chlorzoxazone hydroxylation in microsomes and hepatocytes from cytochrome p450 oxidoreductase-null...

7
J BIOCHEM MOLECULAR TOXICOLOGY Volume 23, Number 5, 2009 Chlorzoxazone Hydroxylation in Microsomes and Hepatocytes from Cytochrome P450 Oxidoreductase-Null Mice Li Li and Todd D. Porter Department of Pharmaceutical Sciences, College of Pharmacy, University of Kentucky, Lexington, KY, 40536-0082, USA; E-mail: [email protected] Received 9 February 2009; accepted 2 April 2009 ABSTRACT: Previous studies have demonstrated that the NADH-dependent cytochrome b 5 electron transfer pathway can support some cytochrome P450 monooxy- genases in vitro in the absence of their normal re- dox partner, NADPH-cytochrome P450 oxidoreduc- tase. However, the ability of this pathway to sup- port P450 activity in whole cells and in vivo re- mains unresolved. To address this question, liver mi- crosomes and hepatocytes were prepared from hep- atic cytochrome P450 oxidoreductase-null mice and chlorzoxazone hydroxylation, a reaction catalyzed primarily by cytochrome P450 2E1, was evaluated. As expected, NADPH-supported chlorzoxazone hy- droxylation was absent in liver microsomes from oxidoreductase-null mice, whereas NADH-supported activity was about twofold higher than that found in normal (wild-type) liver microsomes. This greater ac- tivity in oxidoreductase-null microsomes could be at- tributed to the fourfold higher level of CYP2E1 and 1.4- fold higher level of cytochrome b 5 . Chlorzoxazone hy- droxylation in hepatocytes from oxidoreductase-null mice was about 5% of that in hepatocytes from wild- type mice and matched the results obtained with wild- type microsomes, where activity obtained with NADH was about 5% of that obtained when both NADH and NADPH were included in the reaction mixture. These results argue that the cytochrome b 5 electron trans- fer pathway can support a low but measurable level of CYP2E1 activity under physiological conditions. C 2009 Wiley Periodicals, Inc. J Biochem Mol Toxicol 23:357–363, 2009; Published online in Wiley InterScience (www.interscience.wiley.com). DOI10:1002/jbt.20299 Correspondence to: Todd D. Porter. Contract Grant Sponsor: United States Public Health Service Grant (to Xinxin Ding) from the National Institute of Environmental Health Sciences, National Institutes of Health. Contract Grant Number: ES07462. c 2009 Wiley Periodicals, Inc. KEYWORDS: Cytochrome b 5 ; Cytochrome P450 Oxidore- ductase; Cytochrome P450 2E1; Chlorzoxazone; Hepatocytes INTRODUCTION Although cytochrome P450 oxidoreductase (POR) is the principal redox partner for the microsomal cy- tochromes P450, cytochrome b 5 has long been sus- pected to serve as an auxiliary source of electrons for this family of monooxygenases (reviewed in Porter [1]). These two redox pathways are readily distinguished by their strong preference for NADPH and NADH, re- spectively. A role for cytochrome b 5 was first suggested when it was noted that NADH stimulated microsomal P450 activity [2], and subsequent studies with antibod- ies to cytochrome b 5 [3,4] and to cytochrome P450 ox- idoreductase [5] provided support for this hypothesis. The ability of cytochrome b 5 to support P450 3A4 and P450 2E1 (CYP2E1) activities was confirmed with re- constituted systems lacking cytochrome P450 oxidore- ductase [6,7] and, for P450 2E1, in a bacterial expression system in vivo [8]. It should be noted that the ability of cytochrome b 5 to support or stimulate P450s is not uni- versal, with some forms exhibiting little or no stimula- tion; with many P450s it appears that cytochrome b 5 is able to donate only the second of the two required elec- trons, thus maintaining a requirement for cytochrome P450 oxidoreductase. Although the above studies demonstrated that cy- tochrome P450 2E1 could accept both required electrons from cytochrome b 5 and function in the absence of cy- tochrome P450 oxidoreductase, reconstituted systems are highly artificial and may not represent the normal physiological milieu of a cell. A similar criticism can be made for recombinant bacterial expression systems, where the selected enzymes are highly overexpressed, other mammalian enzymes are lacking, cofactor levels 357

Upload: li-li

Post on 11-Jun-2016

214 views

Category:

Documents


1 download

TRANSCRIPT

Page 1: Chlorzoxazone hydroxylation in microsomes and hepatocytes from cytochrome P450 oxidoreductase-null mice

J BIOCHEM MOLECULAR TOXICOLOGYVolume 23, Number 5, 2009

Chlorzoxazone Hydroxylation in Microsomesand Hepatocytes from Cytochrome P450Oxidoreductase-Null MiceLi Li and Todd D. PorterDepartment of Pharmaceutical Sciences, College of Pharmacy, University of Kentucky, Lexington, KY, 40536-0082, USA;E-mail: [email protected]

Received 9 February 2009; accepted 2 April 2009

ABSTRACT: Previous studies have demonstrated thatthe NADH-dependent cytochrome b5 electron transferpathway can support some cytochrome P450 monooxy-genases in vitro in the absence of their normal re-dox partner, NADPH-cytochrome P450 oxidoreduc-tase. However, the ability of this pathway to sup-port P450 activity in whole cells and in vivo re-mains unresolved. To address this question, liver mi-crosomes and hepatocytes were prepared from hep-atic cytochrome P450 oxidoreductase-null mice andchlorzoxazone hydroxylation, a reaction catalyzedprimarily by cytochrome P450 2E1, was evaluated.As expected, NADPH-supported chlorzoxazone hy-droxylation was absent in liver microsomes fromoxidoreductase-null mice, whereas NADH-supportedactivity was about twofold higher than that found innormal (wild-type) liver microsomes. This greater ac-tivity in oxidoreductase-null microsomes could be at-tributed to the fourfold higher level of CYP2E1 and 1.4-fold higher level of cytochrome b5. Chlorzoxazone hy-droxylation in hepatocytes from oxidoreductase-nullmice was about 5% of that in hepatocytes from wild-type mice and matched the results obtained with wild-type microsomes, where activity obtained with NADHwas about 5% of that obtained when both NADH andNADPH were included in the reaction mixture. Theseresults argue that the cytochrome b5 electron trans-fer pathway can support a low but measurable levelof CYP2E1 activity under physiological conditions.C© 2009 Wiley Periodicals, Inc. J Biochem Mol Toxicol23:357–363, 2009; Published online in Wiley InterScience(www.interscience.wiley.com). DOI 10:1002/jbt.20299

Correspondence to: Todd D. Porter.Contract Grant Sponsor: United States Public Health Service

Grant (to Xinxin Ding) from the National Institute of EnvironmentalHealth Sciences, National Institutes of Health.

Contract Grant Number: ES07462.c© 2009 Wiley Periodicals, Inc.

KEYWORDS: Cytochrome b5; Cytochrome P450 Oxidore-ductase; Cytochrome P450 2E1; Chlorzoxazone;Hepatocytes

INTRODUCTION

Although cytochrome P450 oxidoreductase (POR)is the principal redox partner for the microsomal cy-tochromes P450, cytochrome b5 has long been sus-pected to serve as an auxiliary source of electrons forthis family of monooxygenases (reviewed in Porter [1]).These two redox pathways are readily distinguishedby their strong preference for NADPH and NADH, re-spectively. A role for cytochrome b5 was first suggestedwhen it was noted that NADH stimulated microsomalP450 activity [2], and subsequent studies with antibod-ies to cytochrome b5 [3,4] and to cytochrome P450 ox-idoreductase [5] provided support for this hypothesis.The ability of cytochrome b5 to support P450 3A4 andP450 2E1 (CYP2E1) activities was confirmed with re-constituted systems lacking cytochrome P450 oxidore-ductase [6,7] and, for P450 2E1, in a bacterial expressionsystem in vivo [8]. It should be noted that the ability ofcytochrome b5 to support or stimulate P450s is not uni-versal, with some forms exhibiting little or no stimula-tion; with many P450s it appears that cytochrome b5 isable to donate only the second of the two required elec-trons, thus maintaining a requirement for cytochromeP450 oxidoreductase.

