Cytochrome P450IIIA activity and cytokine-mediated synthesis of nitric oxide

Cytochrome P450IIIA activity and cytokine-mediated synthesis of nitric oxide Paul C. Kuo,

MD, a

Keith Y. Abe, BS, and Donald C. Dafoe, MD, Stanford, Calif.

Background. Although nitric oxide synthase (NOS) is a cytochrome P450-like hemoprotein with additional sequence homology to cytochrome P450 reductase, the role of the cytochrome P450 system in cytokine-mediated NO synthesis is unknown. Methods. To characterize the role of the P450 system in the synthesis of NO, NO production, NOS enzyme activity, and steady state NOS mRNA and protein expression were characterized in the setting of P450 isoform activity inhibition by using a model of isolated rat hepatocytes in primary culture. Cimetidine (0 to I0 mmol/L) was chosen as a specific inhibitor of P45OIIIA activity. NO production was induced by interleukin-1 (50 ng/ml) and tumor necrosis factor (500 units/ml) and quantified by measurement of its metabolite, nitrite, in the culture medium. Steady state NOS mRNA and protein expression were determined by reverse-transcriptase polymerase chain reaction and immunoblot analysis, respectively. NOS enzyme activity was measured by the conversion of tritiated-L-arginine to tritiated-L-citruUine. Results. Inhibition of P450IIIA activity was associated with a concentration-dependent decrease in cytokine-mediated NO production. Levels of NOS mRNA and protein were not altered. The NOS enzyme assay was notable for stable concentrations of intermediate, N-OH-L-arginine, and decreased production of the final end product, L-citrulline. Dixon plot kinetic analysis of cimetidine-mediated inhibition of NOS yielded an inhibition constant Ki = 1.76 mmol/L. Conclusions. These results indicate that cytochrome P450IIIA isoform may play a posttranslational role in cytokine-mediated NO synthesis in this model of isolated rat hepatocytes in primary culture. (SURCEnY 1995;118.'310-7.) From the Division of Transplantation Surgery, Department of Surgery, Stanford University Medical Center, Stanford, Calif.

Nrrmc OXIDE (NO) is a ubiquitous multifunctional free radical that is synthesized by five-electron oxidation o f the substrate L-arginine. 1 T h e enzyme, NO synthase (NOS), catalyzes this reaction in the presence o f the cofactors, r e d u c e d nicotinamide a d e n i n e dinucleotide p h o s p h a t e (NADPH), molecular oxygen, and tetrahyd r o b i o p t e r i n (BH4), to form the intermediate. N-OH-Larginine, with subsequent generation of" NO a n d its cop r o d u c t L-citrulline. 1 NOS is a monooxygenase with substantial sequence h o m o l o g y to cytochrome P450 reductase a n d has b e e n shown to have enzyme activities similar to those of cytochrome P450 reductase, including reduction o f cytochrome P450. 2,3 Given these observations a n d the striking similarity between N O formation a n d P450-mediated oxidation reactions, we Presented at the Fifty-sixthAnnual Meeting of The Societyof University Surgeons. Denver. Colo.. Feb. 9-11. 1995. Reprint requests: Paul C. Kuo. MD. Department of Surgery, MSOB X300, Stanford UniversityMedical Center. Stanford, CA 94305. aRecipientof the American Societyof Transplant Surgeons-OrthoFaculty Development Award. Copyright 9 1995 by Mosby-YearBook. Inc. 003943060/95/$3.00 + 0 11/6/64961 310

SURGERY

sought to characterize the role of the cytochrome P450 system in a m o d e l o f isolated rat hepatocytes. In particular, having previously shown that inhibition o f CYP450IIIA was specifically associated with altered cytokine-mediated N O synthesis, the role of the P450IIIA isoform was d e t e r m i n e d in the settings o f hepatocyte N O production, N O S enzyme activity, a n d steady state NOS mRNA and p r o t e i n expression. 4 In the p r e s e n t study we demonstrate that the P450IIIA isoform regulates cytokine-mediated hepatocyte N O synthesis by decreasing conversion of the intermediate, N-OH-L-arginine, to the final e n d products, N O a n d L-citrulline.

