ARCHIVES
OF BIOCHEMISTRY
Vol. 284, No. 2, February
AND
BIOPHYSICS
1, pp. 400-406, 1991
Role of Cytochrome P450 IA2 in Acetanilide Hydroxylation as Determined with cDNA Expression and Monoclonal Antibodies Gao Liu, Harry Laboratory
V. Gelboin,
of Molecular
and Michael
Carcinogenesis, National
4-
J. Myers’ Cancer Institute,
Bethesda, Maryland
20892
Received June 26, 1990, and in revised form October 4, 1990
INTRODUCTION The role of P450 IA2 in the hydroxylation of acetanilide was examined using an inhibitory monoclonal antibody (MAb) 1-7-1 and vaccinia cDNA expression producing murine P450 IA1 (mIAl), murine P450 IA2 (mIA2), or human P450 IA2 (hIA2). Acetanilide hydroxylase (AcOH) activity was measured using an HPLC method with more than 500-fold greater sensitivity than previously described procedures. This method, which does not require the use of radioactive acetanilide, was achieved by optimizing both the gradient system and the amount of enzyme needed to achieve detection by uv light. MAb 1-7-1 inhibits up to 80% of the AcOH activity in both rat liver microsomes and cDNA expressed mouse and human P450 IA2. MAb 1-7-1, which recognizes both P450 IA1 and P450 IA2, completely inhibits the aryl hydrocarbon hydroxylase (AHH) activity of cDNA expressed in IAl. The inhibition of only 80% of the AHH activity present in MC liver microsomes by MAb 1-7-1 suggests that additional P450 forms are contributing to the overall AHH activity present in methylcholanthrene (MC)-liver microsomes as MAb 1-7-1 almost completely inhibits the AHH activity of expressed mIA1. Maximal inhibition of IA2 by 1-7-1 results in an 80% decrease in acetanilide hydroxylase activity in both liver microsomes and expressed mouse and human IA2. The capacity of MAb 1-7-1 to produce identical levels of inhibition of acetanilide hydroxylase activity in rat MC microsomes (80%) and in expressed mouse (81%) and human P450 IA2 (80%) strongly suggests that P450 IA2 is the major and perhaps the only enzyme responsible for the metabolism of acetanilide. These results demonstrate the complementary utility of monoclonal antibodies and cDNA expression for defining the contribution of specific P450 enzymes to the metabolism of a given substrate. This complementary approach allows for a more precise determination of the inhibitory capacity of MAb with respect to the metabolic capacity of the target P450. o 1991 Academic
Press,
Inc.
Cytochrome P450 enzymes are responsible for the metabolism and activation of a vast array of endogenous and exogenous chemicals. The capacity to metabolize this large number of substrates is due, in part, to a large multiplicity of P450 enzymes, many of which possess overlapping substrate specificities. Monoclonal antibodies (MAbs)2 have proven to be useful tools toward delineating the role of epitope-specific P450 enzymes in the metabolism and activation of specific substrates. One such MAb, 1-7-1, recognizes and binds to both P450 IA1 and P450 IA2 (1). The overlapping P450 binding specificity of 1-71 is likely due to a common epitope resulting from the high degree of sequence homology of these two enzymes. MAb 1-7-1 has been used in RIA’s, immunopurification procedures, and enzyme inhibition assays (2-4) to characterize the substrate specificity and role of this class of P450 enzymes. However, due to the recognition of both P45Os by MAb l-7-1, it has not always been possible to determine the relative contribution of each of these two P45Os in the metabolism of selected substrates in tissue sources such as liver microsomes (5, 6). As a means to resolve the relative contributions of IA2 toward the metabolism/activation of selected substrates, we are developing site-specific monoclonal antibodies, generated using synthetic peptide antigens from regions unique to a given P450 (7). We have used the metabolism of acetanilide as an indicator of IA2 activity. While the hydroxylation of acetanilide is an activity associated with ’ To whom correspondence should be addressed at: U.S.F.D.A., Division of Veterinary Medical Research, BARC-EAST, Bldg. 328A, Beltsville, MD 20705. ’ Abbreviations used: AcOH, acetanilide hydroxylase; AHH, aryl hydrocarbon hydroxylase; ZOH-AA, 2-hydroxyacetanilide; 30H-AA, 3hydroxyacetanilide; IOH-AA, 4-hydroxyacetanilide; CO, control; MC, 3-methylcholanthrene; PB, phenobarbital; r, rat; m, mouse; h, human; MAb, monoclonal antibody;
400 All
Copyright 0 1991 rights of reproduction
0003.9861/91 $3.00 by Academic Press, Inc. in any form reserved.
ACETANILIDE
HYDROXYLASE
P450 IA2 (8), it is not certain that this is the sole enzyme capable of this activity. Therefore, the role of P450 IA2 in acetanilide hydroxylation was determined in two ways. First, the activity of cDNA-expressed P450 IA2 was determined for acetanilide hydroxylation and second the inhibition of acetanilide hydroxylation by MAb 1-7-1 was determined for the cDNA-expressed IA2 and microsomes. This complimentary approach, using both cDNA expression and monoclonal antibodies, permits a more precise assessment of the inhibitory capacity of the MAb with respect to its target P450, which in turn aids in the elucidation of the role of that form toward the total metabolism of the substrate in question. The present study describes how the combined use of the vaccinia virus cDNA expression system and an inhibitory MAb (MAb 1-7-1) permits a more accurate assessment of the contribution of P450 IA2 toward the metabolism of acetanilide in tissues. In the course of this study, we developed a quantitative HPLC assay for measuring the hydroxylation products of acetanilide using a nonradioactive substrate. Previously published HPLC-based detection methods depended upon the use of radioactive acetanilide (8,9). Our methodology is more sensitive (500-fold); thus, the method we describe may be a more useful assay than previously described methods (8, 9). This report describes the use of both monoclonal antibodies and cDNA-expressed P45Os to ascertain the role of a P450 enzyme in the metabolism of a given substrate in a crude tissue. MATERIALS
AND
METHODS
Preparation of microsomes and uaccinia virus-expressed P450. Liver microsomes were prepared from control, 3-methylcholanthrene-treated or phenobarbital-treated male Sprague-Dawley rats and NZB or Balb/ c mice, as previously described (10). Briefly, after 3 days of injection with 3-methylcholanthrene or 7 days with phenobarbital, the animals were sacrificed and the livers perfused with ice-cold 0.15 M KCl. The microsomal fraction was obtained by differential centrifugation of the S-9 fraction and was stored in 25% glycerol at -70°C. Enzymatically active mouse cytochrome P450 IAl, mouse IA2, and human IA2 cDNAs (kind gifts of Dr. Frank Gonzalez) were expressed in HepG2 cells using the vaccinia virus cDNA expression system as previously detailed (ll14). Twenty-four hours after viral infection of the HepG2 cells, the intact cells were collected and stored as a cell pellet at -70°C until used. Prior to use, the cell pellets were thawed on ice and briefly sonicated after 1 ml of distilled H,O was added to the vial. Protein content was determined using Pierce’s BCA methodology (Pierce). Monoclonal antibody preparation. Cell lines producing monoclonal antibodies were generously supplied by Drs. Sang Shin Park (MAb l7-l) and Sandra Smith-Gill (HyHeL9). MAb l-7-1 has been shown to bind to P450 IA1 and IA2 (1). The capacity of this MAb to inhibit the AHH activity of P450 IA1 has been extensively characterized with respect to rodent liver microsomes whereas its inhibitory capacity toward IA2 has not been completely characterized. HyHeL9 is a MAb which recognizes hen egg lysozyme, does not bind to P45Os, and is used as a negative MAb control (1). Both MAbs are of the IgGl subclass. The monoclonal antibodies were purified from ascites by hydroxylapatite chromatography (15) with the protein content determined as above. Acetanilide hydroxylase (ACOH) and aromatic hydrocarbon hydroxylase (AHH) assays. The metabolism of acetanilide was determined by its
ACTIVITY
OF CYTOCHROME
IA2
401
incubation with liver microsomes or expressed P45Os in the presence of: Tris buffer (50 mM, pH 7.5), MgCls (0.30 mM), NADPH (0.6 pmol), and BSA (1 mg) in a volume of I ml. The reaction was initiated by the addition of 25 ~1 of a 20 mM acetanilide solution (in acetone) unless otherwise specified. This amount represents a saturating concentration of acetanilide. In antibody-mediated enzyme-inhibition experiments, the reaction mixture was preincubated with either purified monoclonal antibody l-7-1 or HyHel9 (1) (Ab control) at room temp for 5 min. The reaction was terminated by the addition of 2.5 ml of ice-cold ethyl acetate (HPLC grade) and 0.1 ng of 3-hydroxyacetanilide (in 10 ~1 of acetone). The samples were vigorously mixed and centrifuged briefly. The ethyl acetate extract was removed, and a second 2.0-ml ethyl acetate extract was pooled with the first. The organic phase was then evaporated under N2 and the residue redissolved in 200 ~1 of methanol (HPLC grade). Twenty microliters of each sample was subsequently used for HPLC analysis. 3.Hydroxyacetanilide was added to each reaction tube as an internal standard to estimate the recovery of 4-hydroxyacetanilide. The 3-hydroxyacetanilide was reported to account for 0.1% or less of the total concentration of metabolites (8,9). In our study, we did not detect 30H-acetanilide. The recovery of 3-hydroxyacetanilide was typically 94 f 4%. AHH activity was determined according to the method of Nebert and Gelboin (16). Analysis of the metabolites was performed using HPLC analysis. reverse phase HPLC. A Spectra Physics Model SP8100 liquid chromatograph with an Valco injector, an autosampler, and a DuPont ODS column (4.6 X 25 cm) was used. A gradient system was used to separate the metabolites from material originating with the enzyme source. The solvents used were water:methanol:acetonitrile (HPLC grade), with a starting ratio of 78:17:5. Acetonitrile was kept at a constant ratio (5%) throughout the entire procedure. The solvent ratio of water and methanol changed by 0.5%/min for 3 min for a solvent ratio of 76.5:18.5:5, then changed again to a ratio of 55:4@5 in the next 2 min. This ratio was held constant for 5 min, after which the solvent conditions returned to the starting ratio of 78:17:5 in the next 2 min. The metabolites were detected using a Spectra-Physics Model 100 detector at a wavelength of 254 nm. The retention times for the 4-hydroxyacetanilide, 3-hydroxyacetanilide, 2-hydroxyacetanilide and the parent substrate acetanilide were 5.3, 7.5, 10.2, and 13.5 min, respectively.
