Isolation and characterization of a UDP–glucuronosyltransferase(UGT1A01) cloned from female rhesus monkey

Brian Dean,a,d,* Steve Chang,a Jenny Stevens,b Paul E. Thomas,d and Christopher Kingc

a Department of Drug Metabolism, Merck & Co., Inc., Rahway, NJ 07065, USAb Department of Quality Engineering, Merck & Co., Inc., Rahway, NJ 07065 USAc Department of Drug Metabolism, Merck & Co., Inc., San Diego, California USA

d Joint Graduate Program in Toxicology, Rutgers, The State University of New Jersey/University of Medicine

and Dentistry of New Jersey—Robert Wood Johnson Medical School, Piscataway, New Jersey, USA

Received 22 February 2002, and in revised form 29 March 2002

Abstract

An isoform (rhesus UGT1A01) orthologus to the human UGT1A1 was cloned and sequenced from female rhesus monkey liver

cDNA using primers designed from the human nucleotide sequences. Open reading frame analysis of the PCR-generated product

encodes a 533-amino acid protein with a proposed 27-residue signal peptide. Nucleotide sequence comparison of rhesus UGT1A01

to other rhesus UGT1A isoforms detected a single-transition mutation at nucleotide 1520 (T! C), resulting in a neutral F to S

substitution at position 507. Rhesus UGT1A01 was greater than 99 and 95% identical to cynomolgus UGT1A01 and human

UGT1A1, respectively. The rhesus UGT1A01 was expressed in HK-293 cells for functional analysis. Catalytic activity of UGT1A01

was determined with 7-hydroxy-4-(trifluoromethyl)-coumarin and more specific human UGT1A1 substrates (1-naphthol, b-estra-diol, 17a-ethinylestradiol, and bilirubin). Expression of UGT1A01 protein was also detected by a Western blot utilizing a polyclonalantibody developed against the human UGT1A family. � 2002 Elsevier Science (USA). All rights reserved.

Keywords: UDP–glucuronosyltrasferase; UGT; UGT1A1; UGT1A01; Rhesus; Monkey; Glucuronidation; Liver; Bilirubin; Estradiol

UDP–glucuronosyltransferases (UGTs)1 are one ofthe several major classes of xenobiotic metabolizingenzymes in addition to the cytochrome P450s, flavinmonooxygenases, glutathione transferases, and sul-fotransferases. UGTs, classified as members of the phaseII metabolizing enzymes involved in conjugation reac-tions, catalyze the transfer of glucuronic acid from aUDPGA cofactor to a nucleophilic substrate or ‘‘agly-cone’’ to form O-, N-, S-, or C-linked b-glycosides or‘‘glucuronides.’’ Several UGTs from rat [1,2], cyno-molgus monkey (Macaca fascicularis) [3–5], and humans[6–12] have been cloned, expressed, and characterized.These enzymes are particularly useful in distinguishing

species differences in safety or toxicity studies due toalterations in enzyme efficiency, specificity, or tissuedistribution and further provide invaluable tools forearly in vitro screening assays.Of the phase II enzymes, those of the UGT family

play a major role in the metabolism of both parent drugs[13] and metabolites. The importance of these metabolicenzymes is due, in part, to their high capacity, broadsubstrate specificity [14,15], and distribution in hepaticand extrahepatic tissues [16–19]. The formation of glu-curonides results in a hydrophilic product, often elimi-nated in the bile or urine with greater efficiency than theaglycone, possibly due in part to anionic transporterslocalized to the bile canaliculi [20] or kidney tubules [21].Once eliminated from the liver by way of the bile, glu-curonides are susceptible to b-glucuronidase activitypresent in the intestinal microflora, possibly leading toenterohepatic recirculation [22].UGTs are divided into two major families based on

amino acid sequence identity: the UGT1 family and

Archives of Biochemistry and Biophysics 402 (2002) 289–295

www.academicpress.com

ABB

* Corresponding author. Fax: +1-732-594-1416.

