~ ~~-~-------------

Use of a Standardised and ValidatedLong-term Human HepatocyteCulture System for Rep'etitive Analysesof Drugs: Repeated AClministrationsof Acetaminophen reduces Albumin andUrea SecretionAnett Ullrich', Christine Berg', Ian G. Hengstler? and Dieter Runge!IpRIMACYT Cell Culture Technology GmbH, FRG, Schwerin; 2Center for Toxicology, University Leipzig, FRG, Leipzig, Germany

SummaryHuman hepatocytes are the in vitro system of choice to studydrug-induced processes in man. Here, we present HEPAC2:

a standardised and validated culture system in whicli humanhepatocytes are maintained in HHMM (Human HepatocyteMaintenance Medium) with HGF (hepatocyte growth factor)and EGF (epidermal growth factor). Cellular viability andhepatocellular functions were monitored daily. Albumin andurea production remained on a relatively constant level for upto 2-3 weeks. Based on this, a standard protocol was estab-lished that allows repeated exposure of hepatocytes to studydrug metabolism. We used acetaminophen (AAP) to assaythe feasibility of this system. Hepatocytes were exposed to AAP(100-2815 mg/l)for 24 h. Subsequently, the culture medium wasreplaced by medium without AAP and the same exposurescenario was repeated at intervals of 4 days. High doses ofAAP (2815 mg/l) diminished urea production by 15-30% andalbumin secretion by 70-80%. These effects were reversible.After removal of AAp, secretion of urea and albumin returnedto control levels. AAP hepatotoxicity is caused by its bio trans-formation to the reactive metabolite Nsacetyl-p-beneoquinone-imine (NAPQl) mediated by CYP2El and CYP lA2. The AAPactivating enzymes were active for at least 21 days and theiractivity was maintained during at least four repeated cycles ofexposure to AAP In conclusion, these data demonstrate thesuitability of our long-term culture system to serve as a toolfor repetitive screening of drug-mediated changes of hepatocel-lular functions. This culture technique may help to overcomethe sparse availability of human hepatocytes for testing drug-media ted responses in man.

Zusammenfassung: Verwendung eines standardisierten undvalidierten Langzeit-Kultursystems für humane Hepatozyten zurrepetitiven Arzneimittelprüfung: Mehrfachgabe von Azetamino-phen reduziert die Albumin- und HamstoffsekretionHumane Hepatozyten sind das in vitro System der Wahl, wenn Me-dikamenten-induzierte Prozesse beim Menschen untersucht werdensollen. Mit HEPAC2 stellen wir hier ein standardisiertes und vali-diertes Zellkultursystem vor, in dem humane Hepatozyten inHHMM (Humanes Hepatozyten Erhaltungsmedium) mit HGF (He-patozyten Wachstumsfaktor) und EGF (Epidermaler Wachstums-faktor) kultiviert werden. Täglich wurden zelluläre Vitalität und le-berzelltypische Funktionen analysiert. Die Albumin- undHamstoff-Freisetzung blieb für 2-3 Wochen auf einem relativ kon-stanten Niveau. Darauf aufbauend wurde ein Standard protokollzum Studium von Wirkstoffen etabliert. Dazu werden die Hepato-zyten einer Kultur wiederholte Male diesen Wirkstoffen ausgesetzt.Als Modellsubstanz wurde Aretaminophen (AAP) eingesetzt, umdie Anwendbarkeit dieses Systems zu untersuchen. Hepatozytenwurden 24 Stunden in Gegenwart von AAP (l00-2815 mg/l) inku-biert. Anschließend wurde das Kulturmedium durchAAP-freies Me-dium ersetzt, und das Applikationsscenario wurde mehrfach. in Ab-ständen von 4 Tagen wiederholt. Hohe Konzentrationen von AAP(2815 mg/l) führten zu einer Reduktion der Harnstoff-Freisetzungum 15-30% und zu einer Verminderung der Albumin-Freisetzungum 75-80%. Beide Effekte waren reversibel. Nach Entfernung desAAP erreichten Harnstoff- und Albuminproduktion wieder die Kon-trollwerte. Die Hepatotoxizitdt des AAP wird durch die CYP2Elund CYP lA2 vermittelte Umwandlung zu N-Acetyl-p-Benzoquino-nimin (NAPQl) verursacht. Scheinbar waren die AAP-aktivieren-den Enzyme mindestens 21 Tage aktiv, und ihre Aktivität wurde über4 AAP-Expositionszyklen hinweg aufrechterhalten. Diese Daten be-legen die Einsetzbarkeit unserer Langzeitkulturen als Modellsystemzur repetitiven Analyse von Wirkstoff-induzierten Leberfunktions-änderungen beim Menschen. Dieses Kultursystem kann dazu bei-tragen, die geringe verfugbarkeit humaner Hepatozyten zu umge-hen und die Anzahl von Tierversuchen zu verringern.