Although the above studies demonstrated that cy-tochrome P450 2E1 could accept both required electronsfrom cytochrome b5 and function in the absence of cy-tochrome P450 oxidoreductase, reconstituted systemsare highly artificial and may not represent the normalphysiological milieu of a cell. A similar criticism canbe made for recombinant bacterial expression systems,where the selected enzymes are highly overexpressed,other mammalian enzymes are lacking, cofactor levels

357

Page 2: Chlorzoxazone hydroxylation in microsomes and hepatocytes from cytochrome P450 oxidoreductase-null mice

358 LI AND PORTER Volume 23, Number 5, 2009

may differ from those in mammalian cells, and the lipidcomposition of the bacterial membrane is very differentfrom that of mammalian membranes. To address theselimitations, we have utilized the hepatic cytochromeP450 oxidoreductase-null (POR-null) mouse, in whichcytochrome P450 oxidoreductase expression in the liveris extinguished during maturation [9]. These mice ex-press little to no hepatic cytochrome P450 oxidoreduc-tase as adults and are being widely used to evaluatehepatic drug and xenobiotic metabolism [10–12]. Mi-crosomes and hepatocytes from the livers of these micemore closely resemble the true physiological milieu of acell in vivo than do reconstituted and recombinant sys-tems and provide an excellent venue to test the ability ofcytochrome b5 to support P450 activity in the absence ofcytochrome P450 oxidoreductase. To evaluate CYP2E1activity in these cells, we chose chlorzoxazone as a sub-strate; chlorzoxazone is widely used as a probe drug forCYP2E1 activity, and kinetic analysis of chlorzoxazonemetabolism in liver microsomes from mice revealedonly a single metabolite, 6-hydroxychlorzoxazone, andevidence for only a single-enzyme Michaelis–Mentenmodel with a relatively high rate of intrinsic clearance[13]. The studies described herein with POR-null hepa-tocytes demonstrate that, consistent with earlier in vitrostudies, the cytochrome b5 pathway can support a lowlevel of CYP2E1 activity in whole cells in the absenceof cytochrome P450 oxidoreductase.

MATERIALS AND METHODS

Chemicals

All chemicals were purchased from Sigma Chemi-cal Co. (St Louis, MO) unless otherwise indicated.

Animals

Cytochrome P450 oxidoreductase liver-specificknockout mice (Alb-Cre+/−/Cprlox+/+) were providedby Drs. Xinxin Ding and Jun Gu at the School ofPublic Health, State University of New York, Albany[9,14]. Littermates lacking the Alb-Cre transgene(Alb-Cre−/−/Cprlox+/+) were designated as wild type.Two to four months-old mice from null and wild-typelittermate groups on mixed C57BL/6 (75%) and 129/Sv(25%) genetic background were used in the studies. An-imals were maintained in a temperature-, humidity-,and light-controlled facility (70–72◦F, 48–52% humidity,12-h light/dark cycle) and were allowed free access towater and food. Animal-use protocols were approvedby the Institutional Animal Care and Use Committeesof the Wadsworth Center of the New York StateDepartment of Health and the University of Kentucky.

Chlorzoxazone Metabolism in Microsomes

Mice were killed by CO2 asphyxia, the liverspromptly removed, and microsomes prepared by ul-tracentrifugation. The microsomal fraction was resus-pended at ∼15 mg of protein/mL in 100 mM sodiumphosphate buffer and 1 mM EDTA, pH 7.6, andaliquoted and stored at −80◦C. Frozen microsomeswere thawed on ice immediately prior to use; all micro-somes used in these studies were frozen and thawedonce only. Protein content was determined with theCoomassie Plus assay reagent kit (Pierce Biotechnol-ogy Inc., Rockford, IL). Incubations were performed asdescribed [13] with minor modifications. Chlorzoxa-zone was dissolved in methanol, and appropriate vol-umes were added to disposable glass culture tubes andevaporated to dryness under reduced pressure in a vac-uum oven at room temperature. Phosphate buffer andthawed microsomes (400 μg of protein) were added,and after brief mixing the tubes were preincubated ina 37◦C agitating water bath. Reactions were started bythe addition of 100 μL of an NADPH-regenerating sys-tem (0.5 mM NADP+, 3.75 mM glucose-6-phosphate,and 1 unit/mL of glucose-6-phosphate dehydroge-nase) or 100 μL of an NADH-regenerating system(0.5 mM NAD+, 3.75 mM formic acid, and 1 unit/mLof formic acid dehydrogenase) to a final reactionvolume of 0.5 mL. After 20 min at 37◦C, the reac-tion was stopped by addition of 100 μL of 43% phos-phoric acid and immediately cooled on ice. Chlorzox-azone and its metabolite were assayed as describedbelow.