E-mail address: brian_dean@merck.com (B. Dean).1 Abbreviations used: UGTs, UDP–glucuronosyltransferases; NHP,

nonhuman primates; DMEM, Dulbecco’s modified Eagle’s medium;

HTFMC, 7-hydroxy-4-(trifluoromethyl)-coumarin; E2, b-estradiol;EE, 17a-ethinylestradiol; LSC, liquid scintillation counting.

0003-9861/02/$ - see front matter � 2002 Elsevier Science (USA). All rights reserved.

PII: S0003 -9861 (02 )00084-X

the UGT2 family. Members of the UGT1 family sharemore than 50% identity to one another and are derivedfrom a complex gene [23]. The individual UGT1 familymembers are produced from the precursor mRNA fol-lowing alternative splicing which joins a unique firstexon to four constant downstream exons. The variablefirst exon is hypothesized to provide substrate specificityto the enzymes, while the conserved exons are believed[2–5] to encode the UDPGA binding domain in additionto the membrane-anchoring stop–transfer sequence.Members of the UGT2 family, in contrast to those ofUGT1, are derived from individual genes.As with many other drug-metabolizing enzymes, the

UGTs are found to be highly variable in expression andmay be induced following administration of certainchemicals [24,25]. Deficiencies in UGT activity also havebeen implicated in bilirubin neurotoxicity [26,27]. Hu-man UGT1A1 is the isoform mainly responsible forbilirubin glucuronidation to the mono- and diglucuro-nide conjugates which aid their elimination from thebody. Crigler–Najjar type I [28] and Gilbert’s syndrome[29], both of which are manifested as hyperbilirubin-emia, are due to absent or decreased human UGT1A1activity, respectively.The rhesus monkey (Macaca mulatta) and other

nonhuman primates (NHP) constitute an important andoften required model for pharmaceutical, medical, andpsychological research due in fact to their genetic simi-larity to humans. The rhesus is of primary interest todrug metabolism scientists because it is the NHP ofchoice by many pharmaceuticals in the United States toemulate both human xenobiotic metabolism and humanpotential toxicity. All of the monkey UGTs to date havebeen isolated from the cynomolgus monkey. The goalsof this study were to discover and sequence an ortho-logus form of human UGT1A1 from rhesus monkeyliver tissue and to characterize that isoform with respectto substrate specificity and kinetic parameters to deter-mine any species differences or unique properties.

Materials and methods

Materials. RNeasy, PolyFect transfection reagent,and other molecular biology products were obtainedthrough Qiagen (Valencia, CA). M-MLV ReverseTranscriptase (RNase H minus) and the pGEM-T Easyvector were purchased from Promega (Madison, WI).The E-gels, tris–acetate NuPAGE gels, Dulbecco’smodified eagle’s media (DMEM), heat-inactivated fetalbovine serum, Hepes, geneticin (G418), and thepcDNA3.1 vector were obtained through Invitrogen(Carlsbad, CA). AmpliTAQ Gold was purchasedthrough Applied Biosystems (Foster City, CA). UD-PGA, alamethicin, and D-saccharic acid 1,4-lactone wereobtained from Sigma (St. Louis, MO). Biciuchoninic

acid protein assay kit was obtained from Pierce (Rock-ford, IL). Primers were synthesized by Sigma Genosys(The Woodlands, TX). All double-stranded DNA se-quencing was performed by ACGT (Northbrook, IL).Signal peptide determination was performed by theSignal P server (http://www.cbs.dtu.dk/services/SignalP/)[30]. Antibodies and human UGT standards were pur-chased through Gentest (Woburn, MS). The substrates7-hydroxy-4-(trifluoromethyl)-coumarin (HTFMC), 1-naphthol, b-estradiol (E2), 17a-ethinylestradiol (EE),and bilirubin were obtained from Sigma. RadiolabeledUDPGA and EE were purchased from NEN (Boston,MA).