Keywords: human hepatocytes, long-term cultures, acetaminophen, repetitive cycles of drug administration

Received 12 October 2006; received in final form and accepted for pub'ication 30 December 2006

ALTEX 24, 1/07 35

LINZ 2006 m.....-------------~-

human hepatocyte culture system inwhich hepatocyte morphology and func-tion is maintained for several weeks. Us-ing acetaminophen as a model substance,we are able to demonstrate the robust-ness of HEPAC2

, showing that repeatedexposure of the hepatocytes to the testsubstance led to a reversible reduction ofhepatocel1ular functions.

Urea synthesis serves as a parameter forthe catabolic process maintained by thehepatocytes. The pathway involves anumber of different enzymatic steps, ishighly specific for hepatocytes and can-not be performed by any other cells with-in the liver. An additional advantage isthat both products can easily be detectedin the cell culture medium without inter-fering with cellular viability. Further-more, the release of lactate dehydroge-nase activity was assayed as a marker forcellular vitality, and the cell cultures un-derwent a visual microscopic control.

1 Introduction

The liver is the centralorgan of energymetabolism, biotransformation and syn-thesis of plasma proteins under physiolog-ical and pathophysiological conditions.Primary hepatocytes are an importantmodel system to analyse and study liverspecific processes and functions. So far,most of the studies with hepatocyte cul-tures have primarily been performed withrat hepatocytes. However, interspeciesdifferences in all aspects of hepatocytefunction exist and have been recognisedand investigated for more than 30 years(for review see Runge et al., 2000a andreferences therein). In addition, from dataobtained in animal experiments one can-not always predict the effect any givendrug might have in man. Therefore, pri-mary human hepatocytes should be thesystem of choice for the evaluation of liv-er specific functions in humans, including(i) the biology of human viral liverpathogens or parasites and (ii) phase I and11 drug metabolism (Moshage and Yap1992, Li et al., 1997).Since human hepatocytes are available

only in limited number, the developmentof culture systems that allow cultivation ofdifferentiated and functional hepatocytesis of great importance. These human hepa-tocyte cultures should allow the establish-ment of screening systems for cytochromeP450 inducers and the investigation ofdrug-drug interactions. Given the fact thatthe read-out for these screening systemscan be non-invasive assays, like for testos-terone-6ß-hydroxylase (CYP3A4) orethoxyresorufin-deethylase (CYPIA), itshould be possible for such a culture sys-tem to allow repetitive treatment/wash-outcycles to screen for drug activity.Here we present HEPAC2

, a standard-ised and validated long-term serum-free

2 Characterisation,standardisation and validationof HEPAC2

Selection of parameters. Which parame-ters should be used to evaluate the func-tionality of a given human hepatocyteculture system? Usually, maintenance ofliver-specific functions is monitored bymeasuring the expression of serum pro-teins and/or drug metabolising enzymesby RT-PCR or Northern blot analysis forRNA detection and by Western blot anal-ysis for protein detection. However, thesemethods are invasive and require the dis-ruption of the cells. In addition, the qual-ity of these markers as parameters forhepatocyte specific function is arguablefor two reasons: First, the detection of acertain mRNA alone does not necessari-ly indicate the presence of the corre-sponding protein. Second, even if theprotein is detectable by Western blotanalysis, one cannot draw any conclusionon its functionality, because cofactors re-quired for enzymatic activities might bemissing (Runge et al., 2000b).Therefore, we decided to monitor hep-

atocellular function by tracking twoproducts that are synthesised and secre-ted by hepatocytes. We chose (i) albuminsynthesis as a marker for anabolic reac-tions performed by the hepatocytes, and(ii) the synthesis and excretion of urea.