RESULTS
Previously published HPLC-based methodologies for assessing acetanilide hydroxylase activity have relied on the use of radioactive acetanilide, with uv detection being achieved by the addition of excess amounts of nonradioactive acetanilide and its hydroxylated metabolites upon termination of the reaction (8, 9). However, the HPLC-solvent gradients used in these studies do not adequately resolve the 40H-acetanilide product from a contaminant originating from the microsomes. Therefore, we developed a new gradient system that permitted resolution of the three hydroxyacetanilide products and their separation from this contaminant. Thus, we were able to use nonradioactive acetanilide as the starting substrate and determine the kinetics of the reaction with respect to time, substrate concentration, and protein (microsome) concentration. With the use of saturating amounts of substrate (as suggested in Refs. (8, 9)) and 100 pg of rat MC microsomes (Fig. 1) the production of 4-hydroxyacetanilide (40H-AA) displays linear kinetics with respect to time. Similarly, the kinetics of acetanilide hydroxylase activity is linear from 10 to 200 pg with MC microsomes and saturating amounts of substrate (Fig. 2a). When the
402
LIU,
GELBOIN.
AND
MYERS
in close agreement with those obtained by Guenthner
et
al. (9).
:
Microsomes obtained from either control or MC- or PB-treated rats were preincubated with monoclonal antibody 1-7-1 followed by assessment of acetanilide hydroxylase activity. MAb 1-7-1 maximally inhibited the AcOH activity of rat MC microsomes at a concentration of 100 pg MAb to 100 pg of microsomes, producing 83% inhibition of AcOH activity (Fig. 4). Both control and PB microsomes exhibited maximal inhibition at a concentration of 200 pg of MAb 1-7-1 to 100 pg of microsomes, resulting in 67 and 73% inhibition of the AcOH activity
/
c
300 : 0
0,.:li, 10
20
30
(
,
40
50
1 60
70
TIME (min)
7rA-
FIG. 1. Formation of 4-hydroxyacetanilide as a function of time. MCinduced rat liver microsomes (100 Kg) were incubated with a saturating amount of acetanilide (0.5 mM) for 5 to 60 min. Assay conditions and product detection are as detailed under Materials and Methods. Each point represents the mean I~I SE of duplicate determinations of a representative assay. Similar results were obtained upon repetition with different microsomal preparations.
concentration of microsomes is further increased, acetanilide hydroxylase activity no longer displays absolute linearity (Fig. 2b) but rather is curvilinear. The lack of linearity of AcOH with high levels of microsomes is in agreement with Guenthner et al. (9). It should also be noted that production of the 2-hydroxyacetanilide metabolite is detected only under conditions in which the concentration of microsomal protein is at or near the nonlinear phase of the reaction (Fig. 2b). In addition, the concentration of SOH-AA does not increase with increasing amounts of microsomes as does the 40H-AA metabolites. The acetanilide hydroxylase activity of control microsomes and phenobarbital microsomes were three- to fourfold less than that observed with rat MC microsomes (Table I). The AcOH activity of a human liver microsomal preparation demonstrated 40HAA levels comparable to those seen with uninduced rat microsomes (Table I). In contrast to the rat, uninduced liver microsomes from either an AHH inducible strain (Balb/c) or a noninducible strain (NZB) had levels of 40H-AA comparible to those seen in rat MC-induced microsomes. MC induction had no effect on AcOH activity in NZB mice, whereas Balb/c microsomes had a threefold induction in activity (Table I). Lastly, under assay conditions in which acetanilide hydroxylase activity was linear with respect to both time and microsomal content, the addition of increasing amounts of acetanilide without a concomitant increase in enzyme activity demonstrates enzyme saturation with substrate (Fig. 3). The apparent Km and V,,, values are 0.303 mM and 3.38 nmol/mg protein/min, respectively (data not shown). These values are
0
50
100
150
200
l v
4 OH-AA 2 OH-AA
250
60 I
10 1
0 td 0
400
800
1200
MC Mic Cone
1600
2000
2400
(ug)
FIG. 2. Formation of 4-hydroxyacetanilide as a function of the concentration of MC rat liver microsomes. Increasing amounts of microsomes were incubated with a saturating amount of acetanilide (0.5 mM). Assay conditions and product detection are as detailed under Materials and Methods and the legend to Fig. 1. The production of 4-hydroxyacetanilide was determined using microsomes at concentrations ranging from 10 to 200 Kg (a) or from 10 to 2000 pg (b). The production of the 2-hydroxyacetanilide metabolite was consistently detected only when the concentration of microsomes was above 200 fig/reaction. Each point represents the mean k SE of duplicate determinations of a representative assay. Similar results were obtained upon repetition with different microsomal preparations.
ACETANILIDE TABLE
HYDROXYLASE
ACTIVITY
OF CYTOCHROME
IA2
I
Acetanilide Hydroxylase Activity of Induced and Uninduced Rat and Mouse Liver Microsomes and Human Liver Microsomes” Species/strain
In uiuo
Mouse/Balb/c
MC None MC None MC PB None None
Mouse/NZB Rat/SD
Human
nmol/OH-AA/20 98.9 28.5 29.4 29.8 33.4 12.3 12.0 16.5
min *
t 6.2
f 1.1 t ?I zk + + k
0.7 4.8 3.2 0.6 0.7 1.6 0
’ ACOH activity was determined as described under Materials and Methods. The tissue samples were obtained as detailed under Materials and Methods. Acetanilide was used at a final concentration of 0.5 mM. * Results are from a representative experiment. All experiments were performed at least twice, with duplicate determinations for each condition. Similar results were obtained with different preparations of microsomes.