Animals. Adult female rhesus monkey tissues (e.g.,liver, intestine, kidney, and brain) were obtained atMerck (Rahway, NJ). All procedures were approved byMerck’s Institutional Animal Care and Use Committeeaccording to regulatory guidelines and rules. All tissueswere harvested, placed in aluminium foil, and immersedin liquid nitrogen immediately following excision to re-duce the risk of cellular disruption and mRNA degra-dation. Samples were stored at )80 �C until use.

RNA Isolation and cDNA preparation. Total RNAwas isolated from tissues using an RNeasy preparationkit and stored at )80 �C. The cDNA was prepared byM-MLV reverse transcriptase (RNase Hus) according tomanufacturers recommendations and stored at )20 �Cuntil use.

Cloning and expression of UGT1A01. Amplification ofUGT1A01 was performed using the sense (50-TTTCTAAGCTTAGGAGCAAAGGCGCCATGGCTGTG-30) and antisense (50-CTCATCTCGAGCACTTCTCAATGGGTCTTGGATTTG-30) primers to the human UGT1A1 sequence. PCR conditions were set for TAQactivation at 95 �C for 5min, followed by 40 cycles at94 �C for 1min, 70 �C for 1min, and 75 �C for 2min anda final extension at 75 �C for 10min. All PCR was per-formed on an Eppendorf Mastercycler Personal andfollowed by electrophoretic separation on 1.2% agaroseE-Gels. Fragments of correct length were excised fromthe gel, purified with a QIAquick gel extraction kit, andligated into the pGEM-T Easy Vector System II forcolor screening of recombinant clones (JM 109 compe-tent cells—Escherichia coli). Positive (white) colonieswere incubated overnight in superbroth and vectorspurified with a QIAprep Spin Miniprep kit. Plasmidswere sequenced and checked for proper product inser-tion by restriction analysis, and the coding region wasligated into the mammalian vector (pcDNA3.1(+)) fortransfection and stable expression into the HK293 cellline. HK293 cells were grown in selective DMEM con-taining 10% heat-inactivated fetal bovine serum, 10mMHepes, and 700lg/ml of the aminoglycoside geneticin(G418).

Preparation of microsomes. HK293 cells were isolatedfrom cell culture flasks and stored as pellets at )80 �C

290 B. Dean et al. / Archives of Biochemistry and Biophysics 402 (2002) 289–295

until needed. Pellets were homogenized (Polytron) in50mM Tris (pH 7.5) with 1.15% KCl and centrifuged at9000g for 10min, followed by centrifugation of the su-pernatant at 400,000g (Beckman Optima TLX Ultra-centrifuge) for 30min. The resulting pellet was washedwith 10mM EDTA (pH 7.4) with 1.15% KCl and respunas above. The final microsomal pellet was solubilized in10mM K2HPO4 (pH 7.4) with 250mM sucrose andstored at )80 �C in 1-ml aliquots containing 5mg/mltotal protein.

Western blot analysis. Expression of female rhesusUGT1A01 was examined by immunoblot using a poly-clonal antibody (WB-UGT1A) for human UGT1Asubfamily detection. Gentest Western blot standards(0:2lg UGT1A1 and 0.32 lg UGT1A6), Supersomes(0:5lg UGT1A9), and SeeBlue Plus2 prestained proteinstandard were used as controls for the detection of

rhesus UGT1A01 (4lg), rhesus kidney (40lg), andrhesus or human liver (40lg) samples. Samples wereseparated on 7% Tris–acetate NuPAGE gels and trans-ferred to a 0.2-lm nitrocellulose membrane followingsupplier’s recommended conditions for reduced samples.The blot was probed with a 1:1000 dilution ofWB-UGT1A primary antibody followed by a 1:10,000dilution of horseradish peroxidase-conjugated goat anti-rabbit IgG secondary antibody. Resulting immuno-complexes were incubated with Supersignal West PicoChemiluminescent substrate and visualized by exposureto Kodak BioMax MR film.

