Expression in Escherichia coli of the flavin ... - ACS Publications

Monooxygenase Form 3: cDNA Expression of the Enzymes Containing Amino Acid Substitutions Observed in Individuals with Trimethylaminuria. John R...
0 downloads 0 Views 717KB Size
Chem. Res. Toxicol. 1993,6, 425-429

425

Expression in Escherichia coli of the Flavin-Containing Monooxygenase D (Form 11) from Adult Human Liver: Determination of a Distinct Tertiary Amine Substrate Specificity Noureddine Lomri, Zicheng Yang, and John R. Cashman, Department of Pharmaceutical Chemistry and Liver Center, School of Pharmacy, Uniuersity of California, S a n Francisco, California 94143-0446 Received March 29, 1993

The cDNA for a major component of the family of flavin-containing monooxygenases (FMOs) present in adult human liver (i.e., HLFMO-D) has been cloned and expressed in a prokaryotic system. Escherichia coli strain NM522 was transformed with pTrcHLFMO-D, and the HLFMO-D cDNA was expressed under the control of the Trc promoter. A variety of tertiary amine substrates Le., chlorpromazine and 10-[(N,N-dimethylamino)alkyl]-2-(trifluoromethy1)phenothiazinesl were efficiently oxygenated by HLFMO-D cDNA expressed in E. coli or by adult human liver microsomes. Approximate dimensions of the substrate binding channel for both adult human liver microsomal FMO and cDNA-expressed HLFMO-D were apparent from an examination of the N-oxygenation of a series of lo-[ (NlN-dimethylamino)a1ky1]-2(trifluoromethy1)phenothiazines.The substrate regioselectivity studies suggest that adult human liver FMO form D possesses a distinct substrate specificity compared with form A FMO from animal hepatic sources. It is likely that the substrate specificity observed for cDNA-expressed adult human liver FMO-D may have consequences for the metabolism and distribution of tertiary amines and phosphorus- and sulfur-containing drugs in humans and may provide insight into the physiologic substrate(s) for adult human liver FMO.

Introduction The flavin-containing monooxygenases (FM0s)l (EC 1.14.13.8) comprise a family of mammalian microsomal monooxygenases which oxygenate nucleophilic nitrogen-, sulfur-, and phosphorus-containing drugs and chemicals to N-, S-,and P-oxides, respectively (1,2).Adult human liver FMO cDNAs have been cloned, but the corresponding enzymes have not been isolated and purified to homogeneity, and attempts to confirm the amino acid sequences of the microsomal proteins by sequencing the purified proteins have been confounded by lack of enzyme stability, and low or no immunoreactivity to animal FMO antibodies. In human liver, two distinct cDNAs encoding quite different proteins have been cloned: FMO-A2 (31, considered to be mainly present in the fetal human liver, and FMO-D (4),considered to be prominent in the adult human liver. Fetal human liver FMO-A shares approximately 88% identity with pig liver FMO and 86% identity with rabbit liver FMO-A deduced from the cDNA data (3). Adult human liver FMO-D is only 52-57% identical to FMO-A from animals (4). Another FMO has been cloned '

* To whom correspondenceand reprint requests should be addressed at IGEN Research Institute, 130 5th Ave., N, Seattle, WA 98109. Telephone (206) 441-6684; FAX (206) 443-0685. 1 Abbreviations: FMO, flavin-contaming monooxygenase;HLFMOD, adult human liver flavin-containing monooxygenase;PMSF, phenylmethanesulfonyl fluoride; IPTG, isopropyl 8-Dthiogalactopyranoside; PCR, polymerase chain reaction; DETAPAC, diethylenetriaminepentaaceticacid;LSIMS,liquidsecondaryionmasa spectrometry;CI,chemical ionization; MOPS, 3-(N-morpholino)propanesulfonicacid;FAD, flavin adeninine dinucleotide. a HLFMO-Dhas ale0 been designated as FMO 11, and HLFMO A has been designated as FMO I on the basis of our previous publications.The new designation is based on the rabbit FMO nomenclature. However, a confoundingreport (ref 5) has described another human liver enzyme quite dissimilar to the enzyme studied here as FM02. 0893-228~/93/ 2706-0426$04.00/ 0