Chlorzoxazone Metabolism in PrimaryHepatocytes

Mice were anesthetized with urethane, and theliver was perfused via the portal vein with 50 mLof liver perfusion medium (Gibco, Invitrogen Corp.,Carlsbad, CA) followed by 30 mL of liver digestionmedium (Gibco). The liver was removed and trans-ferred into a tissue culture plate, and the liver capsulewas removed. Hepatocytes were suspended in hepato-cyte wash buffer (Gibco) and twice-filtered through adouble layer of gauze. The cells were pelleted, washedtwice, and resuspended in William’s E medium. Cellnumber and viability were determined by trypanblue exclusion. Hepatocytes were allowed to attach to35-mm tissue culture plates (BD Primaria™, BD Bio-sciences, San Jose, CA) in William’s E medium sup-plemented with 10% fetal bovine serum (InvitrogenCorp.), 4 mM L-glutamine, 100 units/mL penicillin,and 100 μg/mL streptomycin sulfate. After 3 h, thecells were washed in phosphate-buffered saline, themedium replaced with 2 mL of Hank’s balanced salt

J Biochem Molecular Toxicology DOI 10:1002/jbt

Page 3: Chlorzoxazone hydroxylation in microsomes and hepatocytes from cytochrome P450 oxidoreductase-null mice

Volume 23, Number 5, 2009 CHLORZOXAZONE METABOLISM IN POR-NULL LIVER 359

solution, and the cells were incubated with chlorzoxa-zone for 45 min. Cells were scraped from the platesand transferred with buffer into 10 mL glass tubesand disrupted by sonication for 1 min. Chlorzoxa-zone and its metabolite were measured as describedbelow.

Measurement of 6-Hydroxychlorzoxazone

Five mg of phenacetin was added as an internalstandard to the denatured microsome or hepatocytereaction mixtures, which were then extracted twicewith 1 mL of ethyl acetate and the organic layerstransferred to a microcentrifuge tube and evaporatedunder a stream of nitrogen. Samples were redissolvedin 120 μL of “mobile phase” (50 mM KH2PO4 inwater:acetonitrile (75:25 v/v)), vortexed, and cen-trifuged for 15 min at 16,000 rpm. The supernatewas then transferred to vials for high-performanceliquid chromatographic analysis. The chromatographyapparatus (Shimazu, Japan) consisted of a C18 column(μ-Bondapack, 300 mm × 3.9 mm I. D., Waters Corp.,Milford, MA) with mobile phase set at a flow rate of1.0 mL/min. Eluants were monitored by ultravioletabsorption at 295 nm using a variable-wavelengthdetector. Retention times for 6-hydroxychlorzoxazone,phenacetin, and chlorzoxazone were approximately7.03, 13.29, and 19.56 min, respectively. Product peakidentity was verified by demonstrating coelution withauthentic 6-hydroxychlorzoxazone and by its absenceupon exclusion of cofactors or microsomes from theincubation mixture. For each analysis, a calibrationcurve was prepared using a series of concentrations ofpure 6-hydroxychlorzoxazone dissolved in the mobilephase with the internal standard.

Gel Electrophoresis and Immunoblotting

For immunoquantitation, twofold serial dilutionsof wild-type and POR-null liver microsomes (20–2.5 μg of protein) were fractionated by sodium dodecylsulfate-polyacrylamide gel electrophoresis on 10%gels (CYP2E1 and POR) or 15% gels (cytochrome b5),and then electroblotted to nitrocellulose membranes(Bio-Rad Laboratories, Inc.). The membrane wasblocked with 0.05% tween-20 and 5% defatted milkand then incubated in this same buffer with goatantibodies to CYP2E1 from rat (Daiichi Pure Chemi-cals, Tokyo, Japan), cytochrome P450 oxidoreductasefrom rabbit (a gift from Dr. M. J. Coon, Universityof Michigan), or cytochrome b5 from rabbit (OxfordBiomedical Research, Oxford, MI). The immunoblotswere developed with a secondary antibody conjugatedto horseradish peroxidase (Sigma) and visualized by

chemiluminescence (Supersignal West Pico Chemilu-minescent Substrate, Pierce Biotechnology, Inc.) andquantified on a Kodak Image Station MM2000. Imageswere acquired with 2-min exposures and the signalsobtained with POR-null microsomes were fitted tostandard curves generated from wild-type microsomeswith background subtraction.