METHODS Rat h e p a t o c y t e isolation. Male Lewis rats (200 to 300 gin: H a r l a n Inc., Indianapolis, Ind.) were used for hepatocyte isolation with the technique o f in situ collagenase-D perfilsion a n d Percoll g r a d i e n t centrifugation. 5 Hepatocytes were r e s u s p e n d e d in Williams' E m e d i u m ) with 1 m m o l / L L-arginine, 1 p m o l / L insulin, 15 m m o | / L HEPES (pH 7.4), penicillin, streptomycin, and 10% fetal calf serum. Purity was assessed by leukocyte esterase staining, a n d viabilitywas assessed by trypan blue exclusion. Preparations were routinely greater

Surgery Volume 118, Number 2 than 90% viable and greater than 98% pure. The cell suspension was plated at a density of 5.0 x 105 cells/ml onto collagen-coated wells. Generation of NO. Hepatocytes were stimulated by incubation in Williams' E medium with 10% heat inactivated fetal calf serum and 1 m m o l / L L-arginine supplemented with 50 n g / m l interleukin-1 (IL-1) and 500 units/ml tumor necrosis factor-(TNF-[3). 6 In selected instances the cytochrome P450IIIA inhibitor cimetidine was added. 7 After incubation for 18 hours at 37 ~C in 95% 02/5% CO2 the supernatant was aspirated and the cells were washed twice with Dulbecco's phosphate-buffered saline solution, Measurement of NO. NO was quantified by measurement of the metabolite, nitrite. 8 Conditioned medium was mixed with 1% sulfanilamide in 0.5N HC1 (50% volume/volume). After a 5-minute incubation at room temperature an equal volume of 0.02% naphthylenediamine was added. After incubation at room temperature for 10 minutes, absorbance at 570 nm was compared with an NaNO2 standard. Determination of hepatocyte microsomal monooxygenase activities. Cytochrome P450 isoform-specific enzyme activities, ethoxyresorufin deethylase (IA), aniline hydroxylase (IIE) and erythromicin demethylase (IIIA), were assayed in hepatocyte microsomes by using previously published techniques. 9-n Microsomes were diluted to 0.2 m g / m l in 0.1 m o l / L potassium phosphate, pH 7.4, and incubated at 37 ~C for 5 minutes in the presence of the appropriate substrate. The reaction was initiated by addition of 1 m m o l / L NADPH and allowed to proceed at 37 ~ C. Final substrate concentrations were 5 pmol/L ethyoxyresonLfin, 1 m m o l / L erythromicin, and 1 m m o l / L aniline. Turnover (picomoles of product formed per milligram per unit time) was determined from the initial part of the reaction where substrate oxidation was linear as a function of time.

Measurement of iNOS activity. Hepatocytes were washed twice in phosphate-buffered saline solution (pH 7.4), scraped and resuspended in buffer containing 20 m m o l / L Tris, pH 7.4, 0.5 m m o l / L ethyleneglycol-bis-([3-aminoethylether) = N,N,N',N' = tetraacetic acid, 0.5 mmol/L, ethylenediamine tetraacetic acid 1 m m o l / L dithiothreitol, 1 p m o l / L leupeptin, 1 p m o l / L BH4 and 0.2 m m o l / L phenylmethanesulfonyl fluoride. After homogenization the crude cellular homogenate was used immediately for the NOS enzyme assay. All procedures were performed at 4 ~ C. NOS activity was determined by the conversion of tritiated L-arginine to tritiated L-citrulline, as previously described.12,13 Immunoblot analysis of iNOS. Hepatocytes were washed • in phosphate-buffered saline solution and incubated with boiling • nonreducing electrophoresis

Kuo, Abe, Dafoe

311

sample buffer for 2 minutes. Separation was performed on a 12% sodium dodecylsulfate-polyacrylamide gel electrophoresis and then electrotransferred to a polyester-supported nitrocellulose membrane for 90 minutes at 150 milliamps. The membrane was blocked overnight at 4 ~C in triethanolamine-buffered saline solution (10 m m o l / L Tris-HCI at pH 7.5, 150 m m o l / L NaCI) containing 3% bovine serum albumin. Blocked membranes were incubated with immunoblotting murine inducible NOS (iNOS) monoclonal antibody (Transduction Laboratories, Lexington, Ky.), washed three times in triethanolamine-buffered saline solution 0.1% Tween, and incubated with biotinylated sheep anti-mufine immunoglobulin G for 1 hour. After they were washed again three times, membranes were then incubated with streptavidin-horseradish peroxidase conjugate. After washing, bound antibodies were detected by the ECL detection system (Amersham, Arlington Heights, Ill.). RNA extraction and reverse transcriptase-polymerase chain reaction. Total cellular RNA was extracted by the phenol-chloroform-isoamyl alcohol extraction method (Promega RNAgents Total RNA Isolation System; Promega Corp., Madison, Wis.) and treated with 10 units/pl ribonuclease-free deoxyribonuclease. 14 Cellular RNA (0.5 pg) was reverse transcribed by incubating samples for 45 minutes at 42 ~ C in 10 pl reverse transcriptase btfffer containing 5 m m o l / L MgC12, 10 m m o l / L Tris, 0.25 m m o l / L deoxyribonucleoside triphosphate, 0.5 units/ml ribonuclease inhibitor, 2.5 n m o l / L random hexamer, and 10 units/pl Moloney murine leukemia virus reverse transcriptase (Gibco BRL, Gaithersburg, Md.). The reaction was terminated by heating to 95 ~ C for 5 minutes. Polymerase chain reaction (PCR) was performed as previously described. The sequence of primer pairs for iNOS amplification have been described 15 and are TAGAGGAACATCTGGCCAGG and TGGCCGACCTGATGTTGCCA. The sequence primer pairs for rat [3-acfin messenger RNA, used to determine constitutive expression, have also been described 16and are CATCGTGGGCCGCTCTAGC~_AC and CCGC~CAC~CAAGTCCAGACGC. Amplification was initiated by 1 minute of denaturation at 94 ~ C for one cycle followed by 20 to 30 cycles at 94 ~ C for 30 seconds, 60 ~ C for 30 seconds, and 72 ~ C for 2 minutes. After the last cycle the samples were incubated for 10 minutes at 72 ~ C and held at 4 ~ C. Verification of the amplified PCR products was performed by automated DNA sequencing. Determination of aspartate aminotransferase and protein. Aspartate aminotransferase (AST) and lactate dehydrogenase (LDH) were measured by a colorimetric technique (Schiapparelli Biosystems Inc., Fairfield, N.J.). Total protein was determined by the method of Lowry et al. 17