(Fig. 4). MAb Z-66-3, which inhibits the activities of P45Os IIBl and IIB2, had no effect on acetanilide hydroxylase activity in PB microsomes (data not shown). MAb 1-7-1 binds to both P450 IA1 and P450 IA2. Expressed mIA1, however, does not exhibit AcOH activity (Table II). Thus, the inhibitory effect of MAb 1-7-1 on microsomes suggests that P450 IA2 is the major enzyme responsible for the metabolism of acetanilide. However, since MAb 1-7-1 in-
c 1 0, a
50
100
150
MAD 1-7-l
250
30”
350
(ug)
FIG. 4. Inhibition of acetanilide hydroxylase activity in rat liver microsomes using monoclonal antibody 1-i-l. Control (CO), phenobarbital (PB), or MC rat liver microsomes (100 pg) were incubated with a saturating amount of acetanilide (0.5 mM) for 20 min with increasing amounts of either MAb 1-7-1 or control MAb (HyHeL9). Maximal ACOH activities for CO, PB, or MC microsomes were similar to those detailed in Table I. Assay conditions and product detection are as detailed under Materials and Methods and Fig. 1. The results are expressed as percentages of the control MAb response. Each point represents the mean k SE of duplicate determinations of a representative assay. Similar results were obtained upon repetition with different microsomal preparations.
hibited acetanilide hydroxylase activity maximally at 80%, this suggests that either 1-7-1 was not capable of completely inhibiting AcOH activity or that additional P450 form(s) contribute to the metabolism of acetanilide in microsomes. These possibilities were examined by determining the capacity of MAb 1-7-1 to inhibit the AcOH
30
TABLE Acetanilide
25
200 Cone
Hydroxylase
Activity
II of Vaccinia-Virus
Expressed
Mouse and Human Cytochrome P450 IA1 and IAZ”
F q20
nmol 40H-AA/reaction:
& e
15
E c
10
1
3
5
(mg cell lysate/reaction) 5 C 30
C2
04 Substrate
Cone
06
08
(mM)
FIG. 3. Substrate saturability of the acetanilide hydroxylase activity of 3-methylcholanthrene (MC) rat liver microsomes. Microsomes (100 ug) were incubated with increasing amounts of acetanilide at 37°C for 20 min. The extent of activity was determined by measuring the amount of the main metabolite, 4-hydroxyacetanilide, that was present in each sample via HPLC coupled with uv detection. Each point represents the mean + SE of duplicate determinations of a representative assay. Similar results were obtained upon repetition with different microsomal preparations.
mIAl* mIA2 hIA
0.0 -c 0.51 k 0.10
0.0 1.18 f 0.03 1.16 i- 0.05
0.01 + 0.00 -c -c
a ACOH activity was determined as described under Materials and Methods. The cDNA expressed P45Os were obtained as detailed under Materials and Methods. The acetanilide concentration used was 0.5 mM. No ACOH activity was observed when cell lysates from uninfected Hep G2 cells were examined. b Results are from a representative experiment. All experiments were performed at least twice, with duplicate determinations for each condition. Similar results were obtained with different preparations of expressed P45Os; h, human; m, mouse. ’ Not determined.
404
LIU,
GELBOIN,
activity of vaccinia-virus expressed P450 IA2 (mouse and human) and its closely related form IAl. AcOH activity as previously shown, was exhibited by murine and human P450 IA2, whereas murine P450 IA1 showed negligible AcOH activity (Table II). In contrast, AHH activity is a function only of P450 IA1 (59.0 pmol/mg protein/min). Furthermore, the AHH activity of mIA1 was almost completely inhibited (>97%; 1.4 pmol/mg protein/min) by MAb 1-7-1 while the AHH activity of rat MC liver microsomes (2927.1 pmol/mg/min) was maximally inhibited by only 85% (449.0 pmol/mg/min). MAb 1-7-1 produced an 81% decrease in the AcOH activity of expressed murine P450 IA2 at a concentration of 16 pg MAb/mg cell lysate (Fig. 5), which represents the maximum extent to which this activity can be inhibited by 1-7-1. MAb 1-7-1 inhibited the AcOH activity of uninduced Balb/c and NZB liver microsomes and MC-induced NZB microsomes by 90% at a concentration of 50 pg MAb/lOO /*g microsomes. MAb 1-7-1 inhibits the activity of MC-induced Balb/c microsomes by 90% at a concentration of 300 pg MAb/ 100 pg microsomes. Statistical analysis (Student’s t test) demonstrated no significant differences between the maximum inhibition of the expressed mIA2 and any of the mouse liver microsomes (data not shown). MAb 1-71 resulted in an 80% inhibition of the AcOH activity of expressed human IA2 (Fig. 6), but in contrast to mouse
0 MC-E&lb/c V MC-NZB V CONT-Balb/c 0 CONT-NZB . mlA2
0
50
100
150 MAb l-7-1
200
250
AND
MYERS
0 L-L.-, 0
300
600 MAb l-7-;
900
'200
kc,:,
(us)
FIG. 6. Inhibition of acetanilide hydroxylase activity in human liver microsomes are expressed human IA2. Conditions are as outlined in the legend to Fig. 5.
IA2, produced maximal inhibition at a concentration of 400 pug MAb/mg cell lysate, a 24-fold increase in the amount of MAb needed as compared to the amount needed to produce maximal inhibition of the AcOH activity of expressed murine IA2. MAb 1-7-1 maximally inhibits the AcOH activity in human microsomes by 63% at a concentration of 900 pg MAb/l50 pg microsomes. Attempts to achieve higher inhibition were not possible as the negative MAb control had a nonspecific inhibitory effect on AcOH activity above a concentration of 400 pugMAb/lOO pg microsomes. DISCUSSION
300
3%
(ug)
FIG. 5. Inhibition of acetanilide hydroxylase activity of expressed murine IA2, control (CO) Balb/c, control NZB, MC Balb/c, or MCNZB liver microsomes using monoclonal antibody l-7-1. HepG2 cell lysate (1 mg) or rat liver microsomes (100 pg) were incubated with a saturating amount of acetanilide (0.5 mM) for 20 min with increasing amounts of either MAb 1-7-l or control MAb (HyHeL9). The HepG2 cells were infected with vaccinia virus which contained the cDNA for either human IA2 (hP3) or murine IA2. Maximal AcOH activities for either the expressed P45O’s or MC microsomes were similar to those detailed in Table I. Assay conditions and product detection are as detailed under Materials and Methods and the legend to Fig. 1. The results are expressed as percentages of the control MAb response. Each point represents the mean t SE of duplicate determinations of a representative assay. Similar results were obtained upon repetition with different sample preparations.