3 Methods

Isolation and culture of human hepato-cytes: Human hepatocytes were isolatedfrom liver resections by three-step colla-genase perfusion (Strom et al., 1982;Strom et al., 1987). The viability of cellswas determined by trypan blue exclusiontest. Only hepatocyte preparations withmore than 70% viability were used in theexperiments described below. Cells wereplated onto collagen-coated 6-well-platesat a density of 106 cells per well usingMEM (Minimal Essential Medium con-taining 500 ng/rnl insulin and 500 ug/mlgentamycin). Unless otherwise stated,the medium was changed after 3 to 12 hto HHMM (Human Hepatocyte Mainte-nance Medium; Runge et al., 2000b),containing 10 ng/ml HGF and 20 ng/mlEGF. Medium was changed daily. Allstudies were performed in accordancewith an ethical survey of the ethics com-mission of the Ärztekammer Mecklen-burg-Vorpommern.

Application of Acetaminophen: AAP wasdissolved in HHMM containing 40%DMSO to obtain a 0.1 g/ml stock solu-

IncubationIncubationIncubation IncubationMe

days10 11 12 13 14 15 16 176 8 9o 1 2 3 5 74

Fig. 1: Experimental design for the implementation of a "recyclable" human hepatocyte culture system.

36 ALTEX 24, 1/07

m.... LINZ 2006---~----------------------------------------

tion. Human hepatocytes were incubatedwith AAP, 100-2815 mg/I, or DMSO forcontrol purposes for 24 h. The first incu-bation cyele was started on day 4 of theculture. After 24 h, medium was replacedby AAP-free culture medium and thehepatoeytes we~e eultured for 72 h in theabsence of AAP. This incubation cyelewas repeated every 4 days resulting in 24h incubation with AAP on days 8, 12, 16,and 20. The experimental design is out-lined in Figure 1.

Biochemical Assays: The eell eulturemedium was eolleeted every 24 h foranalysis of eatabolic (urea release) andanabolie processes (albumin seeretion)and for the determination of cellular via-bility (Lactate-dehydrogenase aetivity).LDH activity was detected using theAeroset e8000 system from Abbott(LDH-deteetion kit, 7D69-20). Urea re-lease was deteeted by a two step enzy-matie eonversion (urease / glutamate-de-hydrogenase) using the Aeroset c8000system from Abbott (Urea deteetion kit,7D75-20/-30). Albumin release into theeulture medium was determined using asandwich ELISA with antibodies ob-tained from Bethyl, using goat anti-hu-man albumin (Bethyl, A80-129A) aseoating antibody and goat anti-human al-

bumin-HRP eonjugate (Bethyl, A80-129P) as detection antibody.

4 Results

Cellular morphology: Cell morphologywas assayed by light and eleetron mi-eroseopy. It is evident that although hepa-toeytes are by far the most prominent eelltype, the eultures consist of a mixed pop-ulation of eells. The eells showed a typical

morphology of mature hepatoeytes for upto 24 days, mainly mono- and binueleareells were present (Fig. 2). Eleetron mi-eroseopy revealed round shaped nuelei;mitochondria and Golgi apparatus werewell established and maintained from ear-ly days in the eulture until at least day 24.At all time points, bile eanalieuli werereadily apparent. Glycogen and desmo-somes also developed during the culture,indicating metabolie activity as well aselose cell-cell eontacts.

Fig. 2: Toluidine Blue staining of human hepatocytes cultured for 24 days in HumanHepatocyte Maintenance Medium with 40 ng/ml HGF and 20 ng/ml EGF.A solution of 1% toluidine blue in 1% sodium borate was used to stain epoxide-embedded sections of cell cultures. Hepatocytes were the most prominent cell type.Endothelial- or fibroblast-like cells were detectable in the culture as weil (arrows).