312

Kuo, Abe, Dafoe

Surgery August 1995

Table I. Nitrite a n d AST profile in CMG-treated rat

Table II. Microsomal cytochrome P450 isoform

hepatocytes

activity ( p i c o m o l / m g / m i n )

Control Control + 10 mmol/L CMG Stimulated Stimulated + 1 mmol/L CMG Stimulated + 5 mmol/L CMG Stimulated + 10 mmol/L CMG Stimulated + 100 pmol/L NMMA Stimulated + 10 mmol/L CMG + 100 larnol/L NMMA

Nitrite (nmol/mg)

AST (units/mg)

18 z 1.8# 21 "- 2.3#

0.45 + 0.14 0.52 --- 0.11

82 -* 3.0* 42 +_ 5.6*

0.53 + 0.13 0.58 + 0.22

33 _+ 6.5*

0.46 + 0.19

22 _+ 2.0*

0.52 + 0.13

10 + 1.9+

0.49 + 0.19

15 -~ 2.5

0.49 +_0.17

Stimulated cells were i n c u b a t e d in presence of IL-1 (50 n g / m l ) a n d TNF (500 u n i t s / m l ) . Data are presented as m e a n + SEM of four experiments, CMG. Cimetidine. *p = 0.0001 by single factor ANOVA. #p < 0.01 versus Stimulated, Stimulated + 1 m m o l / L CMG, a n d Stimulated + 5 m m o l / L CMG. t P < 0.01 versus Stimulated.

Control Control+10 mmol/L CMG Stimulated Stimulated + 1 mmol/L CMG Stimulated + 10 mmol/L CMG Stimulated + 10 mmol/L CMG + 100 lamol/L NMMA

IA

liE

IliA

2.9-+ 0.6 2.6+0.7

390 + 27 411+59

93-+ 12 11+_7"

1.8 + 1.2 1.9 _+ 1.1

365 -+ 70 379 +- 47

100 -+ 13 67 -+ 12"#

2.1 +- 0.9

367 -+ 83

18 -+ 8*#t

1.3 + 1.1

384 + 67

21 +- 9*#t

Stimulated cells were incubated in presence of IL-1 (50 ng/ml) and TNF (500 units/ml). Data are presented as mean _+SEM of three experiments. CMG, Cimefidine. *p < 0.01 versus Control. #p < 0.01 versus Stimulated. tP < 0.01 versus Stimulated + 10 mmol/L CMG.

ues were normalized to total cell protein to correct for variability in cell numbers. Cell protein varied by less than 20%. These data indicate that cimefidine decreases cytokine-mediated NO synthesis without associated hepatotoxicity.

Cimetidine and P450 isoform-specific monooxygenase activity. Isoform-specific activities of cytochrome Statistical methods. Data are presented as m e a n + SEM of four experiments and were analyzed with oneway analysis of variance (ANOVA) or Student's t test as appropriate. P values less than 0.05 were considered significant.

