Unconjugated Bilirubin Exhibits Spontaneous Diffusion throughModel Lipid Bilayers and Native Hepatocyte Membranes*

(Received for publication, April 20, 1998, and in revised form, January 27, 1999)

Stephen D. Zucker‡§, Wolfram Goessling¶, and Alison G. Hoppini

From the ‡Division of Digestive Diseases, University of Cincinnati Medical Center, Cincinnati, Ohio 45267-0595, the¶Department of Medicine, Brigham and Women’s Hospital, Boston, Massachusetts 02115, and the iDivision of PediatricGastroenterology, Massachusetts General Hospital, Boston, Massachusetts 02114

The liver is responsible for the clearance and metab-olism of unconjugated bilirubin, the hydrophobic end-product of heme catabolism. Although several putativebilirubin transporters have been described, it has beenalternatively proposed that bilirubin enters the hepato-cyte by passive diffusion through the plasma membrane.In order to elucidate the mechanism of bilirubin uptake,we measured the rate of bilirubin transmembrane diffu-sion (flip-flop) using stopped-flow fluorescence tech-niques. Unconjugated bilirubin rapidly diffuses throughmodel phosphatidylcholine vesicles, with a first-orderrate constant of 5.3 s21 (t1⁄2 5 130 ms). The flip-flop rate isindependent of membrane cholesterol content, phos-pholipid acyl saturation, and lipid packing, consistentwith thermodynamic analyses demonstrating minimalsteric constraint to bilirubin transmembrane diffusion.The coincident decrease in pH of the entrapped vesiclevolume supports a mechanism whereby the bilirubinmolecule crosses the lipid bilayer as the uncharged di-acid. Transport of bilirubin by native rat hepatocytemembranes exhibits kinetics comparable with that inmodel vesicles, suggesting that unconjugated bilirubincrosses cellular membranes by passive diffusionthrough the hydrophobic lipid core. In contrast, there isno demonstrable flip-flop of bilirubin diglucuronide orbilirubin ditaurate in phospholipid vesicles, yet thesecompounds rapidly traverse isolated rat hepatocytemembranes, confirming the presence of a facilitated up-take system(s) for hydrophilic bilirubin conjugates.

Unconjugated bilirubin is the principal degradation productof heme metabolism. Although the physiologic isomer, bilirubinIXa, is a dicarboxylic acid, the molecule has minimal aqueoussolubility at physiologic pH (1, 2) due to the formation of in-tramolecular hydrogen bonds (3). For this reason, bilirubinundergoes biotransformation in the liver to more polar conju-gates prior to secretion in the bile. However, while unconju-gated bilirubin is not hydrophilic, neither can it be character-ized as lipophilic (1), since this compound is equally insoluble

in apolar solvents (e.g. n-hexane, 1-pentanol). These uniquephysical-chemical properties have generated controversy re-garding the mechanism of bilirubin uptake by the liver. Basedon the spontaneous leakage from multilamellar liposomes (4),it has been proposed that bilirubin is able to diffuse throughcellular membranes (5, 6). On the other hand, in vivo (7) andwhole organ (8, 9) studies indicate that hepatic bilirubin up-take is saturable and occurs against a concentration gradient,findings that support a protein-mediated transport mecha-nism. To date, four putative bilirubin transporters have beenidentified in liver cells: BSP/bilirubin-binding protein, organicanion-transporting polypeptide, bilitranslocase, and organicanion-binding protein. Each of these proteins facilitates uptakeof the hydrophilic organic anion, sulfobromophthalein (BSP),1 aprocess that is competitively inhibited by bilirubin (10, 11).

However, despite the substantial body of evidence in supportof protein-mediated bilirubin uptake, none of the aforemen-tioned candidate proteins has been directly shown to transportunconjugated bilirubin. This is due, in part, to the use of BSPas a surrogate for bilirubin, based on the long standing as-sumption that these two compounds share a common trans-porter (12, 13). However, the cellular uptake of bilirubin andBSP can be dissociated (14, 15), suggesting distinct transportmechanisms. While it has been argued that the unique “per-meability” of hepatocytes to bilirubin confirms the presence ofa specific transporter (12, 16), other studies have demonstratednonsaturable and non-energy-dependent bilirubin uptake (17),indicative of a diffusional process. Moreover, fibroblasts trans-fected with cDNA for the bilirubin-conjugating enzyme, UDP-glucuronosyltransferase, do not express any of the candidatetransporters yet are effectively able to take up and conjugatebilirubin (18).

In an attempt to reconcile these disparate results and tofurther delineate the mechanism whereby bilirubin traversescellular membranes, we devised a stopped-flow fluorescencesystem to facilitate detailed kinetic analysis of bilirubin flip-flop in model and isolated native membrane vesicles. Our datasuggest that unconjugated bilirubin is capable of rapid, spon-taneous diffusion through lipid bilayers, findings with impor-tant implications for bilirubin clearance by the liver.

EXPERIMENTAL PROCEDURES

Materials—Nigericin, valinomycin, and essentially fatty acid-freebovine and human serum albumin were purchased from Sigma. Biliru-bin IXa (purity .99.5% by absorbance in chloroform, e453 5 62,000M21zcm21) and bilirubin conjugate (ditauratez2Na) were obtained from

* The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby marked“advertisement” in accordance with 18 U.S.C. Section 1734 solely toindicate this fact.This work was supported by National Institutes ofHealth Research Grant DK-51679 (to S. D. Z.), a Charles H. HoodFoundation Child Health Research Award (to S. D. Z.), a Harvard Di-gestive Diseases Center Pilot/Feasibility Grant (to S. D. Z.), and aBASF Foundation Postdoctoral Research Award (to W. G.).

Preliminary reports of this work have been published in abstractform (54, 55).

§ To whom correspondence should be addressed: Division of DigestiveDiseases, University of Cincinnati Medical Center, 231 Bethesda Ave.(ML 0595), Cincinnati, OH 45267-0595. Tel.: 513-558-5244; Fax:513-558-1744.

1 The abbreviations used are: BSP, sulfobromophthalein; BDT, bil-irubin ditaurate; BDG, bilirubin diglucuronide; BSA, bovine serumalbumin; cbBSA, Cascade Blue-labeled BSA; bLPM, basolateral (si-nusoidal) liver plasma membranes; dansyl-PE, N-(5-dimethylamino-naphthalene-1-sulfonyl)dipalmitoyl-L-a-phosphatidylethanolamine;pyranine, 8-hydroxypyrene-1,3,6-trisulfonic acid.

THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 274, No. 16, Issue of April 16, pp. 10852–10862, 1999© 1999 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A.

This paper is available on line at http://www.jbc.org10852

by guest on February 17, 2018http://w

ww

.jbc.org/D

ownloaded from

Porphyrin Products (Logan, UT). Grade 1 egg lecithin (phosphatidyl-choline) was purchased from Lipid Products (Surrey, United Kingdom),and cholesterol was obtained from Nu-Chek Prep (Elysian, MN). Di-palmitoylphosphatidylcholine, dioleoylphosphatidylcholine, and N-(5-dimethylaminonaphthalene-1-sulfonyl) dipalmitoyl-L-a-phosphatidyl-ethanolamine (dansyl-PE) were purchased from Avanti Polar Lipids(Birmingham, AL). The fluorescent probes 8-hydroxypyrene-1,3,6-trisulfonic acid (pyranine) and Cascade Blue-conjugated bovine serumalbumin as well as anti-Cascade Blue antibodies were obtained fromMolecular Probes, Inc. (Eugene, OR).