from adult human liver, but it is apparently a minor component and not very similar to the FMO-D reported here (5). Substrate specificity differences are apparent for hepatic form A and D FMOs from in vitro animal liver but with few exceptions, little is known enzyme studies (6), about the adult human liver enzymes (7,8).Another form of FMO has been isolated from the rabbit lung (Le., FMOB) (9,IO), has been shown to possess only 55% amino acid identity with hepatic FMOs, and has a distinct substrate specificity (11, 12). A number of studies have shown that adult human liver microsomes are capable of tertiary amine N-oxygenation (13-17) and thiobenzamide S-oxygenation (15). Adult human liver FMO-dependent N- and S-oxygenation activity is quite thermally labile, and activity is maximal at pH 8.4 (13-161, although considerable intersample variation has been observed. Most physical properties of animal liver FMOs are shared by adult human liver FMO-D although differences in substrate specificity (with two notable exceptions) have not been extensively examined (7,8). In another study, it was reported that human liver microsomes did not N-oxygenate imipramine even though imipramine was an excellent substrate for pig liver FMO-A (16). Immunoquantitation of human liver FMO has relied on antibodies directed against animal FMOs. Thus, polyclonal antibodies prepared against pig liver FMO-A recognized a 60 000-Da human liver protein although the immunoblot was characterized as very faint. Antisera raised against rat liver FMO recognized an adult human kidney protein but did not recognize anything in the adult human liver (16), suggesting that multiple forms of FMO were present in the adult human liver and kidney. The possible presence of multiple forms of adult human liver FMO prompted us to clone and express the adult 0 1993 American Chemical Society

426 Chem. Res. Toxicol., Vol. 6, No. 4, 1993 h u m a n liver FMO-D cDNA and t o examine t h e substrate specificity of this enzyme. Herein, we report on t h e kinetic properties of microsomal and cDNA-expressed adult h u m a n liver FMO-D. Compared with FMO-A f r o m pig liver, we find t h a t t h e cDNA-expressed and microsomal FMO-D from adult h u m a n liver possesses a distinct substrate specificity and presumably quite different dimensions for t h e substrate binding channel.