RESULTS Effect of cimetidine on NO production. NO production was measured in primary cultures of rat hepatocytes stimulated with IL-1 a n d TNF in the presence a n d absence of the P450IIIA inhibitor cimetidine (Table I). Unstimulated cells served as controls. I n the absence or presence of cimetidine, control cells p r o d u c e d m i n i m a l a m o u n t s of nitrite (18 +- 1.8 n m o l / m g versus 21 -+ 2.3 n m o l / m g ; p = n o t significant). In contrast, cytokine stimulation resulted in a 4.5-fold increase in NO synthesis. I n the presence of cimetidine, NO p r o d u c t i o n in stimulated cells decreased in a concentration-depend e n t fashion (ANOVA p = 0.0001). Addition of a competitive substrate inhibitor, ]XLmonomethyl-L-arginine (NMMA), to stimulated cells dramatically decreased NO synthesis. In the presence of both NMMA a n d cimetidine NO p r o d u c t i o n in stimulated cells did n o t significantly differ from that of stimulated cells with cimetidine. T r e a t m e n t of hepatocytes with IL-1/TNF, NMMA, a n d / o r CMG was n o t intrinsically toxic as shown by AST or LDH release that was n o t statistically different a m o n g the various groups. AST a n d LDH val-

P450 IA, IIE, a n d IIIA were d e t e r m i n e d in the setting of cytokine stimulation in the absence a n d presence of the P450IIIA inhibitor cimetidine (Table II). Activity of the three P450 isoforms did n o t differ significantly between control a n d stimulated cells. Ethoxyresortffin deethylase a n d aniline hydroxylase activities, representative of P450 IA a n d IIE, respectively, were n o t altered by incubation with IL-1 a n d TNF in the presence or absence of cimetidine (1 m m o l / L a n d 10 m m o l / L ) . In contrast, P450IIIA activity, as measured by erythromicin demethylase, decreased in both control a n d stimulated cells when exposed to cimetidine. Cytokine-stimulated hepatocytes exhibited a cimeddine concentrationd e p e n d e n t response in P450IIIA activity. This P450 activity profile was n o t altered by coadministration of the competitive substrate inhibitor, NMMA, to cytokinestimulated or control hepatocytes. The relative mass of microsomal cytochrome P450 IA, liE, a n d IIIAwere n o t altered by cytokine-stimulation a n d / o r cimetidine exposure as measured by i m m u n o b l o t analysis (data n o t shown). These results indicate that cimeddine is a specific inhibitor of P450IIIA enzyme activity whose effect is n o t altered by the presence of IL-1, TNF, or NO. RT-PCR analysis of steady state iNOS m R N A expression. Complementary DNA (cDNA) from isolated rat hepatocytes was amplified by using a single set of primers specific for iNOS a n d [~-actin. DNA sequence

Surgery Volume 118, Number 2

Kuo, Abe, Dafoe 313

T a b l e H I . Tritiated N-hydroxy-L-arginine a n d

tritiated-L-citrulline formation in cell h o m o g e n a t e s

N-hyUroxyControl Stimulated Stimulated + 1 mM CMG Stimulated + 5 mM C M G Stimulated + 10 mM CMG

L-arginine

L-citruUine

ND 38 -+ 13 33 + 15 40 _+ 19 37 + 11

ND 69 + 9* 47 +--7* 29 + 8* 12 -+ 9*

Activity expressed as pmol/mg/min. Data are presented as mean + SEM of four experiments. ND, Not determined; CMG,cimetidine. *p = 0.0001 within group by single factor ANOVA.

analysis of the amplified PCR products revealed n e a r c o m p l e t e h o m o l o g y to published sequences for rat iNOS (98%) a n d 13-actin (99%). Amplification o f c D N A isolated from hepatocytes incubated with IL-1 a n d TNF exhibited an iNOS PCR p r o d u c t o f the anticipated size (approximately 250 base pairs) after 15 cycles. F u r t h e r amplification showed a linear increase in PCR p r o d u c t between 15 a n d 25 cycles without further increase in PCR p r o d u c t beyond 25 cycles o f amplification. Therefore 23 amplification cycles were used in subsequent studies; negative controls included reverse transcriptase-PCR (RT-PCR) reactions p e r f o r m e d in the absence of reverse transcriptase, deoxyribonuclease, o r template. RT-PCR o f control a n d stimulated r a t hepatocyte RNA was p e r f o r m e d in the presence a n d absence o f l 0 m m o l / L cimetidine (Fig. 1). In controls there was no detectable iNOS cDNA amplification product. In contrast, cytokine stimulation was associated with the a p p e a r a n c e o f iNOS cDNA, a n d the relative density o f this p r o d u c t was n o t altered by the addition o f 10 m m o l / L cimetidine. This same p a t t e r n was p r e s e n t in relative iNOS protein mass, as m e a s u r e d by i m m u n o blot analysis (Fig. 2). In the presence o r absence o f cimetidine, no detectable iNOS protein was p r e s e n t in control cells. In cytokine-stimulated cells, iNOS was present a n d the relative quantity was n o t altered by the presence o f 10 m m o l / L cimetidine. RT-PCR o f cytokine~stimulated hepatocyte RNAwas then p e r f o r m e d in the presence o f increasing concentrations (0, 1, 5, a n d l 0 m m o l / L ) o f c i m e t i d i n e (Fig. 3). T h e relative density o f steady state iNOS m R N A did n o t significantly change at any concentration o f cimetidine. These data indicate that decreased N O p r o d u c t i o n n o t e d in stimulated cells in the presence o f cimetidine was n o t the result o f alteration in steady state iNOS m R N A expression or p r o t e i n content. P r o d u c t i o n o f N-OH-L-arginine a n d L-citrulline. N O synthase activity was d e t e r m i n e d in control a n d stimulated hepatocellular h o m o g e n a t e s in the presence a n d absence o f cimetidine (Table III). Activity was deter-

< Z n" E

0.6. 0.5-

O 0.4.