12

10......::::.UI

i 8E::a....,~ 6>~ftI:c 4Q..I

2

0

r--'

~

\ II

\ I

I\ r T I

~""-

.xL ...•. ..... ... ..n I-- ;;: ".~"!:" ....- '" ~2 31 4 5 6 7 8 9 10 11 12 13 14 15

davs In culture

Fig. 3: Release of lactate dehydrogenase in human hepatocyte cultures.Human hepatocytes were cultured in Human Hepatocyte Maintenance Medium in the presence (blue line) or absence (green line) of HGFand EGF. Cell culture medium was collected every 24 hand assayed for LOH-activity. Oata are means ± SEM of 4 to 6 different donors.

ALTEX 24, 1/07 37

LINZ 2006 m.....-------------~-

Cellular vitality: Cellular integrity wasdetermined by measuring the enzymaticactivities of lactate dehydrogenase(LDH) in the culture medium. We moni-tored the release of LDH in hepatocytecultures in the presence or absence ofgrowth factors, Initially, within 24 h afterplating, LDH activity reached levels ofabout 8 umol/s/l, With ongoing culturethe daily release of LDH enzymatic ac-tivity declined to basal levels of approxi-mately 0.5 umol/s/l. This basal activitywas maintained from day 3 to day 15.The addition of growth factors did not al-ter the release of LDH into the culturemedium (Fig. 3).

Hepatocellular functions: Previously, weshowed that expression of phase I drugmetabolising enzymes like cytochromeP4503A4, lA2, 2El and others as weil ascorresponding testosterone-öß-hydroxy-lase activity (CYP3A4) were maintainedand/or remained inducible for more than 4weeks within these cultures (Runge et al.,2000b). Expression of liver specific tran-scription factors CIEBPa, HNF-3 andHNF-4 was maintained for several weeks(Runge et al., 2000c). These factors areheld responsible for the maintenance ofhepatocyte specific gene expression. Cy-

tochrome P450 3A4 protein expressionand corresponding testosterone-öß-hy-droxylase activity was maintained and/orremained inducible for up to 30 days(Runge et al., 2000b).Here, we determined the release of

urea as an example for catabolic functionof the hepatocytes. Initially, within 24 hafter plating, urea release reached levelsof about 3-4 mmol/1. Within the next 24h after plating, urea release dropped toabout 1.5 mmol/1. In the absence of HOFand EOF, urea release dropped further toabout 1 mmol/I within the next 24 h,reaching a new steady state that was sta-ble at least until day 15. In the presenceof growth factors, urea release was main-tained from day 2 on at approximately1.5 mmol/l (Fig. 4). The presence ofHOF and EOF seemed to improve ureasynthesis and release in human hepato-cytes. Statistical analysis using Student'st-test (two sided distribution, two sampIeequal variance) revealed a significantlyhigher urea release (p < 0.05) in HOFand EOF treated hepatocytes on days 6,11, and 13, compared to hepatocytes cul-tured in the absence of these growth fac-tors. In addition, we monitored albuminrelease as a parameter for hepatocellularanabolic activity. As seen with urea, al-

bumin release reached a steady state be-tween day 3 and 7 during the culture. Inthe absence of growth factors, albuminrelease declined from day 7 on. Again,the presence of HOF and EOF seemed toimprove hepatoceJlular functions, sincealbumin release did not decline but wasmaintained at the level reached at aroundday 3 (data not shown).

4.1 Implementation of HEPAC2

as a recyclable humanhepatocyte culture systemAfter having validated the functional sta-bility of our human hepatocyte culturesystem, we started to implement a proto-col that would take advantage of the ro-bustness of this model system: We cul-ture human hepatocytes in the presenceof HOF and EOF until day 4, a time pointby which the cells have adopted a newsteady state with regard to albumin andurea production. At day 4 the cells are in-cubated for 24 h with the substance of in-terest. Then at day 5 the active substanceis omitted and the hepatocytes are cul-tured for 3 days under standard condi-tions (wash out cycle). At day 8, a second24 h incubation cycle with the substanceof interest is introduced. Again, the nextday the foreign substance is omitted, and

s

......4:::::.