Communications

in a modified MOPS medium (22) supplied with 20 mg/L FAD. After inoculation with a fresh overnight culture of NM522 containing pTrcHLFMO-D, the culture was grown at 37 “C until A m was approximately 0.5. Finally, IPTG (0.8 mM) was added to the culture to induce expression of protein, and the culture was left to grow overnight at 37 “C. Cell Lysis a n d Solubilization. The bacterial pellet was resuspended in 5 volumes of 50 mM phosphate buffer (pH 8.4) containing 1% Triton X-100,0.1% L-a-phosphatidylcholinefrom egg yolk, 500 KM PMSF, and 1 mM EDTA. The mixture was Experimental Procedures sonicated and then centrifuged to remove cellular debris. The solubilized protein from the bacterial lysate was used directly. Materials. Chlorpromazine was purchased from Aldrich Protein concentration was determined by the BCA protein assay Chemical Co. (Milwaukee, WI). The 10-[(N,N-dimethylamifrom Pierce (Rockford, IL). no)alkyl]-2-(trifluoromethy1)phenothiazineswere a kind gift of Enzyme Preparations a n d 1ncubations.Adult humanliver Professor D. M. Ziegler (University of Texas, Austin, TX) (17). samples were obtained from the Department of Surgery (UCSF) The chemical synthesis of tertiary amine N-oxides and sulfoxides and with a protocol from the Medical College of Wisconsin (Steve of the phenothiazines was accomplished as previously described (18,19). Chlorpromazine and the l ~ [ ( N ~ - d i m e t h y l a m i n o ) ~ l l - Wrighton, now a t Lilly Inc., Indianapolis, IN) and were screened for FMO-D N-oxygenase activity (13). One individual sample 2-(trifluromethy1)phenothiazineN-oxides were biosynthesized (i.e., J microsomes from a normal female without a previous with pig liver microsomes as described before (17). All of the history of drug administration) possessed significant cytochrome substrates and N-oxide derivatives were completely characterized P-450 activity [i.e., 360 pmol of 7-hydroxycoumarin formedl by 1H NMR, mass spectrometry, and UV-vis and were identical (min-mgofprotein)] and FMO-D activity [Le., 133pmolof transto authentic standards. Triton X-100, L-a-phosphatidylcholine, nicotine ”oxide formed/(min.mg of protein)] (13)and was used and the components of the NADPH-generating system were from in this study. Approval from the UCSF and Medical College of Sigma Chemical Co. (St. Louis, MO). Flavin adenine dinucleWisconsin Committees on Human Research, respectively, were otide (FAD) was from Kodak (Rochester, NY). Restriction obtained for these studies. Pig and human liver microsomes endonucleases, T4 DNA ligase, and polymerase 1and polymerase were isolated as described before (23). Detergent-solubilized 1 Klenow fragment were from Boehringer-Mannheim (Indiahuman liver HLFMO-D (Le., cDNA-expressed HLFMO-D) was napolis, IN). The pTrc99A expression vector was purchased from expressed in competent NM522 E. coli cells after transformation PharmaciaLKB Biotechnology Inc. (Milwaukee,WI). Gene Amp with the pTrc HLFMO-D plasmid and was used directly as the DNA amplification reagent kits were purchased from Perkincell lysate. Elmer Cetus (Emeryville, CA). The sequence kit was from U.S. A typical incubation mixture (final volume 0.25 mL) contained Biochemical (San Diego, CA). Isopropyl /3-D-thiogdactopyrahepatic microsomes (0.4 mg of protein) or bacterial lysate (2.3 noside (IPTG) and all other chemicals used were of the highest mg of protein), NADP+ (0.5mM), glucose 6-phosphate (2.0 mM), purity available and were purchased from Fisher Chemical Co. glucose-6-phosphate dehydrogenase (1IU), DETAPAC (0.8 mM), (Richmond, CA). and potassium phosphate buffer (50 mM, pH 8.4). Reaction Extension of t h e 5’- a n d 3’-Ends of HLFMO-D cDNA products were quantified by HPLC as described previously (7, Coding Strand. To obtain the full-length open reading frame 18). The [(N,N-dimethylamino)alkyl1-2- (trifluoromethy1)phenofor human liver flavin-containing monooxygenase (form D) thiazine metabolites were separated by an HPLC method pre(HLFMO-D) containing convenient ends for subcloning into the viously described (24). The HPLC system efficiently separated expression vector pTrc99A, the polymer chain reaction (PCR) the tertiary amine, the alkyl side-chain tertiary amine N-oxide, technique was employed. For PCR, two oligonucleotide primers were designed. Primer A (5’-GGTACCACATGTCCATGGG- and the sulfoxide. In all cases examined, the substrate, N- or S-oxide metabolite, or other metabolites were recovered in 285 % GAAGAAAG-3’) consisted of an AflIII site a t the 5‘-end of the efficiency and the substrate and N- or S-oxide accounted for HLFMO-D cDNA. Primer B (5’-GACGTCGACGGATCCT1 9 5 % of the material present. TAGGTCAACACA-3‘) had a SalI site a t the 5’-end and a 13nucleotide sequence complementary to the 3’-end of the HLFMO-D cDNA coding strand. The PCR was carried out with a Results Perkin-Elmer Cetus (San Jose, CA) DNA thermal cycler, using Expression of Adult Human Liver FMO-D cDNA a GeneAmp DNA amplification reagent kit with the largest HLFMO-D cDNA clone from X g t l l (4) as a template. After 30 in E. coli. The PCR product which was designed t o obtain cycles of PCR reaction, a full-length HLFMO-D cDNA coding the full-length open reading f r a m e cDNA of HLFMO-D strand was obtained with an AflIII restriction site and a SalI site was inserted into a pTrc99A expression vector to give the attached to the 5’- and 3’-ends of the cDNA, respectively. expression plasmid, p T r c H L F M 0 - D . Restriction enzyme Prokaryotic Expression of HLFMO-D. After the PCR and DNA sequence analyses (21) of t h e p T r c H L F M 0 - D product was purified and digested with AflIII and SalI restriction DNA sequence confirmed that the entire HLFMO-D enzymes, the generated fragment was subjected to fractionation cDNA coding strand was successfully extended and by electrophoresis on a 1% agarose gel, purified, and subcloned correctly inserted into t h e pTrc99A vector. The expression into the NcoI-Sal1 sites of the inducible prokaryotic expression of p T r c H L F M 0 - D i n the E. coli host bacteria NM522 vector pTrc99A. The recombinant clones were identified by following incubation in the presence of t h e inducing agent screening with a 32P-labeled HLFMO-D cDNA probe (20) using IPTG produced active HLFMO-D. As a control, we the AflIII-SalI-generated PCR fragment. Two clones were used for sequencing by the dideoxy method (21) with the sequence examined bacterial lysate extracts from pTrc99A-transkit. The sequence of the entire insert was obtained from both formed NM522 host bacteria, but t h e bacteria did n o t strands by using a combination of exonuclease I11 deletions, produce a n y detectable N-oxygenase activity when grown subcloning techniques, and custom-made primers. The correct in the presence or absence of IPTG. insertion was also confirmed by digestion with NcoI and SalI Substrate Oxygenation. T h e regioselective oxygenand DNA sequence analysis. The sequence determined was ation of various tertiary amines by human liver microsomes identical to the HLFMO-D cDNA reported previously (4), and and cDNA-expressed HLFMO-D was examined to selecthe modified 5’- and 3’-sites provided the anticipated sequence. tively monitor FMO enzyme action as well as to examine Competent NM522 Escherichia coli cells were transformed t h e possible involvement of cytochromes P-450 or nonwith the pTrcHLFM0-D plasmid. Expression was performed