Preparation and Loading of Unilamellar Phospholipid Vesicles—Small unilamellar vesicles were prepared by probe sonication using amodification (19) of the technique of Barenholz et al. (20) A chloroformsolution of phosphatidylcholine was evaporated to dryness under argonatmosphere, solubilized in ether, and re-evaporated to form a uniformfilm. The lipids then were desiccated overnight under vacuum to re-move traces of ether, suspended in 0.1 M potassium phosphate solutionat pH 7.4 (unless otherwise stated), and sonicated until clear. For thepreparation of fluorescence-labeled vesicles, dansyl-PE was added tothe lecithin-containing chloroform solution so as to comprise 1 mol % oftotal phospholipid. Large unilamellar vesicles were prepared using amodification (21) of the injection method of Kremer et al. (22) Constit-uent phospholipids were suspended in ethanol and slowly injected intoa magnetically stirred aqueous solution of 0.1 M potassium phosphate(pH 7.4), with vesicle size regulated by the concentration of phospho-lipid in the injected ethanol. Cascade Blue-conjugated bovine serumalbumin (cbBSA) was entrapped at the time of vesicle preparation byinjecting the ethanolic phospholipid solution into phosphate buffer con-taining 50 mM cbBSA. The vesicle suspension (5 ml) was subjected todialysis against 100 volumes of phosphate buffer three successive timesto remove retained ethanol, and unentrapped cbBSA was separatedfrom the vesicles by elution on a Sepharose 4B column (42 3 2.5 cm).Mean vesicle hydrodynamic diameter was determined by quasielasticlight scattering (21).

Isolation of Membrane Vesicles from Rat Liver—Rat liver microsomalmembranes were isolated from fasted male Sprague-Dawley rats, asdescribed previously (21), and protein concentration was quantified bythe Bio-Rad assay. Hepatocyte basolateral (bLPM) and canalicular liverplasma membranes were prepared by the method of Meier and Boyer(23), with appropriate enrichment documented by enzymatic assay (21),and by Western blotting with polyclonal antibodies to the a-subunit ofthe plasma membrane Na1/K1 ATPase and with monoclonal antibodiesto the canalicular ecto-ATPase (generously provided by Dr. R. Green).Membranes were loaded with cbBSA (50 mM) by homogenization in thepresence of the probe, followed by exhaustive washes.

Analysis of Bilirubin Transmembrane Diffusion—The flip-flop ratefor unconjugated bilirubin was determined from fluorescence record-ings of bilirubin equilibration between a suspension of dansyl-PE-la-beled small unilamellar phosphatidylcholine vesicles and bovine serumalbumin (Fig. 1A). The binding of bilirubin to dansyl-PE-labeled vesi-cles causes a reduction in fluorescence intensity due to resonance en-ergy transfer between bilirubin and the dansyl moiety (19). Since un-conjugated bilirubin is poorly soluble at neutral pH, incorporation intophospholipid vesicles was accomplished by dissolving bilirubin in alka-line (pH 12) potassium phosphate (19). The addition of a small aliquot(#1%, v/v) of the bilirubin solution to a suspension of phospholipidvesicles buffered at pH 7.4 in 0.1 M potassium phosphate caused nodetectable alteration in pH. Alternatively, when indicated, bilirubinwas dissolved in dimethyl sulfoxide (Me2SO), and a small aliquot (#1%,v/v) was added to the vesicle suspension (24). At this low concentration,Me2SO has no discernible affect on membrane physical properties (25,26), and the use of either method to solubilize bilirubin yielded identicalkinetic results. All steps were performed in diminished light to mini-mize bilirubin photodegradation.

An Aminco-Bowman II fluorescence spectrophotometer equippedwith an SLM-Aminco MilliFlow stopped-flow reactor (mixing time: 3ms) facilitated the rapid mixing of a suspension of dansyl-labeled ves-icles containing bound bilirubin with an equal volume of a bovine serumalbumin solution. Bilirubin equilibration is manifest by a time-depend-ent reemergence of dansyl fluorescence (excitation, 340 nm; emission,525 nm), reflecting the transfer of bilirubin from the vesicles to BSA(27). The time course was analyzed by fitting the fluorescence curves tothe function,

F~t! 5 OAiexp~ 2 kit! 1 C (Eq. 1)

where t is time, ki is the ith order rate constant, Ai is the amplitude, andC is a constant term. Quality of fit was assessed by regression analysis,

and an F-test was applied to determine if the inclusion of an additionalterm provided a statistically significant improvement in curve fit (19,21). Due to the rapid rate of bilirubin solvation (19), analysis of biliru-bin-membrane dissociation was performed with an Applied Photophys-ics (Leatherhead, UK) fluorescence spectrophotometer equipped withan SPF-17 stopped-flow device (mixing time: 0.7 ms).

An alternative method utilized for monitoring bilirubin flip-flop in-volved the preparation of a suspension of bilirubin-containing phos-phatidylcholine vesicles, which were mixed, using stopped-flow tech-niques, with a solution of free cbBSA. Bilirubin binding to cbBSAinduces a concentration-dependent decrease in Cascade Blue fluores-cence (excitation, 380 nm; emission, 430 nm), and flip-flop rates werecalculated from the time course of the quenching of cbBSA. The rate ofbilirubin transmembrane diffusion also was determined from thequenching of cbBSA entrapped within model or native membrane ves-icles (Fig. 1B). Membranes were preincubated with anti-Cascade Blueantibodies in order to completely neutralize all residual unincorporatedprobe; therefore, the quenching of Cascade Blue fluorescence is predi-cated on bilirubin traversing the acceptor membrane and binding to theentrapped cbBSA. For these studies, bilirubin was incorporated intounlabeled small unilamellar phosphatidylcholine donor vesicles or, al-ternatively, solubilized in 0.1 M potassium phosphate (pH 7.4) contain-ing Me2SO (20%, v/v). In order to minimize light-scattering effects,

FIG. 1. Schematic illustration of experimental systems utilizedfor studying bilirubin transmembrane diffusion. A depicts theexperimental schema for monitoring bilirubin equilibration betweendonor vesicles, composed of phosphatidylcholine (dark circles) with 1mol % dansyl-phosphatidylethanolamine (light circles) and BSA. Attime 0 (t0), bilirubin-containing vesicles are rapidly mixed with a solu-tion of BSA, using stopped-flow techniques. The equilibration rate isdetermined from the time-dependent re-emergence of dansyl fluores-cence caused by the spontaneous transfer of bilirubin from the vesiclesto albumin. The resultant fluorescence curve reflects both rapid solva-tion (Dissociation) from the external hemileaflet of the bilayer andslower transmembrane diffusion (Flip-flop) of bilirubin from the innerto the outer hemileaflet. Bilirubin flip-flop across unlabeled (nonfluo-rescent) phospholipid vesicles was monitored by substituting cbBSA forBSA and recording the quenching of Cascade Blue fluorescence bybilirubin. B illustrates an alternative method for measuring bilirubinflip-flop rates in which cbBSA is entrapped within acceptor vesicles.Bilirubin flip-flop is reflected in the time-dependent quenching of Cas-cade Blue fluorescence following mixing of cbBSA-containing vesicleswith bilirubin-incorporated donor vesicles.

Transmembrane Diffusion of Bilirubin 10853

by guest on February 17, 2018http://w

ww

.jbc.org/D

ownloaded from

experiments utilizing isolated native membrane vesicles employed anemission wavelength of 450 nm with a 440-nm cut-on filter.

Kinetic Analysis of Bilirubin Transmembrane Diffusion—We previ-ously have shown that the equilibration of bilirubin between unilamel-lar phospholipid vesicles and BSA occurs via aqueous diffusion (27).Assuming that bilirubin flip-flop is significantly slower than membranedissociation, the equilibration rate (R) can be described by the expres-sion (27, 28),

R 5

KaBSA

KaV 1

@V#

@BSA#

KaBSA

KaV S 1

kffD1

@V#

@BSA#S 1koff

BSAD (Eq. 2)

where kff represents the flip-flop rate constant, koffBSA represents the

dissociation rate constant from BSA, Ka represents the associationconstant, and [BSA] and [V] represent the concentrations of bovineserum albumin and vesicle phospholipid, respectively. Based on thisequation, a plot of R versus [V]/[BSA] intersects the y axis at kff andasymptotically approaches koff

BSA at high molar ratios of phospholipid toBSA. Values for the various kinetic parameters were determined bymeasuring the bilirubin equilibration rate over a range of phospholipid:albumin molar ratios and fitting the data to Equation 2.