Data Analysis

The formation of 6-hydroxychlorzoxazone was fitto a single-enzyme Michaelis–Menten model usingnonlinear regression: V = VmaxS/(Km + S), where Vis the reaction velocity, S is the substrate concentra-tion, Vmax is the maximum reaction velocity, and Kmis the substrate concentration corresponding to 50%Vmax. Data points were fitted to a line using non-linear regression with the computer program PRISM(GraphPad, San Diego, CA). The differences betweenthe cytochrome P450 oxidoreductase-null group andthe wild-type group were compared using a one-wayANOVA, setting p < 0.05.

RESULTS

We utilized chlorzoxazone as a probe for CYP2E1activity in microsomes from wild-type and cytochromeP450 oxidoreductase-null mouse liver microsomes, asshown in Figure 1. As expected, the Vmax for chlor-zoxazone hydroxylation with NADH in wild-type mi-crosomes was very low (approximately 10% of thatobtained with NADPH, Table 1), whereas addition ofNADH to NADPH synergistically increased the Vmax toa level nearly twice that obtained with NADPH alone.In contrast, with microsomes from POR-null livers ac-tivity with NADPH was undetectable, but with NADHchlorzoxazone was hydroxylated at a rate twice that ob-tained with NADH in wild-type microsomes. No syn-ergistic effect was seen when NADH and NADPH werecombined, and the small increase in activity over thatseen with NADH alone was not statistically significant.Km and Vmax values were within the range of values re-ported previously for chlorzoxazone hydroxylation byCYP2E1 in mice and other species [13,15].

The twofold higher rate of chlorzoxazone hydrox-ylation with NADH in POR-null liver microsomesas compared to wild-type microsomes suggested thatCYP2E1 or cytochrome b5 expression might be higherin POR-null liver. Indeed, immunochemical quantita-tion of these proteins revealed a fourfold increase inCYP2E1 and a 40% increase in cytochrome b5 (Figure 2).The increase in CYP2E1 is similar to the twofold in-crease reported by Gu et al. [9]; cytochrome b5 levels

J Biochem Molecular Toxicology DOI 10:1002/jbt

Page 4: Chlorzoxazone hydroxylation in microsomes and hepatocytes from cytochrome P450 oxidoreductase-null mice

360 LI AND PORTER Volume 23, Number 5, 2009

Wild-type

0 250 500 750 10000.0

0.5

1.0

1.5

2.0

2.5

NADH

NADPH

NADPH + NADH

A

[Chlorzoxazone] μ( M)

6O

H-C

LZ F

orm

ati

on

(nm

ol/m

in/m

g p

rote

in)

POR-null

0 250 500 750 1000

0.1

0.2

0.3

0.4

0.0

NADH

NADPH + NADH

B

H-C

LZ F

orm

ati

on

ol/m

in/m

g p

rote

in)

[Chlorzoxazone] μM)(

6O

(nm

NADPH

FIGURE 1. Chlorzoxazone hydroxylation in microsomes from wild-type and POR-null mice. Pooled liver microsomes from wild-type (A) orPOR-null (B) mice (n = 6) were incubated with the indicated concentrations of chlorzoxazone for 20 min at 37◦C in the presence of 3.7 mM NADPHand/or 3.7 mM NADH and 6-hydroxychlorzoxazone (6OH-CLZ) formation was measured by high-performance liquid chromatography. Assayswere carried out in duplicate, and values represent the mean ± SE. Lines were fit using PRISM.

TABLE 1. Kinetic Analysis of NADH-Dependent and NADPH-Dependent Metabolism of Chlorzoxazone in Wild-Type andPOR-Null Mouse Liver Microsomes

Vmax (percent ofwild-type with NADH + NADPH)

Liver Cofactor(s) Km (chlorzoxazone) (μM) (nmol/min/mg protein)

Wild-type NADH 24.01 ± 4.99 0.114 ± 0.005 (5%)NADPH 13.01 ± 2.28 1.278 ± 0.039 (55%)

NADH + NADPH 28.16 ± 3.61 2.315 ± 0.063 (100%)POR-null NADH 61.07 ± 18.19 0.245 ± 0.020 (11%)

NADPH —a —a

NADH + NADPH 33.68 ± 7.93 0.330 ± 0.017 (14%)

Data represent mean ± SE of duplicate determinations, from Figure 1.a Insufficient activity for determination.

in POR-null mouse liver have not been reported previ-ously. Antibody to cytochrome b5 was able to decreasechlorzoxazone hydroxylation by up to 25%, but wasnot able to block activity completely (Figure 3). Fur-ther increases in antibody did not increase the extentof inhibition (data not shown). Cytochrome P450 ox-

idoreductase expression was below the limits of de-tection by immunoblotting in liver microsomes fromPOR-null mice (data not shown).