Communications

Figure 1. Structure of the lo-[ (N,N-dimethylamino)alkyl1-2substitutad-phenothiazinesused in this study. For chlorpromazine, X = C1 and n = 3; for compounds 2-6, X = CFs and n = 2-6, respectively.

enzymatic oxidation of the same substrate. Tricyclic antidepressants (Figure 1)provided excellent probes for monooxygenase action because these chemicals possessed a tertiary amine center known to be selectively N-oxygenated by FMO (17-19) and a nonnucleophilic sulfur atom known to be oxidized by cytochromes P-450 or by nonenzymatic oxidants (Le., Hz02 or ROOH). In all cases examined, product formation was directly determined by HPLC analysis of organic extracts. cDNA-expressed HLFMO-D was evaluated for N- and S-oxygenase activity with compound 5. Solubilization of the recombinant HLFMO-D protein resulted in 84 % N-oxygenase activity [i.e., 220 pmol/(min.mg of protein)] in the lOOOOOg supernatant fraction with the remainder of the activity in the pellet. No detectable amount of nonenzymatic N-oxygenation was observed during the metabolic incubations examined. In addition, no detectable N-oxide reduction was observed during the course of our studies. Preliminary studies showed that cDNA-expressed HLFMO-D supplemented with NADPH catalyzed the N-oxygenation of a variety of substrates. As a standard for comparison, the oxidation of substrates with adult human liver microsomes was also studied. For compound 6, formation of the tertiary amine N-oxide was a linear function of bacterial lysate protein concentration (0-4.6 mg of protein) and of time (0-10 min) (data not shown). In the presence of adult human liver microsomes, tertiary amine N-oxide formation of 6 was a linear function of microsomal protein concentration (0-0.5 mg of protein) and of time (0-30min) (data not shown). N-Oxygenation of 6 was dependent upon the pH of the reaction mixture, and microsomal protein and cDNA-expressed HLFMO-D gave virtually identical pH-rate profiles. The pH optimum for tertiary amine N-oxygenation was approximately 10, and the pH-rate profile resembled a titration curve with the midpoint near avalue of pH 9.5. The true pH optimum is probably 8.5, and it is likely that deprotonation of the amines (pK, = 9.5) provided the nonprotonated substrate and thus influenced the apparent pH optimum of the enzyme. The effect of various incubation conditions and metabolism inhibitors on the N- and S-oxygenation of 6 was examined with cDNA-expressed HLFMO-D and with human liver microsomes. Data in Table I show that N-oxygenation of 6 in the presence of human liver microsomes showed a profile similar to that of the cDNAexpressed HLFMO-D. Thus, the N-oxygenation of compound 6 was dependent upon the presence of NADPH and active protein and was quite sensitive to temperature in the absence of NADPH. Heat inactivation of human liver microsomes or cDNA-expressed HLFMO-D under conditions that destroyed FMO activity, but did not inactivate cytochromes P-450(25, 26), significantly decreased HLFMO-D-mediated N-oxygenation. In the pres-