Preparation of Bilirubin Diglucuronide—Bilirubin diglucuronide(BDG) was isolated from bile enriched by the intravenous administra-tion of 20 mM unconjugated bilirubin in 3.2% bovine serum albumin toanesthetized male Sprague-Dawley rats (29). BDG was extracted withchloroform/ethanol and NaCl-saturated acidified glycine and isolatedby thin layer chromatography. The purity of the BDG preparation(.95%) was measured by reverse-phase high pressure liquid chroma-tography (30), with the initial mobile phase consisting of a 35% solutionof 1% acetic acid and 65% 25 mM sodium acetate in methanol/chloroform(95:5, v/v). After 30 s, the sodium acetate solution was increased in alinear fashion to 100% over 8 min, followed by a linear decrease to 80%over 5 min. The flow rate was maintained constant at 1 ml/min.

Effect of Bilirubin on Membrane Vesicle pH—In order to determinethe effect of bilirubin on the internal pH of membrane vesicles, 0.5 mM

pyranine, a hydrophilic pH-sensitive fluorescent probe, was dissolved in25 mM HEPES/KOH (pH 7.4) and entrapped within small unilamellar

phosphatidylcholine vesicles by sonication (31). Unentrapped pyraninewas removed from the medium by elution on a Sephadex G-25 (coarsegrade) column (40 3 1 cm). Pyranine-loaded vesicles were diluted to aconcentration of 0.5 mM phospholipid, and 2 ml of total volume wasplaced in a temperature-controlled, stirred cuvette. The internal pH ofthe vesicles, as reflected by the fluorescence intensity of the entrappedpyranine, was monitored continuously at 25 °C using steady-state tech-niques. Excitation (405 nm) and emission (520 nm) wavelengths wereselected so as to minimize bilirubin inner filter effects. At these wave-lengths, pyranine fluorescence intensity correlates inversely with pHsuch that an increase in hydrogen ion concentration manifests as acorresponding increase in probe fluorescence.

Flip-flop of bilirubin from the outer to the inner bilayer hemileafletwas monitored by changes in pyranine fluorescence following the addi-tion of a small (10-ml) aliquot of unconjugated bilirubin. To facilitatedistinction between fluorescence changes induced by bilirubin absorb-ance versus those due to alterations in the pH of the entrapped volume,vesicles were pH-clamped by pretreatment with 3 ml (1 mg/mg phospho-lipid) of an ethanolic solution of the proton/potassium ionophore, nigeri-cin (31). Control experiments were performed to document that thesmall volume of added ethanol (0.15%, v/v) did not alter the permeabil-ity of the vesicles to protons and that there was no effect of the bilirubinvehicle on pyranine fluorescence, either in the presence or absence ofnigericin. Flip-flop of bilirubin from the inner to the outer hemileaflet ofthe membrane bilayer was examined by adding a 10-ml aliquot of bovineserum albumin dissolved in 25 mM HEPES/KOH (pH 7.4) to bilirubin-containing vesicles, producing a final BSA concentration of 5 mM.

RESULTS

In order to determine whether unconjugated bilirubin is ableto spontaneously traverse a membrane bilayer, we studied theequilibration of bilirubin between 1 mol % dansyl-PE-labeledsmall unilamellar phosphatidylcholine vesicles and bovine se-rum albumin (Fig. 2). The resultant fluorescence curve is bestdescribed by a double exponential function, suggesting that theequilibration process consists of two kinetically distinct events.We postulate that the fast phase reflects bilirubin dissociationfrom the external hemileaflet of the vesicle bilayer, while the

FIG. 2. Bilirubin equilibration between dansyl-labeled vesicles and bovine serum albumin. A representative fluorescence curveobtained after mixing a suspension of 1 mol % dansyl-PE-labeled small unilamellar phosphatidylcholine vesicles containing unconjugated bilirubin(0.5 mM) with a solution of bovine serum albumin is displayed. The time-dependent reemergence of fluorescence reflects bilirubin equilibrationbetween the donor vesicles (100 mM phospholipid) and BSA (50 mM). The tracing, which represents the mean of six stopped-flow injections performedat 25 °C, was generated by obtaining a total of 200 fluorescence readings over a time interval of 50 ms, followed by an additional 200 readingsduring the ensuing 500 ms. The data are normalized to a scale of 0 to 1 and fit to both single (dotted line) and double (solid line) exponentialfunctions. The double exponential equation provides the best fit of the data, supporting the presence of two distinct kinetic components to theequilibration process: Afast 5 0.202 6 0.003, kfast 5 222 6 5 s21, Aslow 5 0.112 6 0.001, kslow 5 10.8 6 0.4 s21 (6 S.E.), where A is the amplitudeand k is the rate constant. In the inset, the slow phase of bilirubin transfer is isolated from the fast phase by recording fluorescence intensity every200 ms over a 20-s interval. The curve reflects the mean of 10 stopped-flow injections performed at 25 °C and is best described by a singleexponential (solid line), with a first-order rate constant of 3.7 6 0.4 s21.

Transmembrane Diffusion of Bilirubin10854

by guest on February 17, 2018http://w

ww

.jbc.org/D

ownloaded from

slow phase, which is readily resolved using a longer samplinginterval (Fig. 2, inset), represents bilirubin flip-flop from theinner to the outer hemileaflet of the phospholipid bilayer. Insupport of this hypothesis, the rate constant for the fast com-ponent (kfast) is identical to previously reported bilirubin off-rates from small unilamellar phosphatidylcholine vesicles (19).The finding that the ratio of the amplitudes of the fast versusslow phases of transfer (Afast:Aslow) is identical to the outer:inner surface area of the vesicles (1.8:1) further supports thecontention that the slower process corresponds to transmem-brane diffusion from the internal to the external hemileaflet ofthe membrane bilayer.

A fluorescence system utilizing phosphatidylcholine vesiclesloaded with cbBSA also was employed to verify that the slowcomponent of bilirubin transfer represents bilirubin transmem-brane flip-flop. Small unilamellar phosphatidylcholine donorvesicles were preincubated with bilirubin and then rapidlymixed with a suspension of cbBSA-loaded acceptor vesicles(Fig. 3, left panel). Alternatively, when bilirubin was solubi-lized in phosphate buffer containing 20% Me2SO (in the ab-sence of donor vesicles), a similar flip-flop signal was observed,since bilirubin must traverse the acceptor vesicle bilayer inorder to quench cbBSA fluorescence. While the curve obtainedin the presence of Me2SO equilibrates at a slightly higher baseline due to the enhanced aqueous solubility of bilirubin, thefinding that the equilibration rate constant is unaffected by thepresence or absence of donor vesicles indicates that diffusion ofbilirubin across the acceptor membrane is rate-limiting. Con-trol experiments demonstrate that neither vesicles nor Me2SOhave a direct effect on cbBSA fluorescence (Fig. 3, left panel).

The transfer of bilirubin from unlabeled phosphatidylcholinedonor vesicles to free cbBSA (Fig. 3, right panel) exhibits kinet-ics comparable with cbBSA-loaded acceptor vesicles, although,in this case, the fluorescence signal reflects bilirubin flip-flopfrom the inner to the outer hemileaflet of the donor vesicles.This hypothesis is supported by the nearly instantaneous de-

cline in fluorescence that occurs when bilirubin is solubilized in20% Me2SO, since there is no intervening membrane to impedebilirubin access to cbBSA, and the association rate of bilirubinfor cbBSA is more rapid than can be resolved by the stopped-flow apparatus. Hence, the quenching of cbBSA by bilirubin isnearly complete before the first data point is recorded. Thesefindings confirm that a flip-flop signal is observed only whenbilirubin and cbBSA are separated by a membrane bilayer.Moreover, the similarity in the equilibration rate between un-labeled (Fig. 3, right panel) and dansyl-labeled (Fig. 2, inset)vesicles indicates that the presence of the dansyl-PE probe doesnot affect bilirubin flip-flop kinetics.