The results presented in this report demonstrate the utility of the combined use of monoclonal antibodies and cDNA-expressed cytochrome P45Os. The cDNA-expressed P45Os permit a precise definition of the substrate and product specificity of a given P450 form. The inhibitory MAb measures the minimum contribution of the P45O(s) to which it binds. Thus, the cDNA-expressed P450 yields substrate specificity information but does not portray specific P450 contributions toward substrate metabolism in a crude tissue, whereas the MAb measures the minimum contribution of a given P450 form (or forms) to the metabolism of a substrate. This complementary approach was demonstrated in this report with respect to the metabolism of acetanilide and benzo(a)pyrene. The expressed P45Os as reported previously demonstrate that benzo(a)pyrene is metabolized by P450 IA1 and not IA2, whereas acetanilide is metabolized by IA2 but not IA1 (Table II). MAb 1-7-1 inhibits both the AHH activity and the AcOH activity of MC-induced microsomes by 80%. These results taken independently suggest that while the metabolism of these two substrates is mutually exclusive with respect to P45Os
ACETANILIDE
HYDROXYLASE
IA1 and IA2, the residual 20% of AHH and AcOH activities may be other P450 enzymes which are capable of metabolizing these two substrates. A different picture emerges however if both methodologies are used together. MAb l-7-1 almost completely inhibits the metabolism of benzo(a)pyrene by expressed P450 IAl. Thus when MAb 1-7-1 is maximally inhibiting the AHH activity of P450 IAl, any residual activity almost certainly is the result of other P45Os. In contrast, MAb l-7-1 inhibited the AcOH activity of tissue-culture-expressed mouse and human IA2 by 80%; absolute inhibition was never obtained. This same degree of inhibition was also obtained when rat MC-induced liver microsomes or mouse liver microsomes (induced or uninduced) were examined. Thus, when MAb 1-7-1 is maximally inhibiting AcOH activity, the residual AcOH activity is possible due to P450 IA2, which is not inhibitable by MAb 1-7-1 rather than by additional P450 enzymes. One possible explanation may be that the portion of AcOH activity of cDNAexpressed IA2 not inhibited by 1-7-1 could also be due to enzyme which is inserted into the microsomal membrane such that the 1-7-l binding site is not accessible. Another aspect of the inhibition of AcOH activity of expressed mouse and human IA2 by MAb 1-7-1 is that a 24-fold higher concentration of antibody is needed to inhibit the expressed human IA2 compared to the amount needed to obtain a similar level of inhibition with the murine IA2. The need for more antibody is not due to differences in the amount of enzyme present since both mouse and human IA2 preparations contained equal amounts of enzyme activity (Table II) as well as similar spectral activity and equi-intense staining upon Western blot analysis using polyclonal antisera (data not shown). Human liver microsomes also required increased amounts of MAb 1-7-1 as compared to the rodent to achieve maximal inhibition of AcOH. Thus, the increased amount of 1-7-l needed to induce maximal inhibition of AcOH activity in human IA2 may be due to differences in the binding affinity of 1-7-l for human IA2 compared to rodent. IA2. This suggests that either the epitope recognized by 1-7-1 is not completely accessible in human P450 IA2 or, that the complete 1-7-1 binding epitope is not present in human IA2. This report also demonstrated that MAb 1-7-1 interacts with human IA2. Previous attempts failed to detect IA2 in human liver microsomes in Western blot analysis and standard alkaline phospatase detection methods (Friedman, unpublished results). That IA2 is indeed present in human liver microsomes was recently demonstrated using either antisera generated against synthetically prepared peptide sequences from IA2 (7) or polyclonal anti-rat P450 IAl/IA2 antisera (17). cDNA-expressed human IA2 also did not yield a positive signal with Western blot analysis using MAb 1-7-1 (data not shown). However, cDNA-expressed mouse IA2 also did not produce a positive signal after Western blot analysis using 1-7-1 whereas IA1 and
ACTIVITY
OF CYTOCHROME
IA2
405
IA2 in mouse MC liver microsomes are detectable with 1-7-1. The inability to immunodetect the cDNA-expressed forms of IA2 may be due to the relatively low levels of P450 present in the cell lysate or the lower affinity of l7-l for IA2 relative to IA1 (18). Both forms of expressed IA2 yield a positive ELISA signal (data not shown) which, as expected, is lower than that obtained for an equal amount of tissue-culture-expressed IA1 due to the lower affinity of 1-7-1 for IA2 relative to that for IAl. The inhibition kinetics of MAb 1-7-1 toward the AcOH activity of PB and CO microsomes show that more MAb 1-7-1 is needed to achieve maximum inhibition of the AcOH activity in these tissues than in MC-induced microsomes which have more P450 IA2 than either PB or CO microsomes. This suggests that either 1-7-1 is binding to another P450 which has AcOH activity or that it more likely is binding to both active and inactive proteins. Guengerich et al. (19) and Robinson et al. (20) have shown that for many P450 forms, the total amount of P450 protein is greater than the amount of spectrally active protein. The HPLC-based methodology used in this report for determining acetanilide hydroxylase activity has several advantages over previously published methodologies. One advantage is in the use of nonradioactive acetanilide as the substrate. The capacity to use a nonradioactive substrate was due to the use of a new HPLC gradient system. This system permitted resolution of the 40H-AA product from a substance of microsomal origin which comigrated with the 40H-AA product when previously published HPLC gradient systems were used. Another is in the amount of tissue sample needed to detect AcOH activity; as little as 10 pg of MC-induced liver microsomes resulted in detectable activity (Fig. 2a). This by itself represents a 50-fold more sensitive assay system than any previously published. Both of the previous methods used the entire sample for HPLC analysis (8,9). However, as we routinely examined only l/10 of the total reaction sample on the HPLC, this actually represents a 500-fold increase in sensitivity. Therefore, if all of the sample was to be examined, the lower level of sensitivity of the assay with respect to the amount of tissue sample needed drops to approximately 1 pg. Thus this methodology should be of value in determining the AcOH activity in tissues which have a lower P450 content than the liver or for tissues which are difficult to obtain in large quantities. The second main metabolite, 2-hydroxyacetanilide, only appeared when the concentration of liver microsomes was at or near the nonlinear portion of the curve, and in contrast to the 4-hydroxy metabolite, this metabolite did not increase with increasing amounts of microsomes (Fig. 2b). Two possible explanations are that SOH-AA is further metabolized by other enzymes or is not a product of IA2mediated metabolism. 20H-AA has been shown to be further metabolized to 2-acetamidohydroxyquinone through an NADPH-dependent pathway.
406
LIU,
GELBOIN,
The mechanism by which MAb 1-7-1 inhibits AcOH activity is unknown. In this regard, the results of Furuya et al. are intriguing (21). They demonstrated that almost any single amino acid alteration in the primary sequence of rat IA2 resulted in significant loss of enzyme activity. Their results suggest one possible mechanism by which MAb 1-7-1 is affecting IA2, that of conformation alteration of the enzyme by 1-7-1. In summary, the major P450 responsible for the metabolism of acetanilide is P450 IA2. MAb l-7-1, which recognizes both IA1 and IA2, inhibits up to 80% of the AcOH activity in both liver microsomes and tissue-culture-expressed IA2. In addition, MAb 1-7-1 completely inhibits the AHH activity of expressed IAl. As MAb l7-l maximally inhibits AHH activity of MC-induced liver microsomes by BO-85% this suggests that additional P450 forms are contributing to the overall AHH activity present in liver microsomes. We have also described a new HPLCbased methodology for detection of ACOH activity which is at least 500-fold more sensitive than previously published procedures. ACKNOWLEDGMENTS We thank Drs. Sang Shin Park and Sandra Smith-Gill for providing the monoclonal antibodies and Dr. Frank Gonzalez for the vaccinia virus containing the P450 cDNAs. We would also like to thank Dr. Fred Friedman for critically reviewing this manuscript. The assistance of Mr. Richard Robinson in the development of the HPLC-gradient system is deeply appreciated.
REFERENCES 1. Park, S. S., Fujinio, T., West, D. Guengerich, F. P., and Gelboin, H. V. (1982) Cancer Res. 42, 1798-1808. 2. Friedman, F. K., Park, S. S., and Gelboin, H. V. (1985) Biochen. Pharmacol. 34, 2225-2234. 3. Cheng, K. C., Gelboin, H. V., Song, B. J., Park, S. S., and Friedman, F. K. (1985) Reu. Drug Metab. Drug Interact. 5, 159-192.
AND
MYERS
4. Song, B. J., Fujino, T., Park, S. S., Friedman, F. K., and Gelboin, H. V. (1984) J. Biol. Chem. 259, 1394-1397. 5. Helm, K. A., Park, S. S., Gelboin, H. V., and Kupfer, D. (1989) Arch. Biochem. Biophys. 269, 664-667. 6. Heitanen, E., Malaveille, C., Friedman, F. F., Park, S. S., Bereziat, J.-C., Brun, G., Bartsch, H., and Gelboin, H. V. (1986) Cancer Res. 46, 524-531. 1. Myers, M. J., Liu, G., Robinson, R. C., Miller, H., Gelboin, H. V., and Friedman, F. K. (1990) Biochem. Biophys. Res. Commun. 169, 171-176. 8. Selander, H. G., Jerina, D. M., and Daly, J. W. (1974) Arch. Biochem. Biophys. 164, 241-246. 9. Guenthner, T. M., Negishi, M., and Nebert, D. W. (1979) Anal. Biochem. 96, 201-207. 10. Cheng, K.-C., Park, S. S., Krutzsch, H. C., Grantham, P. H., Gelboin, H. V., and Friedman, F. K. (1986) Biochemistry 25,2397-2402. 11. Battula, N., Sagara, J., and Gelboin, H. V. (1987) Proc. Natl. Acad. Sci. USA 84,4073-4077. 12. Aoyama, T., Gonzalez, F. J., and Gelboin, H. V. (1989) Mol. Carcinogen. 2, 192-198. 13. Aoyama, T., Gonzalez, F. J., and Gelboin, H. V. (1989) Mol. Carcinogen. 1, 253-259. 14. Aoyama, T., Yamano, S., Waxman, D. J., Lapenson, D. P., Meyer, U. R., Fischer, V., Tyndale, R., Inaba, T., Kalow, W., Gelboin, H. V., and Gonzalez, F. J. (1989) J. Biol. Chem. 26, 10,388-10,395. 15. Stanker, L. H., Vanderlaan, M., and Juarez-Salinas, H. (1985) J. Immunol. Methods 76, 157-169. 16. Nebert, D. W., and Gelboin, H. V. (1968) J. Biol. Chem. 243,62426249. 17. De Waziers, I., Cugnenc, P. H., Yang, C. S., Leroux, J. P., and Beaune, P. H. (1990) J. Pharmacol. Exp. Z’her. 253, 387-394. 18. Friedman, F. K., Miller, H., Park, S. S., Graham, S. A., Gelboin, H. V., and Carubelli, R. (1989) Biochem. Pharmacol. 38,3075-3081. 19. Guengerich, F. P., Dannan, G. A., Wright, S. T., Martin, M. V., and Kaminski, L. S. (1982) Biochemistry 21,6019-6031. 20. Robinson, R. C., Nagata, K., Gelboin, H. V., Rifkind, J., Gonzalez, F. d., and Friedman, F. K. (1990) Arch. Biochem. Biophys. 277,4246. 21. Furuya, H., Shinji, T., Hatano, M., and Fujii-Kuriyama, Y. (1989) Riochem. Biophys. Res. Commun. 160, 661-676.