O.3e-

9

o.2-

.-~

o.1-

~

o B

C

S

C+CMG

S+CMG

Fig, 1. A, RT-PCR of hepatocyte iNOS and [~-actin mRNA expression in control, stimulated (50 ng/ml IL-1, 100 units/ml TNF), Control + CMG (10 mmol/L), and Stimulated + CMG (10 mmol/L) cells. Control PCR reactions were performed as follows: no deoxyribonuclease, no template, and no reverse transcriptase. Gel is representative of four experiments. B, Relative density of iNOS mRNA normalized to [3-actin. Data presented as mean _+ SEM of four experiments. C, Control; S, Stimulated; CMG, cimetidine (10 mmol/L).

Kuo, Abe, Dafoe

314

Surgery August 1995

5O

m

e-

40-

o 0

.->

30 i

u~ o a.

20

O 10o a..

0

B

C

S

C+CMG

S+CMG

Fig. 2. A, Immunoblot of iNOS protein expression in Control, Stimulated (50 ng/ml IL-1, 100 units/ml TNF), Control + CMG (10 mmol/L), Stimulated + CMG (10 mmol/L) rat hepatoeytes and murine iNOS positive control. Blot is representative of three experiments. B, Relative iNOS protein content expressed as percentage of positive muriue control. Data presented as mean -+ SEM of three experiments. C, Control; S, Stimulated; CMG,cimetidine.

'r z

1.25-

Iz:

E

1-

0 z .m 0.75 -

m i n e d by the NOS catalyzed reaction converting tritiated L-arginine to tritiated L-citrulline, a c o m m o n e n d product with NO. I n controls n o detectable N-OH-Larginine or L-citrulline was present. In stimulated hepatocytes the addition of cimetidine resulted in a concentration-dependent decrease in formation of the final product, L-citrulline. This paralleled the decrease in NO production seen with cimetidine treatment of cytokinestimulated hepatocytes. O f note, the rate of formation of the intermediate, N-OH-L-arginine,was n o t altered by the presence of cimetidine in stimulated cells. These data indicate that inhibition of P450IIIA activity is associated with decreased conversion of N-OH-L-arginine to L-citrulline. In addition, the lack of effect o n N-OH-Larginine synthesis suggests that the site of action for cimetidine is n o t NOS.

C

>

:~,

0.5-

0.25-

he 0 mM

1 mM

[ C M G ] in S t i m u l a t e d

5 mM

10 mM

Hepatocytes

Fig. 3. A, RT-PCR of hepatocyte iNOS and [3-actin mRNA expression in Stimulated (50 ng/ml IL-1, 100 units/ml TNF), Stimulated + 1 mmol/L cimetidine (CMG), Stimulated + 5 mmol/L CMG, and Stimulated + 10 mmol/L CMG cells. Control PCR reactions were performed as follows: no template and no reverse transcriptase. Gel is representative of four experiments. B, Relative density of iNOS mRNA normalized to [3-actin.Data presented as mean _+SEM of four experiments.

Surgery Volume 118, Number 2

Kuo, Abe, Dafoe 315

H

a

I@

N p~ ILF .'NHNADP_-Hoz Ne~l H~

CO0e

,--Aa'gini-,e

HsN

0.~ NADPH

Fl4Bioplg~ oz

--NO"

"]"

H

COOe

N. H~UroxT-L-m.~u~

~lde

/.-Cll~iline

Fig. 4. Reaction scheme for NO synthesis by iNOS.

Dixon plot analysis was performed to determine the kinetic constants associated with cimetidine-mediated inhibition of NO synthesis. Given that L-arginine was added in excess, the substrate concentration was assumed to approach infinity and the plot of 1/velocity versus [cimetidine] yielded an inhibition constant Ki of 1.76 m m o l / L .