Transmembrane Diffusion of Bilirubin Conjugates—We ex-amined whether the non-hydrogen-bonded, water-soluble bili-rubin derivative, bilirubin ditaurate (BDT), is able to diffusespontaneously through phospholipid bilayers. Since BDT is lessefficient than unconjugated bilirubin at quenching cbBSA flu-orescence (Fig. 4, inset), a 10-fold higher concentration wasutilized to ensure that a flip-flop signal would be readily de-tectable. In contradistinction to unconjugated bilirubin, biliru-bin ditaurate causes no significant attenuation in the fluores-cence of entrapped cbBSA over a 150-s time course (Fig. 4),indicating that spontaneous diffusion of BDT across phosphati-dylcholine bilayers is extremely slow. In control experiments,20% Me2SO had no effect on the rate of BDT flip-flop, verifyingthat Me2SO does not alter membrane permeability to thisbilirubin conjugate. Results identical to BDT were obtainedwith the principal physiologic bilirubin conjugate, BDG, whilemesobilirubin XIIIa (kindly provided by Dr. D. Lightner), abilirubin isomer that retains internal hydrogen bonding (26),exhibited a flip-flop signal identical to that of bilirubin IXa.These findings suggest that internal hydrogen bonds are per-missive to bilirubin transmembrane diffusion.

Kinetic Analysis of Bilirubin Flip-flop—In order to deter-mine the rate of bilirubin transmembrane diffusion, equilibra-tion of bilirubin between dansyl-PE-labeled small unilamellar

FIG. 3. Detection of bilirubin flip-flop with cascade blue-labeled bovine serum albumin. In the left panel, bilirubin flip-flop is monitoredby the quenching of cbBSA entrapped within phospholipid vesicles. In these experiments, bilirubin (2 mM) is bound to small unilamellarphosphatidylcholine donor vesicles (SULV) or alternatively solubilized in 0.1 M potassium phosphate (pH 7.4) containing 20% dimethyl sulfoxide(DMSO) and then mixed with an equal volume of cbBSA-loaded acceptor vesicles. Each curve represents an average of five stopped-flow injectionsconducted at 25 °C. Similar kinetics are observed in the presence or absence of donor vesicles, since bilirubin must traverse the acceptor membranebilayer to quench the entrapped cbBSA. No change in cbBSA fluorescence is observed in the presence of small unilamellar vesicles (or Me2SO) butin the absence of bilirubin (Control). The right panel displays the results of experiments in which bilirubin equilibration is monitored by thequenching of free (unentrapped) cbBSA. A characteristic flip-flop curve is obtained when vesicles (SULV) serve as the bilirubin donor, reflectingbilirubin diffusion across the donor bilayer. However, when bilirubin is dissolved in Me2SO, equilibration is instantaneous (DMSO), since there isno intervening membrane to impede access of bilirubin to the albumin, and the bilirubin on-rate for cbBSA exceeds the mixing time of thestopped-flow apparatus. Again, no quenching of cbBSA is observed in the absence of bilirubin (Control).

Transmembrane Diffusion of Bilirubin 10855

by guest on February 17, 2018http://w

ww

.jbc.org/D

ownloaded from

phosphatidylcholine vesicles and BSA was studied over a broadrange of donor and acceptor concentrations, and the measuredfirst-order rate constants were plotted against the phospholip-id:BSA molar ratio (Fig. 5). It is notable that individual trans-fer curves (Fig. 5, inset) constitute approximately one-third ofthe total change in fluorescence signal from base line (0 mM

BSA), consistent with the internal:external surface area of thevesicles. According to Equation 2, at low ratios of phospholipidto BSA, the equilibration rate (R) approaches kff, the value ofwhich is calculated to be 5.3 6 0.6 s21 (t1⁄2 5 130 6 15 ms;6S.E.) based on the best fit of the data. Conversely, at highratios of phospholipid to albumin, R asymptotically approacheskoff

BSA. The value of 0.16 s21 obtained from the curve fit isconsistent with previously reported bilirubin off-rates fromBSA (27). Comparable results were obtained when bilirubinequilibration between unlabeled (nonfluorescent) phosphati-dylcholine vesicles and cbBSA was examined (data not shown).

While it previously has been shown that membrane lipidpacking and cholesterol content are key determinants of bili-

rubin solvation (21), we found that neither vesicle hydrody-namic diameter (Table I), nor phospholipid:cholesterol ratiosup to 50 mol % (data not shown) significantly altered the rateof bilirubin flip-flop. There also was no difference in the flip-flop rate for vesicles composed of dioleoylphosphatidylcholineor dipalmitoylphosphatidylcholine as compared with egg phos-phatidylcholine, suggesting that acyl chain length and satura-tion have little impact on the flip-flop process. To investigatewhy membrane composition and lipid packing do not alter therate of bilirubin flip-flop, we determined thermodynamic acti-vation parameters by examining bilirubin equilibration be-tween small unilamellar phosphatidylcholine vesicles andcbBSA over a temperature range of 10–40 °C (19). From anArrhenius plot of the data (Fig. 6), the free energy of activation(DG‡) at 25 °C was calculated to be 16.9 kcal/mol, the principaldeterminant of which is the activation enthalpy (DH‡ 5 12.6kcal/mol), with only a minor entropic contribution (TDS‡ 5 -4.3kcal/mol). These findings suggest a relatively modest energybarrier to bilirubin transmembrane diffusion, with minimal

FIG. 4. Transmembrane diffusion ofunconjugated bilirubin and bilirubinditaurate. The flip-flop of unconjugatedbilirubin (2 mM) and bilirubin ditaurate(20 mM) across unilamellar phosphatidyl-choline vesicles (1 mM phospholipid) wasmonitored by the time-dependent quench-ing of entrapped cbBSA. A solution of un-conjugated bilirubin (UCB) solubilized in20% Me2SO or BDT dissolved in 0.1 M

potassium phosphate, was mixed with asuspension of cbBSA-loaded vesicles, us-ing stopped-flow techniques. Each curvereflects the average of six stopped-flowinjections performed at 25 °C. The insetdemonstrates that unconjugated bilirubin(circles) and bilirubin ditaurate (squares)quench the fluorescence of free cbBSA (5mM) in a linear manner up to a 1:1 molarratio of bilirubin to albumin. Each pointreflects the mean 6 S.D. of three experi-ments and, following correction for biliru-bin inner filter effects, is normalized to ascale of 0–1.

FIG. 5. Influence of the phospho-lipid:albumin molar ratio on the bili-rubin equilibration rate. The rate ofbilirubin (2 mM) equilibration betweensmall unilamellar phosphatidylcholinevesicles containing 1 mol % dansyl-PEand BSA was measured at 25 °C over arange of phospholipid (PL) and albuminconcentrations. The first-order rate con-stant is plotted against the molar ratio ofphospholipid to albumin, with each pointreflecting the mean of five stopped-flowinjections. The curve (solid line) is gener-ated from best-fit (p , 0.0005) parame-ters for Equation 2: kff 5 5.3 6 0.6 s21,koff

BSA 5 0.16 6 0.18 s21, KaBSA/Ka

V 5 778 61000 (6S.E.). The inset displays fluores-cence recordings of bilirubin equilibrationbetween dansyl-labeled vesicles (1 mM

phospholipid) and increasing concentra-tions of BSA (0 –20 mM) obtained at200-ms intervals. Each curve reflects theaverage of six stopped-flow injections andis fit to a single exponential function (sol-id line).

Transmembrane Diffusion of Bilirubin10856

by guest on February 17, 2018http://w

ww

.jbc.org/D

ownloaded from

steric constraints.Bilirubin Flip-flop in Isolated Rat Liver Membranes—Since

long-chain fatty acid flip-flop has been shown to be slower innative as compared with model membranes (32), we deter-mined bilirubin flip-flop rates from the time-dependentquenching of cbBSA entrapped within isolated native rat hep-atocyte membrane vesicles. Under the conditions utilized, bil-irubin off-rates (21) are over 40 times faster than transmem-brane flip-flop; hence, solvation from the surface of the donorvesicle is not rate-limiting. Rat liver microsomes initially wereutilized as the bilirubin acceptor, with vesicle latency andadequate entrapment of cbBSA confirmed by mannose-6-phos-phatase assay and treatment with anti-Cascade Blue antibod-ies. The rate of bilirubin flip-flop across microsomal mem-branes was found to be identical to that for phosphatidylcholinevesicles (Fig. 7).