OF BIOCHEMISTRY
Vol. 284, No. 2, February
AND
BIOPHYSICS
1, pp. 400-406, 1991
Role of Cytochrome P450 IA2 in Acetanilide Hydroxylation as Determined with cDNA Expression and Monoclonal Antibodies Gao Liu, Harry Laboratory
V. Gelboin,
of Molecular
and Michael
Carcinogenesis, National
4-
J. Myers’ Cancer Institute,
Bethesda, Maryland
20892
Received June 26, 1990, and in revised form October 4, 1990
INTRODUCTION The role of P450 IA2 in the hydroxylation of acetanilide was examined using an inhibitory monoclonal antibody (MAb) 1-7-1 and vaccinia cDNA expression producing murine P450 IA1 (mIAl), murine P450 IA2 (mIA2), or human P450 IA2 (hIA2). Acetanilide hydroxylase (AcOH) activity was measured using an HPLC method with more than 500-fold greater sensitivity than previously described procedures. This method, which does not require the use of radioactive acetanilide, was achieved by optimizing both the gradient system and the amount of enzyme needed to achieve detection by uv light. MAb 1-7-1 inhibits up to 80% of the AcOH activity in both rat liver microsomes and cDNA expressed mouse and human P450 IA2. MAb 1-7-1, which recognizes both P450 IA1 and P450 IA2, completely inhibits the aryl hydrocarbon hydroxylase (AHH) activity of cDNA expressed in IAl. The inhibition of only 80% of the AHH activity present in MC liver microsomes by MAb 1-7-1 suggests that additional P450 forms are contributing to the overall AHH activity present in methylcholanthrene (MC)-liver microsomes as MAb 1-7-1 almost completely inhibits the AHH activity of expressed mIA1. Maximal inhibition of IA2 by 1-7-1 results in an 80% decrease in acetanilide hydroxylase activity in both liver microsomes and expressed mouse and human IA2. The capacity of MAb 1-7-1 to produce identical levels of inhibition of acetanilide hydroxylase activity in rat MC microsomes (80%) and in expressed mouse (81%) and human P450 IA2 (80%) strongly suggests that P450 IA2 is the major and perhaps the only enzyme responsible for the metabolism of acetanilide. These results demonstrate the complementary utility of monoclonal antibodies and cDNA expression for defining the contribution of specific P450 enzymes to the metabolism of a given substrate. This complementary approach allows for a more precise determination of the inhibitory capacity of MAb with respect to the metabolic capacity of the target P450. o 1991 Academic
Press,
Inc.
Cytochrome P450 enzymes are responsible for the metabolism and activation of a vast array of endogenous and exogenous chemicals. The capacity to metabolize this large number of substrates is due, in part, to a large multiplicity of P450 enzymes, many of which possess overlapping substrate specificities. Monoclonal antibodies (MAbs)2 have proven to be useful tools toward delineating the role of epitope-specific P450 enzymes in the metabolism and activation of specific substrates. One such MAb, 1-7-1, recognizes and binds to both P450 IA1 and P450 IA2 (1). The overlapping P450 binding specificity of 1-71 is likely due to a common epitope resulting from the high degree of sequence homology of these two enzymes. MAb 1-7-1 has been used in RIA’s, immunopurification procedures, and enzyme inhibition assays (2-4) to characterize the substrate specificity and role of this class of P450 enzymes. However, due to the recognition of both P45Os by MAb l-7-1, it has not always been possible to determine the relative contribution of each of these two P45Os in the metabolism of selected substrates in tissue sources such as liver microsomes (5, 6). As a means to resolve the relative contributions of IA2 toward the metabolism/activation of selected substrates, we are developing site-specific monoclonal antibodies, generated using synthetic peptide antigens from regions unique to a given P450 (7). We have used the metabolism of acetanilide as an indicator of IA2 activity. While the hydroxylation of acetanilide is an activity associated with ’ To whom correspondence should be addressed at: U.S.F.D.A., Division of Veterinary Medical Research, BARC-EAST, Bldg. 328A, Beltsville, MD 20705. ’ Abbreviations used: AcOH, acetanilide hydroxylase; AHH, aryl hydrocarbon hydroxylase; ZOH-AA, 2-hydroxyacetanilide; 30H-AA, 3hydroxyacetanilide; IOH-AA, 4-hydroxyacetanilide; CO, control; MC, 3-methylcholanthrene; PB, phenobarbital; r, rat; m, mouse; h, human; MAb, monoclonal antibody;
400 All
Copyright 0 1991 rights of reproduction
0003.9861/91 $3.00 by Academic Press, Inc. in any form reserved.