'0EE•....CI 3013::Il 2a.

~;:)1

01 2 3 4 5 6 7 8 9 lD U U 13 14 15

days in culture

Fig. 4: Release of urea in human hepatocyte cultures.Human hepatocytes were cultured in Human Hepatocyte Maintenance Medium in the presence (blue line) or absence (green line) of HGFand EGF. Cell culture medium was collected every 24 hand assayed for urea. Data are means ± SEM of 4 to 6 different donars. Statisticalanalysis using Student's t-test (two sided distribution, two sam pie equal variance) revealed a significantly higher urea release (p < 0.05) inHGF and EGF treated hepatocytes on days 6, 11, and 13, compared to hepatocytes cultured in the absence of these growth factars.

38 ALTEX 24, 1/07

~ ~~-~-------------

MP MP AAP AAP MP AAPr

~~'-l/---l c!- l~ --:::-..'"

~,1 ~ ~ I 1~,I "'" I

- - ~ r- - - -~

!- r-

-- r- r- r- - - - - r- r- r- r- r- r- - -

the cells are again cultured under controlconditions to wash out any remainingsubstance or product,As these assays used to determine (i)

metabolism of the active substance and(ii) its influence on hepatocellularmetabolism, are non-invasive and, as-suming that the active substance is non-toxic, this incubationlwash out cycle maybe repeated further until the hepatocyteslose the ability to metabolise the activesubstance (Fig. 1).

4.2 Acetaminophen as a modelsubstanceAAP, also known as paracetamol, is thefirst substance we used to evaluate ourculture system. In the liver, AAP isdetoxified via glucuronidation or sulpha-tion. AAP hepatotoxicity is caused by itsbiotransformation to the reactivemetabolite N-acetyl-p-benzoquinone-imine (NAPQI) mediated by CYP2Eland CYP1A2. Unless NAPQI is conju-gated with glutathione and subsequentlyexcreted, it may bind to proteins and

2,50

2,00

co.....Ci 1,50E!51m'!~

1,00:;:)

0,50

0,00

thereby cause hepatotoxicity. As shownin Figure 1,AAP was applied for 24 h onculture days 4, 8, 12, 16, and 20. At day4 high doses of AAP (2815 mg/l) dimin-ished urea production by approximately20% and albumin secretion by 70-80%.These effects were reversible. After re-moval of AAP, secretion of urea returnedto control levels within 24 hand albuminsecretion returned to normal levels with-in 72 h (Figs. 5, 6). Three days after re-moval of AAP, the hepatocytes wereagain incubated with AAP for 24 h.Again, AAP at a concentration of 2815mg/l led to a decrease in urea productionby 15-30% and albumin secretion by 70-80%. These effects could be repeatedseveral times, no matter whether AAPwas added on day 4 or after 2-3 weeks ofculture.

5 Summary and outlook

In this report we have presented HEPAC2:

a validated human hepatocyte culture

system that allows the serum-free long-term culture of hepatocytes. We haveimplemented an experimental designthat takes advantage of the robustnessof our culture system. This protocolallows repetitive analysis of drugs orother substances, as demonstrated herewith AAP as a first model substance.Therefore, HEPAC2 may serve as asuitable tool for repetitive screening ofdrug-mediated changes. A multi-centrestudy for external validation of HEPAC2

has been initiated.Given the stability of the culture sys-

tem with regard to the maintenance ofhepatocyte specific function and cy-tochrome P450 expression, HEPAC2

may also become an alternative to ani-mal experiments for testing long-termeffects of drugs and other chemicals.Currently, preclinical analyses of newdrugs require oral toxicity studies in ro-dents and non-rodents to evaluate thetoxic characteristics of a chemical. Stud-ies with rodents are designed accordingto OECD guidelines 407 (Repeated