Hepatocytes more closely replicate in vivo phys-iology than do microsomes, and so chlorzoxa-zone metabolism was measured in freshly isolated

J Biochem Molecular Toxicology DOI 10:1002/jbt

Page 5: Chlorzoxazone hydroxylation in microsomes and hepatocytes from cytochrome P450 oxidoreductase-null mice

Volume 23, Number 5, 2009 CHLORZOXAZONE METABOLISM IN POR-NULL LIVER 361

CYP2E1

WT POR-null0

100

200

300

400

500

Rela

tive

amo

un

t o

fC

YP

2E

1 (

%)

Cyt b5

WT POR-null0

50

100

150

200

Rela

tiv

e am

ou

nt

of

b5 (

%)

FIGURE 2. Quantitative analysis of CYP2E1 and cytochrome b5 pro-tein levels in POR-null microsomes. Serial dilutions of wild-type andPOR-null microsomes were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis, electroblotted to nitrocellulosemembranes, and CYP2E1 (n = 8) and cytochrome b5 (n = 4) levelswere determined by immunodetection. WT, wild-type. A representa-tive gel image is presented below each graph, with 0.5 μg (CYP2E1)or 5 μg (cytochrome b5) of microsomal protein.

hepatocytes from wild-type and POR-null mice. Asshown in Figure 4, chlorzoxazone hydroxylation couldbe observed in POR-null hepatocytes at a rate approxi-mately 5% of that found in wild-type cells. This matchesthe ratio of activity found in wild-type microsomeswhen comparing NADH-catalyzed activity to that ob-tained with NADH and NADPH, but is somewhat less

Untreat

ed Ab

5g b

μ0.

8

Ab

5g b

μ1.

6

0.0

0.1

0.2

0.3

0.4

6O

H-C

LZ

Fo

rmati

on

(nm

ol/m

in/m

g p

rote

in)

FIGURE 3. Antibody to cytochrome b5 inhibits chlorzoxazone hy-droxylation in POR-null microsomes. Chlorzoxazone hydroxylationwas measured in POR-null microsomes in the presence of 3.7 mMNADH and goat polyclonal antibody to rabbit cytochrome b5. Anti-body was added to the incubations 30 min prior to the addition ofNADH to start the reaction. Each value represents the mean ± SE ofat least two determinations.

Hepatocytes

0 200 400 600 800 10000.0

0.2

0.4

0.6

0.8

1.0

1.2

Wild-type

POR-null

[Chlorzoxazone] μM)(

6O

H-C

LZ

Fo

rma

tio

n

(nm

ol/

h/5

10

5 ce

lls

)

FIGURE 4. Chlorzoxazone hydroxylation in hepatocytes from wild-type and POR-null mice. Incubations were carried out in duplicatein 2 mL of Hank’s balanced salt solution with indicated concen-trations of chlorzoxazone for 45 min, after which the cells werescraped from the plates and transferred with buffer into 10 mL glasstubes and disrupted by sonication for 1 min. Chlorzoxazone and6-hydroxychlorzoxazone (6OH-CLZ) were measured by high per-formance liquid chromatography. Lines were fit using PRISM.

than what might be expected from comparison of ac-tivity in POR-null microsomes to wild-type prepara-tions, where maximal activity in POR-null microsomeswas 10%–15% of that found in wild-type microsomes(Table 1).

DISCUSSION

While the ability of the cytochrome b5 electrontransport pathway to support some P450s in vitro in theabsence of cytochrome P450 oxidoreductase is well es-tablished, the extent to which this pathway contributesto P450 activity in vivo is less clear. We have utilized thehepatic conditional POR-null mouse [9] to explore thisquestion, taking advantage of the lack of cytochromeP450 oxidoreductase in microsomes and hepatocyteswhich otherwise are largely representative of a normalphysiological state. Our results confirm the ability ofthe b5 pathway (or an alternative, unidentified NADH-dependent pathway) to support CYP2E1 activity witha relatively low efficiency in the absence of cytochromeP450 oxidoreductase in microsomes and hepatocytesfrom these mice and are the first demonstration of thefunctionality of this pathway in an intact mammaliancell.