Chem. Res. Toxicol., Vol. 6, No. 4, 1993 427 ence of human liver microsomes, S-oxidation of compound 6 was largely unaffected by heat inactivation. Thiourea (27) and thiobenzamide ( B )two , well-documented alternative substrates for animal FMO, significantly inhibited N-oxygenation of compound 6. However, thiourea and thiobenzamide also inhibited S-oxidation of compound 6. Because the selectivity of thiourea and thiobenzamide for human liver monooxygenases is unknown, interpretation of this result will have to await further work. n-Octylamine, a good inhibitor of cytochromes P-450 (29)and a modulator of animal liver FMO ( I ) , completely and significantly decreased N-oxygenation of compound 6 in the presence of cDNA-expressed HLFMO-D and human liver microsomes, respectively. Because a previous study showed a direct relationship between FMO substrate specificities with differences in the length of the alkyl side chain for a series of lO-[(N,Ndimethylamino)alkyl]-2-(trifluoromethyl)phenothiazines (13,we examined the same substrates with human liver microsome preparations and cDNA-expressed HLFMO-D. As shown in Table 11, N-oxygenation of chlorpromazine and 2-4 was detectable, but compounds 5 and 6, with longer side chains, were better substrates for human liver microsomes and cDNA-expressed HLFMO-D. Dichloromethane extracts of metabolic incubations with human liver microsomes and cDNA-expressed HLFMO-D were subjected to mass spectral analyses. The +LSIMS of the tertiary amine N-oxide metabolite of 5 isolated from pig liver microsomes was similar to the CI spectrum of the N-oxide metabolite isolated from human liver microsomes and cDNA-expressed HLFMO-D.

Discussion Because form A and D FMOs from animals (30) apparently differ in many important properties including substrate specificity (6, 311, enzyme stability (32),and other physical properties, we expressed HLFMO-D cDNA in bacteria and determined the aliphatic tertiary amine substrate specificity of the cDNA-expressed protein and compared the data with that obtained with adult human liver microsomes. The N- and S-oxygenation of nucleophilic amines and sulfur-containing compounds by hepatic FMO-A is well established (1,2).However, while FMO-A (i.e., pig liver) catalyzes N-oxygenation of secondary and tertiary amines, FMO-D (Le., guinea pig liver) apparently N-oxygenates primary aliphatic alkylamines as well as tertiary and secondary amines (6). As a result, n-octylamine, a positive effector for pig liver FMO, is actually a substrate for guinea pig liver FMO-D (6).n-Octylamine is also a substrate for the rabbit pulmonary form of FMO (11, 12). Results of studies of tertiary amines with adult human liver microsomes were comparable to those obtained with cDNAexpressed HLFMO-D. Thus, formation of N-oxide was strictly dependent on NADPH and active protein, was highly sensitive to heat inactivation under conditions which preserved about 80% of cytochrome P-450activity (13), and was inhibited by thiourea (27) and thiobenzamide (28)(Table I). Taken together, the results are consistent with a prominent role of HLFMO-D in compound 6 N-oxygenation. As discussed above, the effect of n-octylamine on N-oxygenation may stem from its action as a substrate for HLFMO-D. Although studies are limited, we suggest that, like other hepatic form D and rabbit pulmonary FMO-B enzymes examined, primary amines

428 Chem. Res. Toxicol., Vol. 6, No. 4,1993

Communications

Table I. Effect of Incubation Conditions on N- and 9-Oxygenation of 6 by Preparations of Adult Human Liver Microsomes and cDNA-ExDressed Flavin-Containing Monooxygenase (Form D) ~~

human liver microsomes [pmol/(min.mg of protein)] N-oxide S-oxide 1080 f 80 4870 f 130 8f3 ND NDb 120 f 20 570 f 130 3810 f 220 290 f 60 ND 20 f 4 ND 7f9 ND

incubation condition complete’ -NADPH -protein +heat inactivation +n-octylamine (4.5 mM) +thiourea (1.0 mM) +thiobenzamide (1.0 mM)

cDNA-expressed HLFMO-D [pmol/(min-mg of protein)] N-oxide S-oxide ND 180 f 40 ND 20f4 20f2 ND 30 f 3 ND ND ND 104 f 20 ND ND ND

a The complete system contained 50mM phosphate buffer (pH 8.4),the NADPH-generating system, 0.5 m M compound6,0.6mM DETAPAC, and 0.4 mg of humun liver microsomes or 2.3 mg of cDNA-expressed HLFMO-D in a final volume of 0.25 mL. The N-oxide and S-oxide were determined by HPLC as described in the Experimental Procedures. The results are the mean of 4 determinations (fSD). ND, not detectable; limit of detection was approximately 5 pmol/(min-mg of protein).