Catalytic activity. Probe substrates (25 or 100mM)were diluted in water, 75% acetonitrile, or methanolfollowed by 11 twofold serial dilutions. All reactionswere performed in 96-well format with columns re-presenting the different concentrations and rows

Fig. 1. Nucleotide sequence and translational product of UGT1A01 isolated from female rhesus monkey liver (Accession No. AF360121). The

suspected transition mutation is indicated by the box.

B. Dean et al. / Archives of Biochemistry and Biophysics 402 (2002) 289–295 291

representing the different time points. The final reaction(200 ll) consisted of substrate (5ll per well), 5 to 50lgof microsomal protein, alamethicin (1lg/10lg protein),10mM MgCl2, 50mM bis-Tris–propane (pH 7.1), and10mM saccharic acid 1,4-lactone. Plates were preincu-bated for 15min at 37 �C with mixing on a Boekel‘‘Jitterbug’’ incubator and the reaction was initiated byaddition of UDPGA (2mM final concentration). Sam-ples were quenched at selected time points (typically 0,10, 30, and 90min) by addition of 50% acetic acid (20ll)or organic solvent, centrifuged, and analyzed by HPLCor liquid scintillation counting (LSC) for bilirubin (seebelow).

HPLC analysis. Separation of product(s) from parentwas performed over 7.5min using linear gradient elutionfrom a 3� 50-mm Luna 3-lm Phenyl–Hexyl or C8column (Phenomenex) with detection at the UV maxand optional radiomonitoring (b-Ram). The mobilephase consisted of A (10mM NH4OAcþ 0:1% aceticacid) and B ((92.8:7.2 acetonitrile:methanol with 7.2mMNH4OAc) + 0.1% acetic acid). Mobile phases for bi-lirubin analysis consisted of A (10mM NH4OAc) and B(1:1 ACN:MeOH with 10mM NH4OAc). All HPLCseparations were carried out at room temperature with aflow rate of 1ml/min. Turnover was calculated as theglucuronide peak area divided by the sum of glucuro-

Fig. 2. Sequence comparison of UGT1A1 orthologs from human (AF297093), rhesus monkey (AF360121), cynomolgus monkey (AF104339), rat

(Rattus norvegicus) (U20551), and mouse (Mus musculus) (AF093878 residues 1–290 partial; L27122 residues 291–535 conserved).

292 B. Dean et al. / Archives of Biochemistry and Biophysics 402 (2002) 289–295

nide and parent peak areas. Km and Vmax parameterswere calculated using Enzyme Kinetics!Pro software(ChemSW, Fairfield, CA) with the rigorous least squaresmethod.

Bilirubin analysis. Catalytic activity for bilirubin wasdetermined by LSC following liquid–liquid extractionaccording to the method of Matern et al. [31]. Briefly,each 200-ll incubation was quenched with 50 ll of 0.7Mglycine, pH 2.0, containing 1% Triton X-100. An aliquot(200 ll) of this was extracted by 0.5ml of H2O-saturatedEtOAc. Aliquots of 400 ll were placed into 7-ml scin-tillation vials followed by addition of 5.5ml of UltimaFlo-M LSC cocktail.

Results

Detection and isolation of novel UGTs from rhesusmonkey liver cDNA were performed using primers forhuman UGT isoforms based on the high degree of ho-mology between primates. Screening of various primercombinations has led to the detection of a 1645-bp PCRproduct using the human UGT1A1 primers. This cDNAwas sequenced and data analysis confirmed existence ofa 1599-bp open reading frame encoding a 533-aminoacid translational product (Fig. 1). The deduced aminoacid sequence revealed greater than 99 and 95% identity(Fig. 2) with the cynomolgus UGT1A01 (GenBankAccession No. AF104339) and human UGT1A1 (Gen-Bank Accession No. AF297093) isoforms, respectively.Based on homology and function, this orthologousprotein isolated from female rhesus monkey liver cDNAhas been termed ‘‘rhesus UGT1A01’’ (GenBank Ac-cession No. AF360121).