DISCUSSION NO is a multifunctional free radical mediator that is thought to participate in the physiology and pathophysiology of virtually every organ system. Three mammalian gene products are known to act in the capacity of NOS; however, the inducible isoform, as expressed in the hepatocyte and macrophage, is most widely expressed in a multitude of cell types after transcriptional induction. 1 Current evidence suggests that the various NOS isozymes have similar catalytic properties. All require NADPH, BH4, and molecular oxygen as cofactors and contain noncovalently b o u n d flavin adenine dinucleotide and flavin mononucleotide. BH4 is used for the hydroxylation o f L-arginine to form the intermediate, N-OH-L-arginine. Molecular oxygen is then reduced in an flavin adenine dinucleotide-flavin mononucleotide-mediated series of electron transfer steps with NADPH acting as the electron donor. The resulting reduced oxygen species reacts with N-OH+-arginine to produce NO and L-citrulline (Fig. 4).2, 12 NOS has been found to have substantial sequence homology with cytochrome P450 reductase and may participate in cellular electron transfer processes via its reductase domain. 3 Additional work by White and Marletta 18 indicates that NOS is a P450-1ike hemoprotein and, as a result, may be the first example of a catalytically self-sufficient mammalian P450 enzyme, containing both a reductase and heine domain, similar to that of the fatty acid monooxygenase isolated from B. megaterium. These considerations notwithstanding, the relationship between the native cytochrome P450 system and NO synthase is largely unknown.

The term cytochrome P450 system is used to describe a diverse family of mixed function oxidases that are present in all living organisms and are critically dependent on the i r o n / h e m e cofactor. The generic representation of P450-mediated oxidation ofsubstrate S may be depicted as the following: S + NADPH + 02--> S-O + H~O + NADP+ The enzymes involved include a cytochrome P450, an iron-sulfur cluster containing protein, and a flavoprotein reductase. The cytochrome P450 reacts directly with oxygen and receives electrons from the iron-sulfur protein, which is in turn reduced by NADPH by the flavoprotein reductase. The result is an "oxygenating" form of P450. 7 The similarity of this reaction with that of NO synthesis, especially the oxidation of N-OH-Larginine to NO and citrulline, suggests a role for the P450 system. (The nomenclature of the P450 enzymes deserves explanation. P450 proteins with greater than 40% sequence identity are included within a single family and are designated by a number; those with greater than 55% identity are grouped within a subfamily and designated with a capital letter [e.g., P450IIIA] ). 19 Stadler et al.20 have shown that cytokine-mediated production of NO inhibited P450IA isoform activity and steady state mRNA expression. In rat hepatic microsoreal subfractions Wink et al.21 have also demonstrated an NO-dependent decrease in P45IIB activity. Conversely, in this model of rat hepatocytes in primary culture we have previously demonstrated that P450IIIA is involved in cytokine-mediated NO synthesis. Cytochrome P450IIIA, IIE, and IA activities were specifically inhibited by cimetidine (IIIA), clotrimazole (IIIA), benzoflavone (IA), and disulfiram (IIE). Cytokine-induced NO synthesis was significantly decreased in the presence of both cimetidine and clotrimazole. Kinetic analysis revealed a noncompetitive m o d e of inhibition (Ki = 21 m m o l / L - cimetidine; Ki = 13 p m o l / L clotrimazole) .6

316

Kuo, Abe, Dafoe

In this study we sought to further delineate the role o f P450IIIA in cytokine-mediated N O p r o d u c t i o n by rat hepatocytes. Using the specific P450IIIA inhibitor cimetidine, we have d e m o n s t r a t e d a concentration-dep e n d e n t decline in N O p r o d u c t i o n without associated hepatocyte injury. 7 In the presence of cimetidine (10 m m o l / L ) , specific inhibition o f P450IIIA activity was shown without c o n c o m i t a n t alteration in P450IA o r IIE activity, corroborating o u r previous observations a n d those of o t h e r investigators. 4' 7 Steady state levels of iNOS messenger RNA were n o t altered by cimetidine as shown by RT-PCR analysis. In addition, relative iNOS p r o t e i n mass was n o t different. A d d i t i o n of a competitive substrate inhibitor o f N O synthesis did n o t alter these observations. However, in the presence o f cimetidine, N O S enzyme activity was f o u n d to decrease in a c o n c e n t r a t i o n - d e p e n d e n t m a n n e r with a Ki for cimetidine o f 1.76 m m o l / L , Interestingly, whereas accumulation o f the intermediate, N-OH-L-arginine, was n o t affected, subsequent oxidation to form NO a n d L-Citrulline was inhibited. These results suggest that P450IIIA activity may be involved in cytokine-mediated hepatocyte N O synthesis in a posttranslational manner. In contrast to the findings o f Stadler et al. 2~ a n d W i n k et alfi 1 we were n o t able to demonstrate an N O - d e p e n d e n t inhibition o f P450IA, IIE, or IIIA monooxygenase activities in o u r system. Given the characterization o f N O S as an enzyme c o m b i n i n g b o t h reductase activity a n d a P450-1ike h e m e moiety, direct inhibition o f NOS activity by cimetidine is a potential explanation. However, specific inhibition of o t h e r P450 isoforms by using competitive substrate o r mechanism-based inhibitors is n o t associated with alteration in NOS activity. Nonspecific inhibition of P450 has n o t b e e n shown to decrease NOS activity in a n u m b e r o f models. 22, 23 These observations suggest that functional inhibition of P450IIIA rather than NOS underlies o u r observations. In addition, by use of a m o d e l o f rat microsomes devoid o f N O S activity, Boucher et al. 24 have shown that the P450 system can mediate conversion of N-OH-L-arginine to L-citrulline. However, the applicability o f this observation to the intact hepatocyte is unknown. In a system of vascular smooth muscle cell Schott et al. 25 have shown that exogenous N-OH-L-arginine is oxidized to form N O by a cytochrome P450-dep e n d e n t process. Finally, when purified NOS was incub a t e d with cimetidine in an in vitro system replete with the essential cofactors, NOS enzyme activity was n o t alt e r e d (unpublished observations). T h e mechanistic implications o f o u r observations are as yet unknown, b u t some hypotheses can be formulated. First, in the intact hepatocyte the P450IIIA system may r e p r e s e n t a p r e f e r r e d alternative pathway for conversion of N-OH-L-arginine to L-citrulline either directly o r by provision o f essential cofactors such as r e d u c e d