Table 11. N- and S-Oxygenation of Chlorpromazine and lo-[ (NJY-Dimethylamino)alkyl]-2-(trifluoromethyl)phenothiazines by Microsomes and cDNA-Expressed Flavin-Containing Monooxygenase (form D) from Adult Human LiveP

substrate chlorpromazine compound 2 compound 3 compound 4 compound SC

alkyl side chain n 3 2

3 4 5

human liver microsomes [pmol/(min.mg of protein)] N-oxide S-oxide 58.9 f 7.4 29.3 f 3.7 NDb 27.8 f 4.5 12.2 f 3.3 149.9 f 1.43 31.0 f 15.8 201.2 f 18.8 32.5 f 33.7 252.1 f 38.6

cDNA-expressed HFLMO-D [pmol/(min-mg of protein)] N-oxide S-oxide 10.2 f 1.2 ND 30.4 f 6.8 ND 53.2 f 12.9 ND 118.0 f 7.8 ND 200.0 f 45.7 ND

0 Incubations were performed as described in the Experimental Procedures and Table I with the exception that 0.1 mM substrate was present. The values are the mean of 5 determinations f SD. ND, not detectable; limit of detection 5 pmol/(min-mg of protein). In a separate experiment, compound 6 (alkyl side chain n = 6) gave similar values to the values listed above for compound 5.

may be N-oxygenated to provide the hydroxylamine. As expected, cytochrome P-450-mediated S-oxidation of compound 5 was strictly dependent upon NADPH and active microsomalprotein (7). Heat inactivation destroyed approximately 21 % of the S-oxidase activity, but thiourea and thiobenzamide completely inhibited S-oxidation. Inhibition of cytochromes P-450 by thiols or thiones may be via a direct mechanism or via reactive metabolites (Le., sulfenic acids) generated by HLFMO-D as suggested for other hepatic systems (26). 10-[(N,N-Dimethylamino)alkyl]-2-(trifluoromethyl)phenothiazineswith alkyl side chains varying in length from C2 to C7 have been examined as substrates for rabbit pulmonary and pig liver FMO (17). While all of the substituted phenothiazines were substrates for pig liver FMO (form A), only the phenothiazine tertiary amines possessing C6 and C7 alkyl side chains were detectable substrates for rabbit pulmonary FMO. The conclusion of the report (17) was that the hydroperoxyflavin of the active site in rabbit lung FMO :ested 6-8 A below the surface of a substrate binding channel with no more than an 8-A diameter along its longest channel axis. On the other hand, pig liver FMO (form A) binding channel is apparently more open and admits larger substrates to within 3 A of the hydroperoxyflavin, with the channel spanning as much as 12 A in diameter. Because the K,,, values for N-oxygenation of all of the 10-[(N,N-dimethy1amino)alkyllphenothiazines examined are similar and 1100 pM (data not shown), the conclusion from our structure-activity measurements suggest that the substrate binding channel leading to the active site hydroperoxyflavin of HLFMO-D is far more restricted than for the pig liver FMO-A. From the dimensions of the substrates examined (Table 11) it would appear that the active site hydroperoxyflavin of HLFMO-D is about 5-7 A below the surface of a narrow channel. The postulated dimensions of the substrate binding channel provide a

structural basis as to why detectable amounts of imipramine N-oxide was not observed to be formed in the presence of human liver microsomes (16). Although long alkyl side-chain 10-[(NJN-dimethylamino)alkyl1-2-(trifluoromethy1)phenothiazines (i.e., compounds 5 and 6) were oxidized at significant rates by cDNA-expressed HLFMO-D and human liver microsomes (Table 11),it was observed that short-chain phenothiazines were only oxidized with a modest rate. In this regard, HLFMO-D apparently N-oxygenates aliphatic tertiary amines with a specificity more similar to that of rabbit lung than that of pig liver FMO (17). It is possible that, for humans administered tertiary amines, the extent of N-oxygenation may depend on the steric constraints surrounding the nitrogen atom. In agreement with previous reports for other FMOs, HLFMO-D does not catalyze S-oxidation of the phenothiazine tricyclic sulfur atom (17, 18). This result is consistent with the mechanism proposed for animal FMOs which requires a “soft”, highly polarizable nucleophilic atom as oxygenatable substrate. That no detectable amount of 10-[(NJN-dimethylamino)alkyl1-2-(trifluoromethy1)phenothiazine S-oxides was detected in preparations of proteins from pTrcHLFM0-D-transformed or nontransformed E.coli suggeststhat nonenzymatic or nonHLFMO-D enzymatic oxidations were not occurring. The available data suggest that form A and form D FMO oxygenates nitrogen-containing compounds with distinct substrate specificities. The stereoselectivities of FMO forms A and D are different? which further points to distinct substrate binding domains for each FMO form. It is likelythat the determination of the mechanistic details of the adult human liver FMO-D will contribute to an understanding of the role of this monooxygenase in human 3S.B. Park, P.Jacob,III, N.Benowitz, and J. R.Cashman, unpublished results.