The expressed female rhesus UGT1A01 was immu-noblotted with a human UGT1A family antibody(Fig. 3) to determine the possibility of detection invarious tissues and as a preliminary test for future UGTinduction studies. The antibody detected human stan-dards UGT1A1, 6, and 9 (faint band) along with rhesusUGT1A01. Bands of protein at the correct molecularmass (� 55kDa) were detected also in female and malerhesus liver and male rhesus kidney. The identity ofthese bands in rhesus liver and kidney remains to bedetermined. UGT1A protein was not detected in eitherthe female rhesus kidney or the male and female humanliver microsomes.To support sequencing data and Western blot detec-

tion, functional characterization of UGT1A01 wasperformed using select substrates (Fig. 4) for glucuron-

Fig. 3. Immunoblot detection of rhesus UGT1A01 with a human WB-

UGT1A polyclonal antibody. Overlay of chemiluminescent-exposed

Biomax MR film with nitrocellulose blot. Lanes 1, 14, and 15, see

Blue+2 Protein Standard; lane 2, human UGT1A1 (0.2lg); lane 3,human UGT1A6 (0.32 lg); lane 4, human UGT1A9 (0.5lg); lane 5,female rhesus UGT1A01 (4lg); lanes 6 and 7, female rhesus liver

(40lg); lanes 8 and 9, male rhesus liver (40 lg); lane 10, female rhesuskidney (40lg); lane 11, male rhesus kidney (40lg); lane 12, femalehuman liver (40 lg); lane 13, male human liver (40lg).

Fig. 4. Chemical structures of substrates for rhesus UGT1A01: bi-

lirubin (I), 1-naphthol (II), 17a-ethinylestradiol (III), 7-hydroxy-4-(trifluoromethyl)-coumarin (IV), b-estradiol (V).

Table 1

Kinetic parameters for glucuronidation with microsomal preparations

from HK 293 cells expressing female rhesus monkey UGT1A01

Substrate Apparent Km(lM)

Vmax (pmol/min/mg)

Vmax=Km(�100)

7-Hydroxy-4(trifluoro-

methyl)-coumarin

44� 7 2600� 226 5900

b-Estradiola 35� 8 303� 27 866

Ethinylestradiolb 28� 5 205� 32 732

1-Naphthol 330� 57 559� 74 169

Bilirubinc 4� 2 63� 6 1575

a b-Estradiol 3-glucuronide production.b Ethinylestradiol 3-glucuronide production.c Total glucuronide production of all isomers using [14C]UDPGA

and EtOAc extraction.

B. Dean et al. / Archives of Biochemistry and Biophysics 402 (2002) 289–295 293

idation (HTFMC) and those more specific for UGT1A1(1-naphthol, EE, E2, and bilirubin). Kinetic parameters(Vmax and Km) were calculated for several substrates(Table 1) and determined to be comparable to thecynomolgus UGT1A01 and human UGT1A1 values.The stereochemistry of glucuronides generated from

multifunctional substrates (EE and E2) was determined

based on retention time compared to commerciallyavailable or NMR-confirmed standards (data notshown). Glucuronidation of both EE and E2 was spe-cific to the 3-hydroxy position on the molecule. HPLCanalysis of bilirubin incubations produced four glucu-ronides (Fig. 5), but the stereochemistry is unknown atthis time.A group of benzyl alcohols were also screened for

glucuronidation activity by UGT1A01, but HPLCanalysis indicated no turnover for these compounds(Table 2).

Discussion

This paper describes the isolation and characteriza-tion of female rhesus UGT1A01, an ortholog of humanUGT1A1 and cynomolgus monkey UGTlA01 [32] andthe first UGT to be sequenced and characterized fromthis species. The amplified UGT1A01 cDNA encodesfor a functional full-length enzyme of 533 amino acidspossessing UGT characteristics such as an ER-directing

Fig. 5. In vitro glucuronidation of bilirubin depicting the presence of several products. (A) Bilirubin and UGT1A01 incubation without UDPGA

cofactor. Bilirubin with minor isomers are shown with dashed arrows. An unknown contaminant was detected at 4min in all incubations.