ACETANILIDE
HYDROXYLASE
P450 IA2 (8), it is not certain that this is the sole enzyme capable of this activity. Therefore, the role of P450 IA2 in acetanilide hydroxylation was determined in two ways. First, the activity of cDNA-expressed P450 IA2 was determined for acetanilide hydroxylation and second the inhibition of acetanilide hydroxylation by MAb 1-7-1 was determined for the cDNA-expressed IA2 and microsomes. This complimentary approach, using both cDNA expression and monoclonal antibodies, permits a more precise assessment of the inhibitory capacity of the MAb with respect to its target P450, which in turn aids in the elucidation of the role of that form toward the total metabolism of the substrate in question. The present study describes how the combined use of the vaccinia virus cDNA expression system and an inhibitory MAb (MAb 1-7-1) permits a more accurate assessment of the contribution of P450 IA2 toward the metabolism of acetanilide in tissues. In the course of this study, we developed a quantitative HPLC assay for measuring the hydroxylation products of acetanilide using a nonradioactive substrate. Previously published HPLC-based detection methods depended upon the use of radioactive acetanilide (8,9). Our methodology is more sensitive (500-fold); thus, the method we describe may be a more useful assay than previously described methods (8, 9). This report describes the use of both monoclonal antibodies and cDNA-expressed P45Os to ascertain the role of a P450 enzyme in the metabolism of a given substrate in a crude tissue. MATERIALS
AND
METHODS
Preparation of microsomes and uaccinia virus-expressed P450. Liver microsomes were prepared from control, 3-methylcholanthrene-treated or phenobarbital-treated male Sprague-Dawley rats and NZB or Balb/ c mice, as previously described (10). Briefly, after 3 days of injection with 3-methylcholanthrene or 7 days with phenobarbital, the animals were sacrificed and the livers perfused with ice-cold 0.15 M KCl. The microsomal fraction was obtained by differential centrifugation of the S-9 fraction and was stored in 25% glycerol at -70°C. Enzymatically active mouse cytochrome P450 IAl, mouse IA2, and human IA2 cDNAs (kind gifts of Dr. Frank Gonzalez) were expressed in HepG2 cells using the vaccinia virus cDNA expression system as previously detailed (ll14). Twenty-four hours after viral infection of the HepG2 cells, the intact cells were collected and stored as a cell pellet at -70°C until used. Prior to use, the cell pellets were thawed on ice and briefly sonicated after 1 ml of distilled H,O was added to the vial. Protein content was determined using Pierce’s BCA methodology (Pierce). Monoclonal antibody preparation. Cell lines producing monoclonal antibodies were generously supplied by Drs. Sang Shin Park (MAb l7-l) and Sandra Smith-Gill (HyHeL9). MAb l-7-1 has been shown to bind to P450 IA1 and IA2 (1). The capacity of this MAb to inhibit the AHH activity of P450 IA1 has been extensively characterized with respect to rodent liver microsomes whereas its inhibitory capacity toward IA2 has not been completely characterized. HyHeL9 is a MAb which recognizes hen egg lysozyme, does not bind to P45Os, and is used as a negative MAb control (1). Both MAbs are of the IgGl subclass. The monoclonal antibodies were purified from ascites by hydroxylapatite chromatography (15) with the protein content determined as above. Acetanilide hydroxylase (ACOH) and aromatic hydrocarbon hydroxylase (AHH) assays. The metabolism of acetanilide was determined by its
ACTIVITY
OF CYTOCHROME
IA2
401
incubation with liver microsomes or expressed P45Os in the presence of: Tris buffer (50 mM, pH 7.5), MgCls (0.30 mM), NADPH (0.6 pmol), and BSA (1 mg) in a volume of I ml. The reaction was initiated by the addition of 25 ~1 of a 20 mM acetanilide solution (in acetone) unless otherwise specified. This amount represents a saturating concentration of acetanilide. In antibody-mediated enzyme-inhibition experiments, the reaction mixture was preincubated with either purified monoclonal antibody l-7-1 or HyHel9 (1) (Ab control) at room temp for 5 min. The reaction was terminated by the addition of 2.5 ml of ice-cold ethyl acetate (HPLC grade) and 0.1 ng of 3-hydroxyacetanilide (in 10 ~1 of acetone). The samples were vigorously mixed and centrifuged briefly. The ethyl acetate extract was removed, and a second 2.0-ml ethyl acetate extract was pooled with the first. The organic phase was then evaporated under N2 and the residue redissolved in 200 ~1 of methanol (HPLC grade). Twenty microliters of each sample was subsequently used for HPLC analysis. 3.Hydroxyacetanilide was added to each reaction tube as an internal standard to estimate the recovery of 4-hydroxyacetanilide. The 3-hydroxyacetanilide was reported to account for 0.1% or less of the total concentration of metabolites (8,9). In our study, we did not detect 30H-acetanilide. The recovery of 3-hydroxyacetanilide was typically 94 f 4%. AHH activity was determined according to the method of Nebert and Gelboin (16). Analysis of the metabolites was performed using HPLC analysis. reverse phase HPLC. A Spectra Physics Model SP8100 liquid chromatograph with an Valco injector, an autosampler, and a DuPont ODS column (4.6 X 25 cm) was used. A gradient system was used to separate the metabolites from material originating with the enzyme source. The solvents used were water:methanol:acetonitrile (HPLC grade), with a starting ratio of 78:17:5. Acetonitrile was kept at a constant ratio (5%) throughout the entire procedure. The solvent ratio of water and methanol changed by 0.5%/min for 3 min for a solvent ratio of 76.5:18.5:5, then changed again to a ratio of 55:4@5 in the next 2 min. This ratio was held constant for 5 min, after which the solvent conditions returned to the starting ratio of 78:17:5 in the next 2 min. The metabolites were detected using a Spectra-Physics Model 100 detector at a wavelength of 254 nm. The retention times for the 4-hydroxyacetanilide, 3-hydroxyacetanilide, 2-hydroxyacetanilide and the parent substrate acetanilide were 5.3, 7.5, 10.2, and 13.5 min, respectively.
RESULTS
Previously published HPLC-based methodologies for assessing acetanilide hydroxylase activity have relied on the use of radioactive acetanilide, with uv detection being achieved by the addition of excess amounts of nonradioactive acetanilide and its hydroxylated metabolites upon termination of the reaction (8, 9). However, the HPLC-solvent gradients used in these studies do not adequately resolve the 40H-acetanilide product from a contaminant originating from the microsomes. Therefore, we developed a new gradient system that permitted resolution of the three hydroxyacetanilide products and their separation from this contaminant. Thus, we were able to use nonradioactive acetanilide as the starting substrate and determine the kinetics of the reaction with respect to time, substrate concentration, and protein (microsome) concentration. With the use of saturating amounts of substrate (as suggested in Refs. (8, 9)) and 100 pg of rat MC microsomes (Fig. 1) the production of 4-hydroxyacetanilide (40H-AA) displays linear kinetics with respect to time. Similarly, the kinetics of acetanilide hydroxylase activity is linear from 10 to 200 pg with MC microsomes and saturating amounts of substrate (Fig. 2a). When the
402
LIU,
GELBOIN.
AND
MYERS
in close agreement with those obtained by Guenthner
et
al. (9).
:
Microsomes obtained from either control or MC- or PB-treated rats were preincubated with monoclonal antibody 1-7-1 followed by assessment of acetanilide hydroxylase activity. MAb 1-7-1 maximally inhibited the AcOH activity of rat MC microsomes at a concentration of 100 pg MAb to 100 pg of microsomes, producing 83% inhibition of AcOH activity (Fig. 4). Both control and PB microsomes exhibited maximal inhibition at a concentration of 200 pg of MAb 1-7-1 to 100 pg of microsomes, resulting in 67 and 73% inhibition of the AcOH activity
/
c
300 : 0
0,.:li, 10
20
30
(
,
40
50
1 60
70
TIME (min)
7rA-
FIG. 1. Formation of 4-hydroxyacetanilide as a function of time. MCinduced rat liver microsomes (100 Kg) were incubated with a saturating amount of acetanilide (0.5 mM) for 5 to 60 min. Assay conditions and product detection are as detailed under Materials and Methods. Each point represents the mean I~I SE of duplicate determinations of a representative assay. Similar results were obtained upon repetition with different microsomal preparations.
concentration of microsomes is further increased, acetanilide hydroxylase activity no longer displays absolute linearity (Fig. 2b) but rather is curvilinear. The lack of linearity of AcOH with high levels of microsomes is in agreement with Guenthner et al. (9). It should also be noted that production of the 2-hydroxyacetanilide metabolite is detected only under conditions in which the concentration of microsomal protein is at or near the nonlinear phase of the reaction (Fig. 2b). In addition, the concentration of SOH-AA does not increase with increasing amounts of microsomes as does the 40H-AA metabolites. The acetanilide hydroxylase activity of control microsomes and phenobarbital microsomes were three- to fourfold less than that observed with rat MC microsomes (Table I). The AcOH activity of a human liver microsomal preparation demonstrated 40HAA levels comparable to those seen with uninduced rat microsomes (Table I). In contrast to the rat, uninduced liver microsomes from either an AHH inducible strain (Balb/c) or a noninducible strain (NZB) had levels of 40H-AA comparible to those seen in rat MC-induced microsomes. MC induction had no effect on AcOH activity in NZB mice, whereas Balb/c microsomes had a threefold induction in activity (Table I). Lastly, under assay conditions in which acetanilide hydroxylase activity was linear with respect to both time and microsomal content, the addition of increasing amounts of acetanilide without a concomitant increase in enzyme activity demonstrates enzyme saturation with substrate (Fig. 3). The apparent Km and V,,, values are 0.303 mM and 3.38 nmol/mg protein/min, respectively (data not shown). These values are
0
50
100
150
200
l v
4 OH-AA 2 OH-AA
250
60 I
10 1
0 td 0
400
800
1200
MC Mic Cone
1600
2000
2400
(ug)
FIG. 2. Formation of 4-hydroxyacetanilide as a function of the concentration of MC rat liver microsomes. Increasing amounts of microsomes were incubated with a saturating amount of acetanilide (0.5 mM). Assay conditions and product detection are as detailed under Materials and Methods and the legend to Fig. 1. The production of 4-hydroxyacetanilide was determined using microsomes at concentrations ranging from 10 to 200 Kg (a) or from 10 to 2000 pg (b). The production of the 2-hydroxyacetanilide metabolite was consistently detected only when the concentration of microsomes was above 200 fig/reaction. Each point represents the mean k SE of duplicate determinations of a representative assay. Similar results were obtained upon repetition with different microsomal preparations.