While these results confirm that microsomes are permeableto bilirubin, since bilirubin gains access to the entrappedcbBSA only by traversing the microsomal membrane, the pos-sibility that microsomal bilirubin transport is more rapid thanin model vesicles with the observed rate limited by bilirubindiffusion across the donor vesicle is not excluded. For thisreason, flip-flop also was monitored by the equilibration ofbilirubin between isolated native rat liver membranes and freecbBSA (Fig. 8). Experiments were conducted over a range ofmembrane phospholipid and cbBSA concentrations, since the

measured rate is a function of the molar ratio of phospholipid toalbumin (Equation 2). Equilibration rates for native hepatocytemembranes were found to be equivalent to or lower than inmodel vesicles at all phospholipid:albumin molar ratios. Sincemodel vesicles contain no transport proteins, these data indi-cate that the rate of bilirubin uptake by rat liver membranesdoes not exceed spontaneous diffusion. The convergence of

TABLE IEffect of vesicle size on the rate of bilirubin flip-flop

The effect of membrane lipid packing on bilirubin flip-flop was deter-mined by comparing the first-order rate constant for bilirubin equili-bration between unilamellar phosphatidylcholine vesicles (small versuslarge) and cbBSA (10 mM). The mean hydrodynamic diameter 6 S.D. ofthe vesicle preparations (1 mM phospholipid) was determined byquasielastic light scattering. Equilibration rate constants, which reflectthe mean 6 S.D. of four separate sets of 10 stopped-flow injectionsperformed at 25 °C, were not statistically different between the twogroups of vesicles (p 5 0.72).

Vesicle hydrodynamic diameter First-order rate constant

nm s21

Small 31 6 1 1.68 6 0.20

Large 148 6 22 1.72 6 0.05

FIG. 6. Bilirubin transmembrane diffusion: Arrhenius plot.The natural log of the first-order rate constant (k) for bilirubin (2 mM)equilibration between small unilamellar phosphatidylcholine vesicles(1 mM phospholipid) and cbBSA (10 mM) is plotted against inversetemperature (K21). Each point represents the mean 6 S.D. of threeexperiments. Thermodynamic activation parameters are calculatedfrom the slope of the plot (r2 5 0.939).

FIG. 7. Comparison of bilirubin diffusion in model vesiclesversus hepatic microsomes. The changes in fluorescence associatedwith equilibration of bilirubin (2 mM) between small unilamellar donorvesicles (1 mM phospholipid) and cbBSA-loaded phosphatidylcholineacceptor vesicles (upper panel) or 0.5 mg of protein/ml cbBSA-loaded ratliver microsomes (lower panel) were recorded at 25 °C. Each curverepresents the average of six stopped-flow injections. At comparablemembrane phospholipid concentrations, the rate of bilirubin flip-flop issimilar between model and native vesicles, with rate constants of0.071 6 0.003 s21 and 0.077 6 0.004 s21 (6S.E.), respectively.

FIG. 8. Bilirubin flip-flop in model and native hepatocytemembranes. The rate of bilirubin (2 mM) equilibration between mem-brane vesicles and free cbBSA was measured over a range of phospho-lipid (PL):albumin molar ratios. Small unilamellar phosphatidylcholinevesicles (●, short dashed line), microsomal membranes (f, solid line),basolateral liver plasma membranes (Œ, dotted line), or canalicular liverplasma membranes (�, long dashed line), at concentrations rangingfrom 25 to 500 mM phospholipid, were utilized as the bilirubin donor.Each point reflects the mean 6 S.D. of five sets of experiments per-formed at 25 °C, with lines generated from best fit parameters forEquation 2.

Transmembrane Diffusion of Bilirubin 10857

by guest on February 17, 2018http://w

ww

.jbc.org/D

ownloaded from

rates at low phospholipid:cbBSA molar ratios further suggeststhat kff is similar for model and native membranes, consistentwith the hypothesis that bilirubin transport across hepatocytemembranes occurs via spontaneous diffusion. We postulatethat the divergence in equilibration rates at high ratios ofphospholipid to cbBSA is due to variations in the bindingaffinity of the membrane preparations for bilirubin. While italso is conceivable that these differences reflect the effect ofmembrane lipid composition or vesicle size on bilirubin diffu-sion, this possibility seems unlikely in light of our demonstra-tion that membrane hydrodynamic diameter, acyl chain satu-ration, and cholesterol content do not alter flip-flop kinetics(see above).

In order to establish whether water-soluble bilirubin conju-gates are able to traverse hepatocyte membranes, we measuredthe rate of bilirubin diglucuronide equilibration between smallunilamellar phosphatidylcholine vesicles (Fig. 9, upper leftpanel), rat liver microsomes (Fig. 9, upper right panel), baso-lateral liver plasma membranes (Fig. 9, lower left panel), orcanalicular plasma membranes (Fig. 9, lower right panel) andcbBSA. In contrast with model membranes, both BDG and BDT(data not shown) readily traverse microsomal membranes at arate that is 5-fold faster than unconjugated bilirubin. Thesecompounds also exhibit rapid diffusion across bLPM and canal-icular liver plasma membranes, confirming the presence oftransporters for bilirubin conjugates in these hepatocyte mem-brane fractions.

Mechanism of Bilirubin Transmembrane Diffusion—In-creased uptake of unconjugated bilirubin by basolateral liver

plasma membranes in the presence of an inwardly directedpotassium gradient and the potassium ionophore, valinomycin,has been cited as evidence for a bilirubin-specific electrogenictransporter (24). We examined the effect of valinomycin onbilirubin transport by preincubating isolated rat bLPM with 10mM valinomycin (or the ethanol vehicle) prior to recording bil-irubin equilibration, using cbBSA as the bilirubin acceptor.Initial experiments were performed in the absence of a potas-sium gradient (Kin

1 5 Kout1 ) by suspending the membranes and

cbBSA in 0.1 M potassium phosphate, 0.15 M sucrose, 10 mM

HEPES (pH 7.4). In the presence of valinomycin, bilirubinequilibration was best described by a single exponential func-tion, while the equilibration curve was best fit by a doubleexponential equation in the absence of valinomycin (Fig. 10). Inthe latter experiments, the rate constant for the fast compo-nent of transfer (1.8 6 0.8 s21) is essentially identical to thefirst-order rate constant measured in the presence of valino-mycin (1.9 6 0.3 s21), suggesting that the fast process repre-sents bilirubin flip-flop. In the absence of potassium, (0.25 M

sucrose, 10 mM HEPES), bilirubin equilibration fits a doubleexponential function irrespective of whether the vesicles arepretreated with valinomycin (data not shown), demonstratingthat abrogation of the slow component requires the presence ofboth potassium and valinomycin.

We subsequently examined the effect of a potassium gradienton the rate of bilirubin equilibration between rat bLPM andfree cbBSA. Since these experiments monitor bilirubin flip-flopfrom the inner to the outer hemileaflet of the membrane bi-layer, we applied an outwardly directed potassium gradient by

FIG. 9. Transport of unconjugatedbilirubin and bilirubin diglucu-ronide by hepatocyte membranes.Representative fluorescence tracings ofthe equilibration of unconjugated biliru-bin (UCB) and BDG between small unila-mellar phosphatidylcholine vesicles (1mM phospholipid) or isolated rat livermembrane fractions (0.5 mg/ml protein)and free cbBSA (10 mM) are displayed.Each curve reflects the average of 10stopped-flow injections performed at25 °C and is normalized to a scale of 0–1.The rate of flip-flop of unconjugated bili-rubin (2 mM) across model vesicles (upperleft panel) is similar to that measured inisolated rat liver microsomal membranes(upper right panel), basolateral plasmamembranes (lower left panel), and canal-icular plasma membranes (lower rightpanel). In contrast, bilirubin diglucu-ronide (20 mM) exhibits no spontaneousflip-flop in model vesicles but rapidlytraverses native hepatocyte membranes.At these donor and acceptor concentra-tions, the first-order rate constant forBDG flip-flop in microsomal membranesof 0.29 6 0.02 s21 (6S.E.) is 5 times fasterthan unconjugated bilirubin (k 5 0.058 60.002 s21).