4 5 6 7 8 9 W U U U U ß U U U U ~ IIU D N ~ Udays in culture

Fig. 5: Acetaminophen reduces urea release in human hepatocytes.Human hepatocytes were cultured in HHMM with growth factors. Medium was changed every 24 h. Urea was detected in cell culturemedium (control, black line). On the days indicated, AAP was added for 24 h. Urea release before, during and after incubation with AAPwas determined in the culture medium (red bars). Data represent means ± SD from two determinations of one representative donor out ofa total of four different donors,

ALTEX 24. 1/07 39

LINZ 2006 m.....-------------~-

40

3S

30

2S:::-......CI!oS 20.5E::JZIoe lS

10

S

0

AAP AAP AAP AAP AAP AAPT

)r/

<, /"Vr /~t -,r /

11 TA -.--I 1~ ...•.

~ \

~~ .I

1n-- I"

n nli- -

~

i- - - -

ndavs in culture

Fig. 6: Acetaminophen reduces albumin release in human hepatocytes.Human hepatocytes were cultured in HHMM with growth factors. Medium was changed every 24 h. Albumin was detected in cell culturemedium (control, black line). On the days indicated, AAP was added for 24 h. Albumin release before, during and after incubation withAAP was determined in the culture medium (red bars). Data represent means ± SD from two determinations of one representative donorout of a total of four different donors.

Dose 28-day Oral Toxicity Study in Ro-dents) and 408 (Repeated Dose 90-dayOral Toxicity Study in Rodents). Up to60 animals are required for each study.Often, the results obtained with these an-imal studies are not in line with data ob-tained from human studies. HEPAC2 al-lows us to culture differentiated humanhepatocytes for 28 days or longer. Thisshould give us a tool to analyse acute orrepeated dose effects that chemieals mayexert on human hepatocytes. If other hu-man cells can be maintained in a compa-rable fashion as weIl, it might be pos si-ble to reduce the number of animalstudies for the evaluation of repeateddose effects of chemicals.

ReferencesLi, A. P, Maurel, P, Gomez-Lechon, M.

1. et al. (1997). Preclinical evaluationof drug-drug interaction potential: pre-sent status of the application of prima-ry human hepatocytes in the evaluationof cytochrome P450 induction. Chem.Bio!. Interact. 107,5-16.

40

Moshage, H. and Yap , S. H. (1992). Pri-mary cultures of human hepatocytes: aunique system for studies in toxicolo-gy, virology, parasitology and liverpathophysiology in man. J. Hepato!. 3,404-413.

Runge, D., Michalopoulos, G. K., Strom,S. C. and Runge, D. M. (2000a). Re-cent advances in human hepatocyteculture. Biochem. Biophys. Res. Com-mun. 274, 1-3.

Runge, D., Köhler, c., Kostrubsky, V. E.et al. (2000b). Induction of cy-tochrome P450 (CYP)lAl, CYPIA2,and CYP3A4 but not of CYP2C9,CYP2CI9, multidrug resistance(MDR-l) and multidrug resistance as-sociated protein (MRP-l) by prototyp-ical inducers in human hepatocytes.Biochem. Biophys. Res. Commun. 273,333-341.

Runge, D., Runge, D. M., Jäger, D. et al.(2000c). Serum-free, long-term cul-tures of human hepatocytes: mainte-nance of cell morphology, transcrip-tion factors, and liver-specific

functions. Bioehem. Biophys. Res.Commun. 269,46-53.

Strom, S. c., Jirtle, R. L., Jones, R. S. etal. (1982). Isolation, culture, and trans-plantation of human hepatocytes. J.Nat!. Cancer Inst. 68,771-778.

Strom, S. C., Monteith, D. L., Manoha-ran, K. and Novotny, A. L. (1987).Genetic toxicology studies with humanhepatocytes. In E. J. Rauckman andG. M. Padilla (eds.), The isolated hep-atoeyte: use in toxieology and xenobi-otie biotransformation (265). Orlando,FL: Academic Press.

Correspondence toDieter RungePRIMACYT Cell Culture TechnologyGmbHHagenower Str. 7319061 SchwerinGermanyTel: +49-385-3993-600Fax: +49-385-3993-602e-mail: dieter.runge@primacyt.de

ALTEX 24, 1/07