CYP2E1 is unusual in that it is one of only twoP450s (along with CYP3A4) that has been shown to beable to accept both the first and second electron fromthe b5 pathway in vitro. Although a number of otherP450s have been shown in vitro to be stimulated by the

J Biochem Molecular Toxicology DOI 10:1002/jbt

Page 6: Chlorzoxazone hydroxylation in microsomes and hepatocytes from cytochrome P450 oxidoreductase-null mice

362 LI AND PORTER Volume 23, Number 5, 2009

presence of cytochrome b5, with these P450s electrontransfer from b5 appears to be limited to the second re-quired electron, and thus they would not be expectedto be catalytically active in the absence of cytochromeP450 oxidoreductase. Indeed, with some P450s thestimulation afforded by cytochrome b5 appears to bestrictly allosteric and does not involve electron trans-fer. However, the majority of P450s have not been testedfor activity with cytochrome b5 in the absence of cy-tochrome P450 oxidoreductase, and so it remains pos-sible that this alternative redox pathway may be ableto support additional P450s in vivo.

How important is this pathway in vivo whencytochrome P450 oxidoreductase is present? Finn et al.[15] very recently reported that the hepatic deletion ofmicrosomal cytochrome b5 significantly decreased theactivity of a variety of cytochromes P450, includingCYP2E1, both in vitro and in vivo. Most importantly,NADH-dependent metabolism by a number of P450swas significantly reduced or eliminated in microsomesfrom b5-null mice, providing strong evidence thatelectron transfer through cytochrome b5 is functionallyrelevant. Our studies with POR-null hepatocytesdemonstrate that the cytochrome b5 pathway canserve as the sole electron source for cytochrome P4502E1 in an intact cell, and, moreover, suggest that themetabolic state of the hepatocyte is likely to be a signif-icant determinant of NADH-cytochrome b5-mediatedactivity. Chlorzoxazone hydroxylation in POR-nullhepatocytes, which must be attributed entirely to elec-tron transfer through the NADH-dependent pathway,was about 5% of that found in wild-type hepatocytes.However, chlorzoxazone hydroxylation in POR-nullmicrosomes was 11%–14% of that found in wild-typemicrosomes (Table 1). Thus, POR-null hepatocytesmetabolize chlorzoxazone at no more than 50% of theprojected rate, based on these microsomal studies. Oneexplanation for this lower rate of NADH-dependentmetabolism in whole cells may be the level of thecofactors NADH and NADPH; these are set at saturat-ing levels in microsomal assays, but this may not betrue in isolated hepatocytes. Competing physiologicalprocesses that utilize NADH, as well as the rate ofintracellular regeneration of NADH, may all contributeto subsaturating NADH levels and, consequently, thelower activity found in POR-null cells. These samefactors are likely to apply to both wild-type cells and invivo and suggest that NADH-cytochrome b5-mediateddrug metabolism is likely to be dependent on themetabolic state of the hepatocyte. In those situationswhere NADH levels are high, and/or where NADPHlevels are limiting, the b5-mediated redox pathwaymay be able to augment CYP2E1, and perhaps otherP450 activity; one such situation might be duringethanol intoxication, where ethanol oxidation by

alcohol and acetaldehyde dehydrogenases generateshigh levels of NADH [16]. It is of interest that CYP2E1may contribute to ethanol metabolism, and alcoholicliver injury, in these circumstances [17].

ACKNOWLEDGMENTS

We thank Drs. Xinxin Ding and Jun Gu, WadsworthCenter, New York State Department of Health, for pro-viding the mice used in these studies. We thank Dr.Markos Leggas for the use of his high performance liq-uid chromatography system and Dr. Jamie Horn for herguidance and assistance on the use of that instrument.

REFERENCES

1. Porter TD. The roles of cytochrome b5 in cytochrome P450reactions. J Biochem Mol Toxicol 2002;16(6):311–316.

2. Hildebrandt A, Estabrook RW. Evidence for the par-ticipation of cytochrome b 5 in hepatic microsomalmixed-function oxidation reactions. Arch Biochem Bio-phys 1971;143(1):66–79.