Transmembrane Diffusion of Bilirubin10858

by guest on February 17, 2018http://w

ww

.jbc.org/D

ownloaded from

preparing rat bLPM in 0.1 M potassium phosphate, 0.15 M

sucrose, 10 mM HEPES (pH 7.4) and diluting 1:200 in 0.25 M

sucrose, 10 mM HEPES (pH 7.4). This vesicle suspension thenwas mixed with an equal volume of 0.25 M sucrose, 10 mM

HEPES (pH 7.4) containing cbBSA. Findings were no differentfrom those observed in the presence of equimolar potassium,indicating that it is not a potassium gradient per se that drivesbilirubin uptake. Taken together, these data support the exist-ence of two distinct phases of bilirubin uptake, the slower ofwhich is enhanced by the combined presence of valinomycinand potassium. Our results are consistent with those of Pascoloet al. (24) and could be construed as evidence for electrogenictransport. However, when similar experiments are performedusing model vesicles composed of phosphatidylcholine in placeof native bLPM, identical results are obtained (Fig. 10, inset).These findings strongly suggest that the enhanced uptake ofbilirubin in the presence of potassium and valinomycin is notdue to a specific electrogenic transport system.

Alternatively, based on our kinetic data, we postulate thatthe uptake of bilirubin by bLPM occurs via spontaneous trans-membrane diffusion of the uncharged bilirubin diacid (BH2). AtpH 7.4, a significant proportion of bilirubin is present as themono- and dianion (2). If our hypothesis is correct, bilirubinuptake would necessitate the acquisition of a proton(s) at theexternal hemileaflet and subsequent release of a proton(s) atthe internal hemileaflet of the membrane bilayer, in a manneranalogous to fatty acid flip-flop (31). As it previously has beenshown that dispersion of a transmembrane pH potential islimited by slow counterdiffusion of potassium ions (33), wepropose that the slow phase of bilirubin equilibration reflectsgradual dissipation of the proton gradient generated by biliru-bin flip-flop and that valinomycin abolishes this process bycollapsing the bilirubin-induced pH gradient. Based on thispostulated mechanism, bilirubin uptake should cause a de-crease in the pH of the entrapped volume as bilirubin anionsacquire protons from the external medium and release theminto the vesicle interior.

In order to test this hypothesis, we incorporated pyranine, awater-soluble pH-sensitive probe, within small unilamellarphosphatidylcholine vesicles and monitored pH throughchanges in pyranine fluorescence. The addition of bilirubin topyranine-loaded vesicles induces a sharp decline in pyraninefluorescence intensity, despite preincubation of the vesicleswith 1 mg nigericin/mg phospholipid to dissipate any transbi-

layer pH gradient (Fig. 11, left panel). Since the vesicles areeffectively pH-clamped and the bilirubin vehicle does not alterpyranine fluorescence, the decrease in fluorescence intensitydoes not reflect changes in vesicle pH but rather is due tobilirubin inner filter effects and/or resonance energy transfer.When BSA subsequently is added to the vesicle suspension, asharp increase in fluorescence is observed, probably reflectingdecreased quenching of the entrapped pyranine and shifts inthe bilirubin absorbance spectrum associated with binding toBSA (34). Despite these pronounced spectral effects, fluores-cence changes resulting from alterations in the pH of the en-trapped vesicle volume can be distinguished from those causedby bilirubin absorbance. In the absence of nigericin, the addi-tion of bilirubin to pyranine-loaded vesicles also induces anabrupt decline in pyranine fluorescence (Fig. 11, middle panel).However, the fluorescence signal equilibrates at a higher in-tensity as compared with identical studies performed in thepresence of nigericin (Fig. 11, left panel). As pyranine fluores-cence intensity correlates inversely with pH, these findings areindicative of a decrease in the internal pH of the vesicles, aneffect that is promptly reversed by nigericin. The subsequentaddition of BSA brings the fluorescence signal back to baseline. In contrast, when BSA is added to a suspension of pyra-nine-loaded vesicles prior to treatment with nigericin (Fig. 11,right panel), the fluorescence signal equilibrates at a levelbelow base line, consistent with an increase in vesicle pH. Thisbilirubin-induced pH gradient is dissipated by the addition ofnigericin, which allows the vesicle pH to drop back to base linelevels. These data indicate that flip-flop of bilirubin from theexternal to internal bilayer hemileaflet results in acidification,while movement of bilirubin out of the vesicle causes alkalin-ization of the entrapped volume, findings that are consistentwith transbilayer diffusion of the bilirubin diacid.

DISCUSSION

Our results indicate that unconjugated bilirubin is able todiffuse spontaneously through phospholipid bilayers. The first-order rate constant for bilirubin flip-flop (5.3 s21), although 40times slower than solvation from the membrane surface (19),exceeds the bilirubin dissociation rate from human and bovineserum albumin by over 5- and 50-fold, respectively (27). Thesefindings are consistent with the concept of dissociation-limiteduptake, originally proposed by Weisiger (35), in which therate-limiting step in bilirubin clearance is solvation from se-

FIG. 10. Effect of valinomycin onbilirubin flip-flop. The equilibration ofbilirubin (2 mM) between isolated ratbLPM (0.125 mg/ml protein) and cbBSA(20 mM) was monitored after preincubat-ing the membranes in the presence (1val) or absence (2 val) of valinomycin (10mM). Each curve reflects the average of sixstopped-flow injections performed at25 °C and is normalized to a scale of 0–1.When no valinomycin is present, the flu-orescence signal is best fit by a doubleexponential function (solid line), while asingle exponential best describes thecurve generated in the presence of vali-nomycin (dashed line). Similar resultsare obtained when the flip-flop of biliru-bin across small unilamellar phosphati-dylcholine vesicles (100 mM phospholipid)is examined under identical conditions(inset).

Transmembrane Diffusion of Bilirubin 10859

by guest on February 17, 2018http://w

ww

.jbc.org/D

ownloaded from

rum albumin. The value obtained for the flip-flop rate constantfor unconjugated bilirubin was determined from the fit of theequilibration data to Equation 2. While one cannot exclude theexistence of alternative mathematical models that may poten-tially yield different kinetic parameters, the model utilized inthe derivation of Equation 1 (28) is straightforward, and theparameters (koff

BSA, KaBSA/Ka

V) obtained from the fit of the datacorrespond closely with published values (27, 36).

Our results are discordant with the findings of Noy et al. (37)that the rate of bilirubin transmembrane diffusion exceedsbilirubin solvation from unilamellar dioleoylphosphatidylcho-line vesicles (k . 70 s21). The log-order discrepancy betweenthe rate constant for bilirubin flip-flop determined by theseauthors and the present study is unlikely to reflect differencesin the phospholipid content of the vesicles, since we demon-strate that the rate of bilirubin diffusion across dioleoylphos-phatidylcholine vesicles is the same as in vesicles composed ofphosphatidylcholine. However, Noy et al. derive the rate ofbilirubin flip-flop from a single exponential fit of bilirubinequilibration curves, while we found that bilirubin transfer isbest described by a double exponential function. This raises thepossibility that these authors may not have examined a longenough time course to identify a flip-flop signal or that theirfluorescence system was unable to resolve the two componentsof equilibration. Indeed, the value for the bilirubin dissociationrate determined by these authors is intermediate between the

off-rate and flip-flop rate constants obtained in our studies.Moreover, we have shown that it is imperative to examineequilibration rates over a range of phospholipid:albumin ratiosin order to determine the flip-flop rate constant (Equation 2).

Thermodynamic analyses indicate a moderate energy barrierto bilirubin flip-flop, composed primarily of an enthalpic com-ponent. Spectroscopic studies have shown that the negativelycharged propionate groups of bilirubin form ion pairs with thequaternary ammonium of sphingomyelin (38); hence, we pos-tulate that the activation enthalpy reflects the energy requiredto disrupt this carboxylate/quaternary ammonium ion interac-tion (since phosphatidylcholine and sphingomyelin possess anidentical head group). In contrast to long-chain fatty acids (32),unconjugated bilirubin traverses hepatocyte plasma mem-branes at a rate similar to that for phosphatidylcholine vesi-cles. This finding is consistent with the small entropic contri-bution to the free energy of activation, suggesting few steric ororientation constraints to bilirubin transmembrane diffusion.The lack of an effect of cholesterol, lipid packing, phospholipidacyl chain length, or saturation on the bilirubin flip-flop ratesupports this hypothesis. Based on these data, the plasmamembrane does not appear to pose a significant barrier to thediffusion of bilirubin into cells.