Surgery August 1995 molecular oxygen o r o t h e r P450 by-products. Alternatively, the r e d o x milieu o f the cell may d e t e r m i n e the p r e f e r r e d pathway for N-OH-L-arginine metabolism: NOS versus P450. T h e role o f the cytokines IL-1 o r TNF in this process has yet to be addressed. Experiments with n o n c y t o k i n e - d e p e n d e n t mechanisms for NOS induction, such as forskolin, will be valuable in this regard a n d are currently ongoing. Finally, Boucher et al. 24 have suggested that the P450 system represents a pathway by which cells o r cell c o m p a r t m e n t s devoid o f N O S may oxidize N-OH-L-arginine. T h e work by Schott et al. 25 indicates that N-OH-L-arginine is recognized by a cationic L-amino acid c a r d e r a n d may be transported into cells or cell c o m p a r t m e n t s devoid of NOS activity. Certainly, the physiologic significance o f o u r observations in the whole organism o r whole organ remains to be established. However, given e m e r g i n g evidence c o n c e r n i n g the protective role o f N O synthesis in oxidative injury, strategies to induce o r a u g m e n t cytochrome P450IIIA activity may be o f potential utility in the future. In this study, by use o f a system of rat hepatocytes in primary culture, specific inhibition o f cytochrome P450IIIA acdvity was associated with decreased cytokinem e d i a t e d N O production. Steady state levels o f NOS messenger RNA a n d p r o t e i n expression were n o t altered in this setting. NOS enzyme activity was decreased with P450IIIA inhibition, specifically in the secondary step o f N-OH-L-arginine oxidation to form N O a n d L-citnllline. These results suggest a posttranslational role for the P450IIIA isoform in cytokine-mediated synthesis o f N O by the rat hepatocyte. REFERENCES

1. Nathan C. Nitric oxide as a secretoryproduct of mammalimacells. FASEBJ 1992;6:3051-64. 2. Klatt P, Heinzel B, MathiasJ, Kastner M, Bohme E, Mayer B. Ca2+/calmodulin-dependentcytochrome c reductase activityof brain nitric oxide synthase.J Biol Chem 1992;267:1137443. 3. Bredt DS, Hwang PH, Glatt CE, Lowenstein C, Reed R, Snyder SH. Cloned and expressed nitric oxide synthase structurally resembles cytochrome P450 reductase. Nature 1991; 351:71443. 4. Kuo PC, Abe KY. Hepatocyte production of nitric oxide is dependent on cytochrome P450 3A activity[abstract]. Hepatology 1994;20:186A. 5. Schuetz EG, Wrighton SA, Safe SH, Guzelian PS. Regulation of cytochrome P450 by phenobarbital and phenobarbital-like inducers in adult rat hepatocytesin primarymonolayer culture and in vitro.Biochemistry 1986;25:1124-33. 6. Kuo PC, SlivkaA. Nitric oxide decreases oxidant mediated hepatocyte injury.J Surg Res 1994;56:594-600. 7. Guengerich FP. Characterization of human cytochrome P450 enzymes. FASEBJ 1992;6:745-8. 8. Snell FD, SnellCT. Colorimetricmethods of analysis.3rd ed. New York: D Van Norstrand Co, 1984. 9. DiazD, Fabre I, Daujat M, et al. Omeprazole is an aryl hydrocarbon-like inducer of human hepatic cytochrome P450. Gastroenterology 1990;99:737~t7. 10. CombalbertJ, Fabre I, Fabre G, et al. Metabolismof cyclosporin A. Drug Metab Dispos 1989;17:197-207.