Chem. Res. Toxicol., Vol. 6, No. 4, 1993 429

Communications

drug and chemical biotransformations.

Acknowledgment. The authors acknowledge the helpful comments of Drs. Denes Medzihradszky and Katy Kuo Korsmeyer. We thank Prof. D. M. Ziegler for the phenothiazine substrates. The expert typing of Gloria Dela Cruz is gratefully appreciated. The work was financially supported by the National Institutes of Health and the University of California Tobacco-Related Disease Research Program (Grant 1IT0071). We acknowledge the generous help of the UCSF Bioorganic Biomedical Mass Spectrometry Resource (A. L. Burlingame, Director, supported by NIH Division of Research Resources Grant TT 016614).

References (1)Ziegler, D. M. (1980)Microsomal flavin-containing monooxygen-

ase: oxygenation of nucleophilic nitrogen and sulfur compounds. In Enzymatic Basis of Detoxification (Jakoby, W. B., Ed.) Vol. I, pp 201-227, Academic Press, New York, NY. (2) Ziegler, D. M. (1988)Flavin-containing monooxygenases: catalytic mechanism and substrate specificities. Drug Metab. Rev. 19,l-32, (3) Dolphin, C., Shephard, E. A., Povey, S., Palmer, C. N. A., Ziegler, D. M., Ayesh, R., Smith, R. L., and Philips, I. R. (1991)Cloning, primary sequence and chromosomal mapping of a human flavincontaining monooxygenase. J.Biol. Chem. 266, 12379-12385. (4) Lomri, N., Gu, Q., and Cashman, J. R. (1992)Molecular cloning of the flavin-containing monooxygenase (form 11) cDNA from adult human liver. Proc. Natl. Acad. Sci. U.S.A. 89,1685-1689. (5)Dolphin, C.T., Shephard,E. A.,Povey,S., Smith,R. L.,andPhilips, I. R. (1992)Cloning, primary sequenceand chromosomallocalization of FM02, a new member of the flavin-containing monooxygenase family. Biochem. J. 287,261-267. (6) Yamada, H., Yuno, K., Oguri, K., and Yoshimura, H. (1990) Multiplicity of hepatic microsomal flavin-containingmonooxygenase in the guinea pig: ita purification and characterization. Arch. Biochem. Biophys. 280,305-312. (7) Cashman,J.R.,Yang, 2.-C.,Yang,L.,and Wrighton,S. (1993)Stereo-

and regioselectiveN- and S-oxidation of tertiary amines and sulfides in the presence of adult human liver microsomes. Drug Metab. Dispos. (in press). (8) Sadeque, A. J. M., Eddy, A. C., Meier, G. P., and Rettie, A. E. (1992) Stereoselective sulfoxidation by human flavin-containing monooxygenase. Drug Metab. Dispos. 20,832-839. (9) Williams, D. E., Hale, S. E., Muerhoff, A. S., and Masters, B. S. S. (1986)Rabbit lung flavin-containing monooxygenase: purification, characterization and induction duringpregnancy. Mol. Pharmacol. 28,381-390. (10) Tynes, R. E., Sabourin, P. J., and Hodgson, E. (1985)Identification

of distinct hepatic and pulmonary forms of microsomal flavincontaining monooxygenase in the mouse and rabbit. Biochem. Biophys. Res. Commun. 126,1069-1075. (11) Williams, D.E., Hale, S. E., Ziegler, D. M., Nordin, D. J., Hale, S. E., and Masters, B. S. S. (1984)Rabbit lung FMO is immunochemically and catalytically distinct from the liver enzyme. Biochem. Biophya. Res. Commun. 126,116-122. (12) Tynes, R. E., Sabourin, P. J., Hodgson,E., and Philpot, R. M. (1986)

Formation of hydrogen peroxide and N-hydroxylated amines catalyzed by pulmonary flavin-containing monooxygenase in the presence of primary alkylamines. Arch. Biochem. Biophys. 261, 654-664. (13) Cashman, J. R., Park, S. B.,Yang, 2.-C., Wrighton, S., Jacob, P., 111, and Benowitz, N. (1992)Metabolism of nicotine by human liver microsomes: Stereoselective formation of trans- nicotine "-oxide. Chem. Res. Toxicol. 6,639-646.