The proposition that bilirubin is able to freely enter all cellsof the body raises questions as to the mechanism underlyingthe specificity of bilirubin clearance by the liver. While our

FIG. 11. Bilirubin effects on vesicle pH. A representative series of experiments examining the effect of bilirubin (5 mM) on the fluorescenceof pyranine-loaded small unilamellar phosphatidylcholine vesicles (0.5 mM phospholipid) is displayed. In the left panel, vesicles were pretreatedwith nigericin (NIG), at a concentration of 1 mg/mg phospholipid. Injection of unconjugated bilirubin (UCB) induces a sharp decrease influorescence intensity (A), while the subsequent addition of 5 mM BSA causes an increase in pyranine fluorescence (D). When bilirubin is added topyranine-loaded vesicles in the absence of nigericin (middle panel), a lesser decline in fluorescence intensity (B) is observed, indicating acidificationof the entrapped vesicle volume, as supported by the abrupt decrease in fluorescence (A) induced by nigericin. Injection of BSA is again associatedwith a rapid rise in fluorescence to the level (D) previously observed. In the right panel, following the addition of bilirubin, the fluorescence signalagain settles at a level (B) consistent with acidification of the internal vesicle volume. Additionally, when BSA is added prior to nigericin, thefluorescence signal equilibrates below base line (C), indicative of alkalinization of the entrapped volume. Gradual dissipation of the proton gradientis observed before the addition of nigericin, which immediately collapses the pH potential and returns the fluorescence intensity to base line (D).

Transmembrane Diffusion of Bilirubin10860

by guest on February 17, 2018http://w

ww

.jbc.org/D

ownloaded from

studies do not exclude the presence of specific sinusoidal mem-brane transporter(s) that may be particularly important at lowplasma bilirubin concentrations, our findings are entirely con-sistent with a diffusional uptake mechanism for unconjugatedbilirubin. Despite the large body of kinetic evidence supportingthe existence of protein-mediated bilirubin transport and theidentification of several putative bilirubin transport proteins inliver cells (12, 39, 40), none of the candidate transporters hasbeen shown to facilitate bilirubin uptake. The only study todirectly examine bilirubin transport (by organic anion-trans-porting polypeptide) reveals no enhancement in bilirubin up-take by transiently transfected HeLa cells over nontransfectedcontrols (15). Evidence that bilirubin is transported by bili-translocase is limited to the induction of hyperbilirubinemia byinjection of polyclonal anti-bilitranslocase antibodies into rats(41), while support for a role for BSP/bilirubin-binding proteinis derived from partial inhibition of bilirubin uptake by pre-treatment of HepG2 cells (42) and cultured hepatocytes (10)with a monospecific antibody. These latter findings provide themost convincing proof of the existence of protein-mediated bil-irubin transport, although antibody-inhibitable uptake ac-counts for less than one-half of total uptake. We speculate thatthe noninhibitable component of bilirubin uptake representsspontaneous transmembrane diffusion and that high affinitybinding to glutathione S-transferase (the principal cytosolicbilirubin binding protein in the hepatocyte) and rapid bilirubinconjugation by UDP-glucuronosyltransferase provide sufficientdriving force for efficient bilirubin clearance by the liver (43).

Much of the kinetic data supporting the existence of specificbilirubin transporters are derived from studies that utilizehydrophilic organic anions, such as BSP and BDG, as surrogatemarkers for unconjugated bilirubin (10, 11, 44). While uncon-jugated bilirubin competitively inhibits BSP transport (10, 11),hepatocellular uptake of bilirubin can be dissociated from BSP(14, 15), supporting the presence of distinct transport mecha-nisms. We have further demonstrated a marked difference inthe uptake kinetics for unconjugated bilirubin and bilirubindiglucuronide in hepatic microsomes (Fig. 9), suggesting thatthese two compounds also are unlikely to share identical trans-port mechanisms, at least within this membrane compartment.These findings raise questions as to whether uptake kineticsfor hydrophilic organic anions can be appropriately extrapo-lated to unconjugated bilirubin.

Protein-mediated transport has been inferred from the in-creased uptake of unconjugated bilirubin by isolated rat baso-lateral liver plasma membranes in the presence of valinomycinand an inwardly directed potassium gradient (24). While thisfinding has been construed as evidence for electrogenic uptake,it is equally consistent with a model of bilirubin transportacross biologic membranes, originally proposed by Wennberg(5, 45), in which unconjugated bilirubin traverses the hydro-phobic membrane core as the uncharged diacid. We have shownthat bilirubin induces acidification of the entrapped volume ofphospholipid vesicles, supporting a mechanism whereby biliru-bin anions acquire a proton(s), diffuse through the lipid bilayer,and subsequently release hydrogen ions into the vesicle inte-rior (31). Since the dissipation of a transmembrane protongradient is limited by the slow counterflow of potassium ions(46), we propose that increased uptake of bilirubin by bLPM inthe presence of potassium and valinomycin results from rapiddissipation of a bilirubin-generated pH gradient, as opposed toelectrogenic transport of charged bilirubin species.

The UDP-glucuronosyltransferases are a family of enzymesthat convert hydrophobic endo- and xenobiotics (e.g, bilirubin,drugs) into water-soluble glucuronides. Topographical studiesof UDP-glucuronosyltransferase indicate that the active site is

oriented toward the lumen (47, 48) of the endoplasmic reticu-lum, and it has long been postulated that specific transportsystems exist to facilitate the movement of polar glucuronidesback across the microsomal membrane into the cytosol. Al-though uptake of bilirubin diglucuronide by hepatocyte baso-lateral and canalicular plasma membranes has been well char-acterized (49), our demonstration of rapid BDG transportacross microsomal membranes provides some of the first directevidence for the presence of a glucuronide transporter in theendoplasmic reticulum (50). Based on these findings, we pro-pose that unconjugated bilirubin gains access to active site ofUDP-glucuronosyltransferase via spontaneous flip-flop acrossthe microsomal membrane, while bilirubin glucuronide exportis carrier-mediated.

In contrast with newborns who are at significant risk for thedevelopment of kernicterus (bilirubin encephalopathy), adultsare resistant to the neurotoxic effects of bilirubin. The basis forthis observation has been presumed to reside in the lipid com-position of the adult blood-brain barrier, which renders it im-permeant to bilirubin. Our findings regarding the relative easewith which bilirubin traverses membranes composed of a vari-ety of lipid species shed doubt on this hypothesis. UDP-glucu-ronosyltransferase activity has been identified in brain mi-crovessels (51) and choroid plexus (52), and we speculate thatglucuronidation of bilirubin to form membrane-impermeantbilirubin conjugates is an essential mechanism to prevent bil-irubin entry into the adult brain. In the newborn, hepaticUDP-glucuronosyltransferase activity is extremely low, ap-proaching adult levels only after several days of life (53). As-suming that expression of UDP-glucuronosyltransferase inbrain capillary endothelial cells parallels that in the hepato-cyte, there exists a period during early postnatal developmentwhen the blood-brain endothelium is unable to effectively con-jugate bilirubin, permitting access of unconjugated bilirubin tothe central nervous system.

Acknowledgments—We gratefully acknowledge Dr. Richard Greenfor assistance with Western blot analyses, Emma Bootle for the per-formance of stopped-flow experiments, and Drs. David Lightner andMartin Carey for insightful criticism and support.

REFERENCES

1. Brodersen, R. (1979) J. Biol. Chem. 254, 2364–23692. Ostrow, J. D., Mukerjee, P., and Tiribelli, C. (1994) J. Lipid. Res. 35,

1715–17373. Bonnett, R., Davies, J. E., Hursthouse, M. B., and Sheldrick, G. M. (1978) Proc.