3. Mannering GJ, Kuwahara S, Omura T. Immunochemi-cal evidence for the participation of cytochrome b5 inthe NADH synergism of the NADPH-dependent mono-oxidase system of hepatic microsomes. Biochem BiophysRes Commun 1974;57(2):476–481.

4. Noshiro M, Harada N, Omura T. Immunochemical studyon the participation of cytochrome b5 in drug oxidationreactions of mouse liver microsomes. Biochem BiophysRes Commun 1979;91(1):207–213.

5. Kuwahara S, Mannering GJ. Evidence for a pre-dominantly NADH-dependent O-dealkylating sys-tem in rat hepatic microsomes. Biochem Pharmacol1985;34(24):4215–4228.

6. Yamazaki H, Nakano M, Gillam EM, Bell LC, GuengerichFP, Shimada T. Requirements for cytochrome b5 inthe oxidation of 7-ethoxycoumarin, chlorzoxazone, ani-line, and N-nitrosodimethylamine by recombinant cy-tochrome P450 2E1 and by human liver microsomes.Biochem Pharmacol 1996;52(2):301–309.

7. Yamazaki H, Nakano M, Imai Y, Ueng YF, GuengerichFP, Shimada T. Roles of cytochrome b5 in the oxidation oftestosterone and nifedipine by recombinant cytochromeP450 3A4 and by human liver microsomes. Arch BiochemBiophys 1996;325(2):174–182.

8. Mokashi V, Li L, Porter TD. Cytochrome b5 reductaseand cytochrome b5 support the CYP2E1-mediated acti-vation of nitrosamines in a recombinant Ames test. ArchBiochem Biophys 2003;412(1):147–152.

9. Gu J, Weng Y, Zhang QY, Cui H, Behr M, Wu L, Yang W,Zhang L, Ding X. Liver-specific deletion of the NADPH-cytochrome P450 reductase gene: impact on plasmacholesterol homeostasis and the function and regulationof microsomal cytochrome P450 and heme oxygenase.J Biol Chem 2003;278(28):25895–25901.

10. Gu J, Cui H, Behr M, Zhang L, Zhang QY, Yang W,Hinson JA, Ding X. In vivo mechanisms of tissue-selective drug toxicity: effects of liver-specific knock-out of the NADPH-cytochrome P450 reductase gene on

J Biochem Molecular Toxicology DOI 10:1002/jbt

Page 7: Chlorzoxazone hydroxylation in microsomes and hepatocytes from cytochrome P450 oxidoreductase-null mice

Volume 23, Number 5, 2009 CHLORZOXAZONE METABOLISM IN POR-NULL LIVER 363

acetaminophen toxicity in kidney, lung, and nasalmucosa. Mol Pharmacol 2005;67(3):623–630.

11. Fang C, Behr M, Xie F, Lu S, Doret M, Luo H, Yang W,Aldous K, Ding X, Gu J. Mechanism of chloroform-induced renal toxicity: non-involvement of hepatic cy-tochrome P450-dependent metabolism. Toxicol ApplPharmacol 2008;227(1):48–55.

12. Henderson CJ, Pass GJ, Wolf CR. The hepatic cytochromeP450 reductase null mouse as a tool to identify a success-ful candidate entity. Toxicol Lett 2006;162(1):111–117.

13. Court MH, Von Moltke LL, Shader RI, Greenblatt DJ.Biotransformation of chlorzoxazone by hepatic micro-somes from humans and ten other mammalian species.Biopharm Drug Dispos 1997;18(3):213–226.

14. Wu L, Gu J, Weng Y, Kluetzman K, Swiatek P, Behr M,Zhang QY, Zhuo X, Xie Q, Ding X. Conditional knockoutof the mouse NADPH-cytochrome p450 reductase gene.Genesis 2003;36(4):177–181.

15. Finn RD, McLaughlin LA, Ronseaux S, Rosewell I,Houston JB, Henderson CJ, Wolf CR. Defining the in vivorole for cytochrome b5 in cytochrome P450 functionthrough the conditional hepatic deletion of microsomalcytochrome b5. J Biol Chem 2008;283(46):31385–31393.

16. Lieber CS. Metabolism of alcohol. Clin Liver Dis2005;9(1):1–35.

17. Lu Y, Cederbaum AI. CYP2E1 and oxidative liverinjury by alcohol. Free Radic Biol Med 2008;44(5):723–738.

J Biochem Molecular Toxicology DOI 10:1002/jbt