(B) Bilirubin and UGT1A01 incubation with UDPGA cofactor. Solid arrows indicate four major glucuronides of bilirubin and its isomers.

Table 2

Substituted benzyl alcohols not glucuronidated by UGT1A01

4-Monosubsti-

tuted benzyl

alcohols

3,5-Disubsti-

tuted benzyl

alcohols

2,6-Disubsti-

tuted benzyl

alcohols

R Group(s): Cl Cl Cl

Br Br

OH OH

CH3 CF3NO2 NO2

294 B. Dean et al. / Archives of Biochemistry and Biophysics 402 (2002) 289–295

signal peptide (residues 1–27), a transmembrane-span-ning domain (residues 490–508), a UDPGA-bindingdomain (residues 354–402), and potential asparagine-linked NXS or NXT glycosylation motifs (residues 102and 295).A sequence comparison of UGT1A01 to other

UGT1A family members isolated from female rhesusmonkey liver suggests that the same alternative splicingmechanisms in humans exist in rhesus monkey. Thisgenetic splicing of an individual exon 1 to four con-served downstream exons produces a unique N-terminalregion dictating substrate specificity along with a con-stant C-terminal region, encoding 246 amino acidscontaining the UDPGA-binding domain and the trans-membrane-spanning region.The expressed UGT1A01 (�55 kDa) was immuno-

reactive with an antibody specific for human UGT1Aproteins. Similar immunoreactive bands were detectedat 55 kDa in the male and female rhesus liver micro-somes and male kidney microsomes. No bands at thismolecular mass were detected for the female rhesuskidney microsomes or the human liver microsomes. Thisis likely due more to levels of expression, which areknown to be highly variable in individuals, since mi-crosomal incubations with ‘‘selective’’ UGT1A1 sub-strates show turnover suggesting the presence of theenzyme. In addition, due to the high variability of UGTexpression in tissues, it is common to not detect UGTsin microsomes from individuals, as compared to pooledsamples.

References

[1] B.L. Coffman, M.D. Green, C.D. King, T.R. Tephly, Mol.

Pharmacol. 47 (1995) 1101–1105.

[2] S. Haque, J, D.D. Petersen, D.W. Nebert, P.I. Mackenzie, DNA

Cell Biol. 10 (1991) 515–524.

[3] M. Beaulieu, E. Levesque, O. Barbier, D. Turgeon, G. Belanger,

D.W. Hum, A. Belanger, J. Mol. Biol. 275 (1998) 785–794.

[4] G. Belanger, M. Beaulieu, E. Levesque, D.W. Hum, A. Belanger,

DNA Cell Biol. 16 (1997) 1195–1205.

[5] C. Albert, M. Vallee, G. Beaudry, A. Belanger, D.W. Hum,

Endocrinology 140 (1999) 3292–3302.

[6] C.D. King, G.R. Rios, J.A. Assouline, T.R. Tephly, Arch.

Biochem. Biophys. 365 (1999) 156–162.

[7] B.L. Coffman, C.D. King, G.R. Rios, T.R. Tephly, Drug Metab.

Dispos. 26 (1998) 73–77.

[8] M. Ouzzine, J. Magdalou, B. Burchell, S. Fournel-Gigleux, FEBS

Lett. 454 (1999) 187–191.

[9] G. Jedlitschky, A.J. Cassidy, M. Sales, N. Pratt, B. Burchell,

Biochem. J. 340 (1999) 837–843.

[10] Z. Cheng, A. Radominska-Pandya, T.R. Tephly, Arch. Biochem.

Biophys. 356 (1998) 301–305.