ACETANILIDE TABLE
HYDROXYLASE
ACTIVITY
OF CYTOCHROME
IA2
I
Acetanilide Hydroxylase Activity of Induced and Uninduced Rat and Mouse Liver Microsomes and Human Liver Microsomes” Species/strain
In uiuo
Mouse/Balb/c
MC None MC None MC PB None None
Mouse/NZB Rat/SD
Human
nmol/OH-AA/20 98.9 28.5 29.4 29.8 33.4 12.3 12.0 16.5
min *
t 6.2
f 1.1 t ?I zk + + k
0.7 4.8 3.2 0.6 0.7 1.6 0
’ ACOH activity was determined as described under Materials and Methods. The tissue samples were obtained as detailed under Materials and Methods. Acetanilide was used at a final concentration of 0.5 mM. * Results are from a representative experiment. All experiments were performed at least twice, with duplicate determinations for each condition. Similar results were obtained with different preparations of microsomes.
(Fig. 4). MAb Z-66-3, which inhibits the activities of P45Os IIBl and IIB2, had no effect on acetanilide hydroxylase activity in PB microsomes (data not shown). MAb 1-7-1 binds to both P450 IA1 and P450 IA2. Expressed mIA1, however, does not exhibit AcOH activity (Table II). Thus, the inhibitory effect of MAb 1-7-1 on microsomes suggests that P450 IA2 is the major enzyme responsible for the metabolism of acetanilide. However, since MAb 1-7-1 in-
c 1 0, a
50
100
150
MAD 1-7-l
250
30”
350
(ug)
FIG. 4. Inhibition of acetanilide hydroxylase activity in rat liver microsomes using monoclonal antibody 1-i-l. Control (CO), phenobarbital (PB), or MC rat liver microsomes (100 pg) were incubated with a saturating amount of acetanilide (0.5 mM) for 20 min with increasing amounts of either MAb 1-7-1 or control MAb (HyHeL9). Maximal ACOH activities for CO, PB, or MC microsomes were similar to those detailed in Table I. Assay conditions and product detection are as detailed under Materials and Methods and Fig. 1. The results are expressed as percentages of the control MAb response. Each point represents the mean k SE of duplicate determinations of a representative assay. Similar results were obtained upon repetition with different microsomal preparations.
hibited acetanilide hydroxylase activity maximally at 80%, this suggests that either 1-7-1 was not capable of completely inhibiting AcOH activity or that additional P450 form(s) contribute to the metabolism of acetanilide in microsomes. These possibilities were examined by determining the capacity of MAb 1-7-1 to inhibit the AcOH
30
TABLE Acetanilide
25
200 Cone
Hydroxylase
Activity
II of Vaccinia-Virus
Expressed
Mouse and Human Cytochrome P450 IA1 and IAZ”
F q20
nmol 40H-AA/reaction:
& e
15
E c
10
1
3
5
(mg cell lysate/reaction) 5 C 30
C2
04 Substrate
Cone
06
08
(mM)
FIG. 3. Substrate saturability of the acetanilide hydroxylase activity of 3-methylcholanthrene (MC) rat liver microsomes. Microsomes (100 ug) were incubated with increasing amounts of acetanilide at 37°C for 20 min. The extent of activity was determined by measuring the amount of the main metabolite, 4-hydroxyacetanilide, that was present in each sample via HPLC coupled with uv detection. Each point represents the mean + SE of duplicate determinations of a representative assay. Similar results were obtained upon repetition with different microsomal preparations.
mIAl* mIA2 hIA
0.0 -c 0.51 k 0.10
0.0 1.18 f 0.03 1.16 i- 0.05
0.01 + 0.00 -c -c
a ACOH activity was determined as described under Materials and Methods. The cDNA expressed P45Os were obtained as detailed under Materials and Methods. The acetanilide concentration used was 0.5 mM. No ACOH activity was observed when cell lysates from uninfected Hep G2 cells were examined. b Results are from a representative experiment. All experiments were performed at least twice, with duplicate determinations for each condition. Similar results were obtained with different preparations of expressed P45Os; h, human; m, mouse. ’ Not determined.
404
LIU,
GELBOIN,
activity of vaccinia-virus expressed P450 IA2 (mouse and human) and its closely related form IAl. AcOH activity as previously shown, was exhibited by murine and human P450 IA2, whereas murine P450 IA1 showed negligible AcOH activity (Table II). In contrast, AHH activity is a function only of P450 IA1 (59.0 pmol/mg protein/min). Furthermore, the AHH activity of mIA1 was almost completely inhibited (>97%; 1.4 pmol/mg protein/min) by MAb 1-7-1 while the AHH activity of rat MC liver microsomes (2927.1 pmol/mg/min) was maximally inhibited by only 85% (449.0 pmol/mg/min). MAb 1-7-1 produced an 81% decrease in the AcOH activity of expressed murine P450 IA2 at a concentration of 16 pg MAb/mg cell lysate (Fig. 5), which represents the maximum extent to which this activity can be inhibited by 1-7-1. MAb 1-7-1 inhibited the AcOH activity of uninduced Balb/c and NZB liver microsomes and MC-induced NZB microsomes by 90% at a concentration of 50 pg MAb/lOO /*g microsomes. MAb 1-7-1 inhibits the activity of MC-induced Balb/c microsomes by 90% at a concentration of 300 pg MAb/ 100 pg microsomes. Statistical analysis (Student’s t test) demonstrated no significant differences between the maximum inhibition of the expressed mIA2 and any of the mouse liver microsomes (data not shown). MAb 1-71 resulted in an 80% inhibition of the AcOH activity of expressed human IA2 (Fig. 6), but in contrast to mouse
0 MC-E&lb/c V MC-NZB V CONT-Balb/c 0 CONT-NZB . mlA2
0
50
100
150 MAb l-7-1
200
250
AND
MYERS
0 L-L.-, 0
300
600 MAb l-7-;
900
'200
kc,:,
(us)
FIG. 6. Inhibition of acetanilide hydroxylase activity in human liver microsomes are expressed human IA2. Conditions are as outlined in the legend to Fig. 5.
IA2, produced maximal inhibition at a concentration of 400 pug MAb/mg cell lysate, a 24-fold increase in the amount of MAb needed as compared to the amount needed to produce maximal inhibition of the AcOH activity of expressed murine IA2. MAb 1-7-1 maximally inhibits the AcOH activity in human microsomes by 63% at a concentration of 900 pg MAb/l50 pg microsomes. Attempts to achieve higher inhibition were not possible as the negative MAb control had a nonspecific inhibitory effect on AcOH activity above a concentration of 400 pugMAb/lOO pg microsomes. DISCUSSION
300
3%
(ug)
FIG. 5. Inhibition of acetanilide hydroxylase activity of expressed murine IA2, control (CO) Balb/c, control NZB, MC Balb/c, or MCNZB liver microsomes using monoclonal antibody l-7-1. HepG2 cell lysate (1 mg) or rat liver microsomes (100 pg) were incubated with a saturating amount of acetanilide (0.5 mM) for 20 min with increasing amounts of either MAb 1-7-l or control MAb (HyHeL9). The HepG2 cells were infected with vaccinia virus which contained the cDNA for either human IA2 (hP3) or murine IA2. Maximal AcOH activities for either the expressed P45O’s or MC microsomes were similar to those detailed in Table I. Assay conditions and product detection are as detailed under Materials and Methods and the legend to Fig. 1. The results are expressed as percentages of the control MAb response. Each point represents the mean t SE of duplicate determinations of a representative assay. Similar results were obtained upon repetition with different sample preparations.