Surgery Volume 118, Number 2 11. Bonfils C, Combalbert J, Pineau T, et al. Ontogenesis of rabbit liver cytochrome P450. EurJ Biochem 1990;188:187-94. 12. Stuehr DJ, KwonNS, Nathan CF, et al. N-hydroxy-L-arginineis an intermediate in the biosynthesis of nitric oxide from L-arginine. J Biol Chem 1991;266:6259-63. 13. Evans T, Carpenter A, CohenJ. Purification of a distinctive form of endotoxin-induced nitric oxide synthase from rat liver. Proc Nail Acad Sci USA 1992;89:5361-5. 14. SambrookJ, Fritsch EF, Maniatis T. Molecular cloning, a laboratory manual. Cold Spring, New York: Cold Spring Harbor Press, 1989. 15. Beasley D, McGuiggin M. Interleukin-1 activates soluble guanylate cyclase in human vascular smooth muscle cells through a novel nitric oxide-independent pathway. J Exp Med 1994; 179:71-80. 16. Bishop GA, Rokahr tL NapoliJ, McCanghan GW. Intragraft cytokine mRNA levels in human allograft rejection analysed by the reverse transcription and semiquantitative polymerase chain reaction amplification. Transplant Immunol 1993;1: 253-61. 17. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the Folin phenol reagent. J Biol Chem 1951; 193:265-75. 18. White KA, Marletta MA. Nitric oxide synthase is a cytochrome P450 type hemoprotein. Biochemistry 1992;31:6627-31. 19. Coon MJ, Ding XX, Pernecky SJ, Vaz ADN. Cytochrome P450: progress and predictions. FASEBJ 1992;6:669-73. 20. Stadler J, Trockfeld J, Schmalix WA, et al. Inhibition of cytochromes P4501A by nitric oxide. Proc Nail Acad Sci USA 1994; 91:3559-63. 21. Wink DA, Osawa Y, Darbyshire JF, Jones CR, Eshenaur SC, Nims RW. Inhibition of cytochromes P450 by nitric oxide and a nitric oxide-releasing agent. Arch Biochem Biophys 1993; 300:115-23. 22. Oyekan AO, McGiffJC, Rosencrantz-WeissP, QuilleyJ. Relaxant responses of rabbit aorta: influence of cytocbrome P450 inhibitors. J Pharmacol Exp Ther 1994;268:262-9. 23. Fulton D, McGiffJC, QuilleyJ. Contribution of NO and cytochrome P450 to the vasodilator effect of bradykinin in the rat kidney. BrJ Pharmacol 1992;107:722-5. 24. BoucherJL, Genet A, Vadon S, Delaforge M, Mansuy D. Formation of nitrogen oxides and citrulline upon oxidation of N-hydroxy-L-arginine by"hemeproteins. Biochem Biophys Res Commun 1992;184:1158-64. 25. Schott CA, Bogen CM, VetrovskyP, Berton CC, StocletJC. Exogenous N-hydroxy-L-argininecauses nitrite production in vascular

Kuo, Abe, Dafoe 317

smooth muscle cells in the absence of nitric oxide synthase activity. FEBS Lett 1994;341:203-7.

DISCUSSION Dr. Timothy R. Billiar (Pittsburgh, Pa.). I have two questions that I think n e e d to be addressed before you can move on in your studies. Would the cilnefidine inhibit NO formation by the purified enzyme? Have you purified iNOS away from the crude lysate to homogeneity and then attempted to block the formation of NO to assure that the cimefidine is not acting directly on the enzyme? Because hepatocytes are unique in t h a t they contain P450 and many other cell types that produce NO do not contain P450 enzymes, have you used cimefidine on a cell such as a macrophage that expresses iNOS but does not contain P450 to determine whether cimefidine can inhibit in that cell type? Dr. Kuo, The answer to the first question is that we actually have done partial purification of the iNOS enzyme by adenosine diphosphate sepharose chromatography and have found that incubation with cimetidine did not alter NO production. In answer to your second question, we haven't tried a P450 IIIA inhibitor or a P450 inhibitor at all in any nonhepatocyte model for NO production. Dr. Nathan Coates (Houston, Texas). The AST release really comes when the cells actually lyse and allow some of the enzyme to escape. Have you looked at some of the active function of the hepatocytes, for example, using lidocaine for monoethylglycinexylidide production or any of the other enzymatic activity of the hepatocytes? Dr. Kuo. We have not tried to look at lidocaine metabolism or any other metabolic function in this model. Dr. R. Armour Forse (Boston, Mass.). I have a methodology question. These seemed to be variable 13-actin gene expression, and I wonder whether [3-actin is appropriate to correct for differences in gene expression or whether it represents some technical problems that you were having with your PCR. Dr. Kuo. All of our reverse transcriptase-PCR was performed in the setting of a log linear dose-response curve. Twenty-three cycles were chosen for all of our studies, so we applied 23 cycles to the [3-actin expression and iNOS expression.