(14) Gold, M. S.,and Ziegler, D. M. (1973)Dimethylaniline N-oxidase

and aminopyrine N-demethylase activities of human liver tissues. Xenobiotica 3, 179-189. (15) McManus, M. E., Stupans, I., Burgess, W., Koenig, J. A, Hall, P. dela M., and Birkett, D. J. (1987)Flavin-containing monooxygenase activity in human liver microsomes. Drug Metab. Dispos. 16,256261. (16) Lemoine,A., Johann, M., and Cresteil, T. G. (1990)Evidence for the

presence of distinct flavin-containing monooxygenases in human tissue. Arch. Biochem. Biophys. 276,336-342. (17)Nagata, T., Williams, D. E., and Ziegler, D. M. (1990)Substrate specificities of rabbit lung and porcine liver flavin-containing monooxygenases: Differences due to substrate specificity. Chem. Res. Toxicol. 3, 372-376. (18)Lomri, N., Thomas, J., and Cashman, J. R. (1993)Expression in Escherichia coli of the cloned flavin-containingmonooxygenasefrom pig liver. J. Biol. Chem. 268,5048-5059. (19) Sofer, S.S.,and Ziegler, D. M. (1978)Microsomal mixed-function

amine oxidase: oxidation producta of piperazine-substituted piperazine-substituted phenothiazine drugs. Drug Metab. Dispos. 6 , 232-239. (20) Rigby, P. W. J., Dieckmann, M., Rhodes, C., and Berg, P. (1977) Labeling deoxyribonucleic acid to high specific activity in vitro by nick translation with DNA polymerase I. J. Mol. Biol. 113,237251. (21) Sanger, F., Nicklen, S., and Coulsen, A. R. (1977)DNA sequencing with chain-twminating inhibitors. Proc. Natl. Acad. Sci. U.S.A. 74,5463-5467. (22) Craig, S.P., 111,Yuan, L., Kuntz, D. A., McKarrow,J. H., and Wang, C. C. (1991)High level expression in Escherichia coli of soluble,

enzymatically active schistosomal hypoxanthine/guanine phosphoribosyltransferase and trypanosomal omithiie decarboxylase. Proc. Natl. Acad. Sci. U.S.A. 88,2500-2504. (23) Cashman, J. R. (1987)A radiometric assay for flavin-containing monooxygenase activity. Anal. Biochem. 160,294-300. (24)Cashman,J. R., and Yang, Z.4. (1990)Analysisof amine metabolites

by high-performanceliquid chromatography on silicagel with a nonaqueous ionic eluent. J. Chromatogr. 632,405-410. (25) Cashman, J. R., Proudfoot, J., Pate, D. W., and Hogberg, T. (1988) Stereoselective N-Oxygenation of Zimeldine and Homozimeldine by the flavin-containing monooxygenase. Drug Metab. Dispos. 16, 616-622. (26) Decker, C. J.,Cashman, J.R., Sugiyama,K., Maltby,D., andcorreia, M. A. (1991)Formation of glutathionyl-spironolactonedisulfides by rat liver cytochrome P-450 or hog liver flavin-containing monooxygenase: A functional probe of 2 electron oxidation of the thiosteroid? Chem. Res. Toxicol. 4, 670-677. (27)Poulsen, L. L., Hyslop, R., and Ziegler, D. M. (1974)S-Oxidation of

thiourylenes catalyzed by a microsomalflavoprotein mixed-function oxidase. Biochem. Pharmacol. 23,3431-3440. (28) Cashman, J. R., and Hanzlik (1982)Oxidation and other reactions of thiobenzamide derivatives of relevance to their hepatotoxicity.

J. Org. Chem. 47,4645-4650. (29) Jefcoate, C. R. E., Gaylor, J. L., and Calabrese, R. L. (1969)Ligand interactionswith cytochromeP-450 I. Binding of primary amines. Biochemistry 8, 3455-3465. (30) Gasser, R., Tynes, R. E., Lawton, M. P., Korsmeyer, K. K., Ziegler, D. M., and Philpot, R. M. (1990)The flavin-containing monooxygenase expressed in pig liver: primary sequence, distribution, and evidence for a single gene. Biochemistry 29,119-124. (31) Hlavica, P., and Golly, I. (1991)On the genetic polymorphism of the flavin-containingmonooxygenase, in N-OxidationofDrugs (Hlavica, P., and Damani, L. A., Eds.) pp 72-90,Chapman and Hall,London, U.K. (32) Ozole.,J. (1991)Multiple forms of liver microsomalflavin-containing monooxygenases: Complete covalent structure of form 2. Arch. Biochem. Biophys. 290,103-116.