mammalian liver. J. Biol. Chem. 273, 10046–10050. 31. Bahar, R. & Stolz, A. (1999) Bile acid transport. Gastroenterol. Clin. North Am. 28, 27–58. 32. Petzinger, E. (1994) Transport of organic anions in the liver. Rev. 12. Murtazina, R., Booth, B.J., Bullis, B.L., Singh, D.N. & Fliegel, L. (2001) Functional analysis of polar amino acids residues in membrane associated regions of the NHE1 isoform of the mam- malian Na+/H+ exchanger. Eur. J. Biochem. 268, 4674–4685. 13. Nicoll, D.A., Hryshko, L.V., Matsuoka, S., Frank, J.S. & Phi- lipson, K.D. (1996) Mutation of amino acid residues in the putative transmembrane segments of the cardiac sarcolemmal Na+–Ca++ exchanger. J. Biol. Chem. 23, 13385–13391. Physiol. Biochem. Pharmacol. 123, 47–211.

14. Lanyi, J.K. (1997) Mechanism of ion transport across membranes. Bacteriorhodopsin as a prototype for proton pumps. J. Biol. Chem. 272, 31209–31212.

33. Eckhardt, U., Schroeder, A., Stieger, B., Ho¨ chli, M., Landmann, L., Tynes, R., Meier, P.J. & Hagenbuch, B. (1999) Poly- specific uptake of the hepatic organic anion transporter Oatp1 in stably transfected CHO cells. Am. J. Physiol. 276, G1037– G1042. 34. Trauner, M., Meier, P.J. & Boyer, J.L. (1998) Molecular patho- genesis of cholestasis. N. Engl. J. Med. 339, 1217–1227. 35. Kullak-Ublick, G.A., Beuers, U. & Paumgartner, G. (2000) 15. Martin, S., Botto, J.M., Vincent, J.P. & Mazella, J. (1999) Pivotal role of an aspartate residue in sodium sensitivity and coupling to G proteins ot neurotensin receptors. Mol. Pharmacol. 55, 210–215. 16. Quick, M. & Jung, H. (1997) Aspartate 55 in the Na+/proline permease of Escherichia coli is essential for Na+-coupled proline uptake. Biochemistry 36, 4631–4636. Hepatobiliary transport. J. Hepatol. 32 (Suppl. 1), 3–18.

17. Jung, H. (2001) Towards the molecular mechanism of Na+/ solute symport in prokaryotes. Biochim. Biophys. Acta 1505, 131–143.

36. Strautnieks, S.S., Kagalwalla, A.F., Tanner, M.S., Knisely, A.S., Bull, L., Freimer, N., Kocoshis, S.H., Gardiner, R.M. & Thompson, R.J. (1997) Identification of a locus for progressive familial intrahepatic cholestasis PFIC2 on chromosome 2q24. Am. J. Hum. Genet. 61, 630–632. 18. Poolman, B., Knol, J., van der Does, C., Henderson, P.J.F., Liang, W.-J., Leblanc, G., Pourcher, T. & Mus-Veteau, I. (1996) Cation and sugar selectivity determinats in novel family of trans- port proteins. Molec. Microbiol. 19, 911–922.

expression in two unusual 19. Blumrich, M. & Petzinger, E. (1990) Membrane transport of conjugated and unconjugated bile acids into hepatocytes is susceptible to SH-blocking reagents. Biochim. Biophys. Acta 1029, 1–12.

37. Shneider, B.L., Fox, V.L., Schwarz, K.B., Watson, C.L., Anan- thanarayanan, M., Thevananther, S., Christie, D.M., Hardikar, W., Setchell, K.D.R., Mieli-Vergani, G., Suchy, F.J. & Mowat, A.P. (1997) Hepatic basolateral sodium-dependent bile acid transporter cases of hyper- cholanemia and in extrahepatic biliary atresia. Hepatology 25, 1176–1183. 20. Blumrich, M. & Petzinger, E. (1993) Two distinct types of SH-groups are necessary for bumetanide and bile acid uptake into isolated rat hepatocytes. Biochim. Biophys. Acta 1149, 278–284.

21. Halle´ n, S., Fryklund, J. & Sachs, G. (2000) Inhibition of the human sodium/bile acid cotransporters by side-specific methane- thiosulfonate sulfhydryl reagents: substrate-controlled accessi- bility of site of inactivation. Biochemistry 39, 6743–6750.

38. Gartung, C., Ananthanarayanan, M., Rahman, M., Shuele, S., Nundy, S., Soroka, C. & Stolz, A. (1996) Down-regulation of expression and function of the rat liver Na/bile acid cotransporter in extrahepatic cholestasis. Gastroeneteology 110, 199–209. 39. Oelkers, P., Kirby, L.C., Heubi, J.E. & Dawson, P.A. (1997) Primary bile acid malabsorption caused by mutations in the ileal sodium-dependent bile acid transporter gene (SLC10A2). J. Clin. Invest. 99, 1880–1887.

22. Mukhopadhayay, S., Ananthanarayanan, M., Stieger, B., Meier, P.J., Suchy, F.J. & Anwer, M.S. (1998) Sodium taurocholate cotransporting polypeptide is a serine, threonine phosphorprotein and is dephosphorylated by cyclic AMP. Hepatology 28, 1629– 1636.

40. Lidofsky, S.D., Fitz, J.G., Weisiger, R.A. & Scharschmidt, B.F. (1993) Hepatic taurocholate uptake is electrogenic and influenced by transmembrane potential difference. Am. J. Physiol. 264 (Gastrointest. Liver Physiol. 27), G478–G485. 23. Gru¨ ne, S., Engelking, L.R. & Anwer, M.S. (1993) Role of intracellular calcium and protein kinases in the activation of

Site-directed mutagenesis of Ntcp (Eur. J. Biochem. 270) 1127

(cid:2) FEBS 2003

44. Weinman, S.A. (1997) Electrogenicity of Na+-coupled bile acid transporters. Yale J. Biol. Med. 70, 331–340. 41. Kirsch, G.E., Pascual, J.M. & Shieh, C.C. (1995) Functional role of a conserved aspartate in the external mouth of voltage-gated potassium channels. Biophys. J. 68, 1804–1813.

45. Edmondson, J.W., Miller, B.A. & Lumeng, L. (1985) Effect of glucagon on hepatic taurocholate uptake: relationship to mem- brane potential. Am. J. Physiol. 249 (Gastrointest. Liver Physiol. 12), G427–G433. 42. Sun, A.Q., Arrese, M.A., Zeng, L., Swaby, I., Zhou, M.-M. & suchy, F.J. (2001) The rat liver Na+/bile acid cotransporter. Importance of the cytoplasmic tail to function and plasma membrane targeting. J. Biol. Chem. 276, 6825–6833.

46. Weinman, S.A., Carruth, M.W. & Dawson, P.A. (1998) Bile acid uptake via human apical sodium-bile acid cotransporter is elec- trogenic. J. Biol. Chem. 273, 34691–34695.

43. Sun, A.Q., Ananthanarayanan, M., Soroka, C.J., Thevananther, S., Shneider, B.L. & Suchy, F.J. (1998) Sorting of rat liver ileal sodium-dependent bile acid transporters in poilarized epithelial cells. Am. J. Physiol. 275, G1045–G1055.