R. Soc. Lond. B Biol. Sci. 202, 249–2684. Hayward, D., Schiff, D., Fedunec, S., Chan, G., Davis, J. P., and Poznansky,

M. J. (1986) Biochim. Biophys. Acta 860, 149–1535. Wennberg, R. P. (1988) Pediat. Res. 23, 443–4476. Schmid, R. (1972) N. Engl. J. Med. 287, 703–7097. Scharschmidt, B. F., Waggoner, J. G., and Berk, P. D. (1975) J. Clin. Invest. 56,

1280–12928. Stollman, Y. R., Gartner, U., Theilmann, L., Ohmi, N., and Wolkoff, A. W.

(1983) J. Clin. Invest. 72, 718–7239. Paumgartner, G., and Reichen, J. (1976) Clin. Sci. Mol. Med. 51, 169–176

10. Stremmel, W., and Berk, P. D. (1986) J. Clin. Invest. 78, 822–82611. Kullak-Ublick, G. A., Hagenbuch, B., Stieger, B., Wolkoff, A. W., and Meier,

P. J. (1994) Hepatology 20, 411–41612. Sottocasa, G. L., Passamonti, S., Battiston, L., Pascolo, L., and Tiribelli, C.

(1996) J. Hepatol. 24, 36–4113. Berk, P. D., Potter, B. J., and Stremmel, W. (1997) Semin. Liver. Dis. 7,

165–17614. Gartner, U., Stockert, R. J., Levine, W. G., and Wolkoff, A. W. (1982) Gastro-

enterology 83, 1163–116915. Kanai, N., Lu, R., Bao, Y., Wolkoff, A. W., and Schuster, V. L. (1996) Am. J.

Physiol. 270, F319–F32516. Baldini, G., Passamonti, S., Lunazzi, G. C., Tiribelli, C., and Sottocasa, G. L.

(1996) Biochim. Biophys. Acta 856, 1–1017. Iga, T., Eaton, D. L., and Klaassen, C. D. (1979) Am. J. Physiol. 236, C9–C1418. Seppen, J., Tada, K., Hellwig, S., Bakker, C. T. M., Prasad, V. R., Roy-

Chowdhury, N., Roy-Chowdhury, J., Bosma, P. J., and Oude Elferink,R. P. J. (1996) Biochem. J. 314, 477–483

19. Zucker, S. D., Storch, J., Zeidel, M. L., and Gollan, J. L. (1992) Biochemistry 31,3184–3192

20. Barenholz, Y., Gibbes, D., Litman, B. J., Goll, J., Thompson, T. E., andCarlson, R. D. (1977) Biochemistry 16, 2806–2810

21. Zucker, S. D., Goessling, W., Zeidel, M. L., and Gollan, J. L. (1994) J. Biol.

Transmembrane Diffusion of Bilirubin 10861

by guest on February 17, 2018http://w

ww

.jbc.org/D

ownloaded from

Chem. 269, 19262–1927022. Kremer, J. M. H., Esker, M. W. J., Pathmamanoharan, C., and Wiersema,

P. H. (1977) Biochemistry 16, 3932–393523. Meier, P. J., and Boyer, J. L. (1990) Methods Enzymol. 192, 534–54524. Pascolo, L., del Vecchio, S., Koehler, R. K., Bayon, J. E., Webster, C. C.,

Mukerjee, P., Ostrow, J. D., and Tiribelli, C. (1996) Biochem. J. 316,999–1004

25. Anchordoguy, T. J., Carpenter, J. F., Crowe, J. H., and Crowe, L. M. (1992)Biochim. Biophys. Acta 1104, 117–122

26. Nogales, D., and Lightner, D. A. (1995) J. Biol. Chem. 270, 73–7727. Zucker, S. D., Goessling, W., and Gollan, J. L. (1995) J. Biol. Chem. 270,

1074–108128. Nichols, J. W., and Pagano, R. E. (1982) Biochemistry 21, 1720–172629. Blanckaert, N., Gollan, J. L., and Schmid, R. (1979) Proc. Natl. Acad. Sci.

U. S. A. 76, 2037–204130. Spivak, W., and Carey, M. C. (1985) Biochem. J. 255, 787–80531. Kamp, F., and Hamilton, J. A. (1992) Proc. Natl. Acad. Sci. U. S. A. 89,

11367–1137032. Kleinfeld, A. M., Chu, P., and Romero, C. (1997) Biochemistry 36,

14146–1415833. Kamp, F., and Hamilton, J. A. (1993) Biochemistry 32, 11074–1108634. Reed, R. G. (1977) J. Biol. Chem. 252, 7483–748735. Weisiger, R. A. (1985) Proc. Natl. Acad. Sci. U. S. A. 82, 1563–156736. Leonard, M., Noy, N., and Zakim, D. (1989) J. Biol. Chem. 264, 5648–565237. Noy, N., Leonard, M., and Zakim, D. (1992) Biophys. Chem. 42, 177–18838. Yang, B., Morris, M. D., Xie, M., and Lightner, D. A. (1991) Biochemistry 30,

688–694

39. Wolkoff, A. W. (1996) Semin. Liver Dis. 16, 121–12740. Berk, P. D., and Noyer, C. (1994) Semin. Liver Dis. 14, 331–34341. Sottocasa, G. L., Tiribelli, C., Luciani, M., Lunazzi, G. C., and Gazzin, B. (1979)

in Function and Molecular Aspects of Biomembrane Transport (Quagliariello,E., Palmieri, F., Papa, S., and Klinkenberg, M. eds) pp. 451–458, Elsevier/North-Holland Biomedical Press, Amsterdam

42. Stremmel, W., and Diede, H. E. (1990) J. Hepatol. 10, 99–10443. Tiribelli, C., and Ostrow, J. D. (1993) Hepatology 17, 715–73644. Torres, A. M., Lunazzi, G. C., Stremmel, W., and Tiribelli, C. (1993) Proc. Natl.

Acad. Sci. U. S. A. 90, 8136–813945. Bratlid, D. (1990) Clin. Perinatol. 17, 449–46546. Deamer, D. W., and Nichols, J. W. (1983) Proc. Natl. Acad. Sci. U. S. A. 80,

165–16847. Shepherd, S. R. P., Baird, S. J., Hallinan, T., and Burchell, B. (1989) Biochem.

J. 259, 617–62048. Jansen, P. L. M., Mulder, G. J., Burchell, B., and Bock, K. W. (1992) Hepatology

15, 532–54449. Oude Elferink, R. P. J., Meijer, D. K. F., Kuipers, F., Jansen, P. L. M., Groen,

A. K., and Groothuis, G. M. M. (1995) Biochim. Biophys. Acta 1241, 215–26850. Banhegyi, G., Braun, L., Marcolongo, P., Csala, M., Fulceri, R., Mandl, J., and

Benedetti, A. (1996) Biochem. J. 315, 171–17651. Ghersi-Egea, J. F., Minn, A., and Siest, G. (1988) Life Sci. 42, 2515–252352. Leininger-Muller, B., Ghersi-Egea, J. F., Siest, G., and Minn, A. (1994)

Neurosci. Lett. 175, 37–4053. Wishart, G. J. (1978) Biochem. J. 174, 485–48954. Zucker, S. D., and Hoppin, A. G. (1996) Hepatology 24, 132 (abstr.)55. Goessling, W., Zucker, S. D. (1997) Hepatology 26, 385 (abstr.)

Transmembrane Diffusion of Bilirubin10862

by guest on February 17, 2018http://w

ww

.jbc.org/D

ownloaded from

Stephen D. Zucker, Wolfram Goessling and Alison G. HoppinBilayers and Native Hepatocyte Membranes

Unconjugated Bilirubin Exhibits Spontaneous Diffusion through Model Lipid

doi: 10.1074/jbc.274.16.108521999, 274:10852-10862.J. Biol. Chem. 

  http://www.jbc.org/content/274/16/10852Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/274/16/10852.full.html#ref-list-1

This article cites 54 references, 16 of which can be accessed free at

by guest on February 17, 2018http://w

ww

.jbc.org/D

ownloaded from