[11] B.L. Coffman, G.R. Rios, C.D. King, T.R. Tephly, Drug Metab.

Dispos. 25 (1997) 1–4.

[12] B. Mojarrabi, R. Butler, P.I. Mackenzie, Biochem. Biophys. Res.

Commun. 225 (1996) 785–790.

[13] J.J. Bouska, R.L. Bell, C.L. Goodfellow, A.O. Stewart, C.D.

Brooks, G.W. Carter, Drug Metab. Dispos. 25 (1997) 1032–1038.

[14] C.D. King, M.D. Green, G.R. Rios, B.L. Coffman, I.S. Owens,

W.P. Bishop, T.R. Tephly, Arch. Biochem. Biophys. 332 (1996)

92–100.

[15] C.D. King, G.R. Rios, M.D. Green, P.I. MacKenzie, T.R.

Tephly, Drug Metab. Dispos. 25 (1997) 251–255.

[16] J.F. Ghersi-Egea, B. Leininger-Muller, G. Suleman, G. Siest, A.

Minn, J. Neurochem. 62 (1994) 1089–1096.

[17] M.K. Martinasevic, C.D. King, G.R. Rios, T.R. Tephley, Drug

Matab. Dispos. 26 (1998) 1039–1041.

[18] A. Radominska-Pandya, J.M. Little, J.T. Pandya, T.R. Tephly,

C.D. King, G.W. Barone, J.P. Raufman, Biochem. Biophys. Acta

1394 (1998) 199–208.

[19] C.P. Strassburg, N. Nguyen, M.P. Manns, R.H. Tukey, Gas-

troenterology 116 (1999) 149–160.

[20] D. Keppler, I. Leier, G. Jedlitschky, R. Mayer, M. Buchler, Adv.

Enzyme Regul. 36 (1996) 17–29.

[21] N. Kanai, R. Lu, Y. Bao, A.W. Wolkoff, M. Vore, V.L. Schuster,

Am. J. Physiol. 270 (1996) F326–F331.

[22] M.A. Brink, J.F. Slors, Y.C. Keulemans, K.S. Mok, D.R.

DeWaart, M.C. Carey, A.K. Groen, G.N. Tytgat, Gastroenterol-

ogy 116 (1999) 1420–1427.

[23] J.K. Ritter, F. Chen, Y.Y. Sheen, H.M. Tran, S. Kimura, M.T.

Yeatman, I.S. Owens, J. Biol. Chem. 267 (1992) 3257–3261.

[24] P.L. Jansen, Eur. J. Pediatr. 158 (Suppl 2) (1999) S89–S94.

[25] J.K. Ritter, F.K. Kessler, M.T. Thompson, A.D. Grove, D.J.

Auyeung, R.A. Fisher, Hepatology 30 (1999) 476–484.

[26] T.W. Hansen, J. Med. Liban. 47 (1999) 22–27.

[27] R.P. Wennberg, Cell. Mol. Neurobiol. 20 (2000) 97–109.

[28] J. Seppen, P.J. Bosma, B.G. Goldhoorn, C.T. Bakker, J.R.

Chowdhury, N.R. Chowdhury, P.L. Jansen, R.P. Elferink, J. Clin.

Invest. 94 (1994) 2385–2391.

[29] S. Aono, Y. Adachi, E. Uyama, Y. Yamada, H. Keino, T. Nanno,

O. Koiwai, H. Sato, Lancet 345 (1995) 958–959.

[30] H. Nielsen, J. Engelbrecht, S. Brunak, G. von Heijne, Protein

Eng. 10 (1997) 1–6.

[31] H. Matern, H. Heinemann, S. Matern, Anal. Biochem. 219 (1994)

182–188.

[32] M. Vallee, C. Albert, G. Beaudry, D.W. Hum, A. Belanger, J.

Steroid Biochem. Mol. Biol. 77 (2001) 239–249.

B. Dean et al. / Archives of Biochemistry and Biophysics 402 (2002) 289–295 295