The results presented in this report demonstrate the utility of the combined use of monoclonal antibodies and cDNA-expressed cytochrome P45Os. The cDNA-expressed P45Os permit a precise definition of the substrate and product specificity of a given P450 form. The inhibitory MAb measures the minimum contribution of the P45O(s) to which it binds. Thus, the cDNA-expressed P450 yields substrate specificity information but does not portray specific P450 contributions toward substrate metabolism in a crude tissue, whereas the MAb measures the minimum contribution of a given P450 form (or forms) to the metabolism of a substrate. This complementary approach was demonstrated in this report with respect to the metabolism of acetanilide and benzo(a)pyrene. The expressed P45Os as reported previously demonstrate that benzo(a)pyrene is metabolized by P450 IA1 and not IA2, whereas acetanilide is metabolized by IA2 but not IA1 (Table II). MAb 1-7-1 inhibits both the AHH activity and the AcOH activity of MC-induced microsomes by 80%. These results taken independently suggest that while the metabolism of these two substrates is mutually exclusive with respect to P45Os
ACETANILIDE
HYDROXYLASE
IA1 and IA2, the residual 20% of AHH and AcOH activities may be other P450 enzymes which are capable of metabolizing these two substrates. A different picture emerges however if both methodologies are used together. MAb l-7-1 almost completely inhibits the metabolism of benzo(a)pyrene by expressed P450 IAl. Thus when MAb 1-7-1 is maximally inhibiting the AHH activity of P450 IAl, any residual activity almost certainly is the result of other P45Os. In contrast, MAb l-7-1 inhibited the AcOH activity of tissue-culture-expressed mouse and human IA2 by 80%; absolute inhibition was never obtained. This same degree of inhibition was also obtained when rat MC-induced liver microsomes or mouse liver microsomes (induced or uninduced) were examined. Thus, when MAb 1-7-1 is maximally inhibiting AcOH activity, the residual AcOH activity is possible due to P450 IA2, which is not inhibitable by MAb 1-7-1 rather than by additional P450 enzymes. One possible explanation may be that the portion of AcOH activity of cDNAexpressed IA2 not inhibited by 1-7-1 could also be due to enzyme which is inserted into the microsomal membrane such that the 1-7-l binding site is not accessible. Another aspect of the inhibition of AcOH activity of expressed mouse and human IA2 by MAb 1-7-1 is that a 24-fold higher concentration of antibody is needed to inhibit the expressed human IA2 compared to the amount needed to obtain a similar level of inhibition with the murine IA2. The need for more antibody is not due to differences in the amount of enzyme present since both mouse and human IA2 preparations contained equal amounts of enzyme activity (Table II) as well as similar spectral activity and equi-intense staining upon Western blot analysis using polyclonal antisera (data not shown). Human liver microsomes also required increased amounts of MAb 1-7-1 as compared to the rodent to achieve maximal inhibition of AcOH. Thus, the increased amount of 1-7-l needed to induce maximal inhibition of AcOH activity in human IA2 may be due to differences in the binding affinity of 1-7-l for human IA2 compared to rodent. IA2. This suggests that either the epitope recognized by 1-7-1 is not completely accessible in human P450 IA2 or, that the complete 1-7-1 binding epitope is not present in human IA2. This report also demonstrated that MAb 1-7-1 interacts with human IA2. Previous attempts failed to detect IA2 in human liver microsomes in Western blot analysis and standard alkaline phospatase detection methods (Friedman, unpublished results). That IA2 is indeed present in human liver microsomes was recently demonstrated using either antisera generated against synthetically prepared peptide sequences from IA2 (7) or polyclonal anti-rat P450 IAl/IA2 antisera (17). cDNA-expressed human IA2 also did not yield a positive signal with Western blot analysis using MAb 1-7-1 (data not shown). However, cDNA-expressed mouse IA2 also did not produce a positive signal after Western blot analysis using 1-7-1 whereas IA1 and
ACTIVITY
OF CYTOCHROME
IA2
405
IA2 in mouse MC liver microsomes are detectable with 1-7-1. The inability to immunodetect the cDNA-expressed forms of IA2 may be due to the relatively low levels of P450 present in the cell lysate or the lower affinity of l7-l for IA2 relative to IA1 (18). Both forms of expressed IA2 yield a positive ELISA signal (data not shown) which, as expected, is lower than that obtained for an equal amount of tissue-culture-expressed IA1 due to the lower affinity of 1-7-1 for IA2 relative to that for IAl. The inhibition kinetics of MAb 1-7-1 toward the AcOH activity of PB and CO microsomes show that more MAb 1-7-1 is needed to achieve maximum inhibition of the AcOH activity in these tissues than in MC-induced microsomes which have more P450 IA2 than either PB or CO microsomes. This suggests that either 1-7-1 is binding to another P450 which has AcOH activity or that it more likely is binding to both active and inactive proteins. Guengerich et al. (19) and Robinson et al. (20) have shown that for many P450 forms, the total amount of P450 protein is greater than the amount of spectrally active protein. The HPLC-based methodology used in this report for determining acetanilide hydroxylase activity has several advantages over previously published methodologies. One advantage is in the use of nonradioactive acetanilide as the substrate. The capacity to use a nonradioactive substrate was due to the use of a new HPLC gradient system. This system permitted resolution of the 40H-AA product from a substance of microsomal origin which comigrated with the 40H-AA product when previously published HPLC gradient systems were used. Another is in the amount of tissue sample needed to detect AcOH activity; as little as 10 pg of MC-induced liver microsomes resulted in detectable activity (Fig. 2a). This by itself represents a 50-fold more sensitive assay system than any previously published. Both of the previous methods used the entire sample for HPLC analysis (8,9). However, as we routinely examined only l/10 of the total reaction sample on the HPLC, this actually represents a 500-fold increase in sensitivity. Therefore, if all of the sample was to be examined, the lower level of sensitivity of the assay with respect to the amount of tissue sample needed drops to approximately 1 pg. Thus this methodology should be of value in determining the AcOH activity in tissues which have a lower P450 content than the liver or for tissues which are difficult to obtain in large quantities. The second main metabolite, 2-hydroxyacetanilide, only appeared when the concentration of liver microsomes was at or near the nonlinear portion of the curve, and in contrast to the 4-hydroxy metabolite, this metabolite did not increase with increasing amounts of microsomes (Fig. 2b). Two possible explanations are that SOH-AA is further metabolized by other enzymes or is not a product of IA2mediated metabolism. 20H-AA has been shown to be further metabolized to 2-acetamidohydroxyquinone through an NADPH-dependent pathway.
406
LIU,
GELBOIN,
The mechanism by which MAb 1-7-1 inhibits AcOH activity is unknown. In this regard, the results of Furuya et al. are intriguing (21). They demonstrated that almost any single amino acid alteration in the primary sequence of rat IA2 resulted in significant loss of enzyme activity. Their results suggest one possible mechanism by which MAb 1-7-1 is affecting IA2, that of conformation alteration of the enzyme by 1-7-1. In summary, the major P450 responsible for the metabolism of acetanilide is P450 IA2. MAb l-7-1, which recognizes both IA1 and IA2, inhibits up to 80% of the AcOH activity in both liver microsomes and tissue-culture-expressed IA2. In addition, MAb 1-7-1 completely inhibits the AHH activity of expressed IAl. As MAb l7-l maximally inhibits AHH activity of MC-induced liver microsomes by BO-85% this suggests that additional P450 forms are contributing to the overall AHH activity present in liver microsomes. We have also described a new HPLCbased methodology for detection of ACOH activity which is at least 500-fold more sensitive than previously published procedures. ACKNOWLEDGMENTS We thank Drs. Sang Shin Park and Sandra Smith-Gill for providing the monoclonal antibodies and Dr. Frank Gonzalez for the vaccinia virus containing the P450 cDNAs. We would also like to thank Dr. Fred Friedman for critically reviewing this manuscript. The assistance of Mr. Richard Robinson in the development of the HPLC-gradient system is deeply appreciated.
REFERENCES 1. Park, S. S., Fujinio, T., West, D. Guengerich, F. P., and Gelboin, H. V. (1982) Cancer Res. 42, 1798-1808. 2. Friedman, F. K., Park, S. S., and Gelboin, H. V. (1985) Biochen. Pharmacol. 34, 2225-2234. 3. Cheng, K. C., Gelboin, H. V., Song, B. J., Park, S. S., and Friedman, F. K. (1985) Reu. Drug Metab. Drug Interact. 5, 159-192.
AND
MYERS
4. Song, B. J., Fujino, T., Park, S. S., Friedman, F. K., and Gelboin, H. V. (1984) J. Biol. Chem. 259, 1394-1397. 5. Helm, K. A., Park, S. S., Gelboin, H. V., and Kupfer, D. (1989) Arch. Biochem. Biophys. 269, 664-667. 6. Heitanen, E., Malaveille, C., Friedman, F. F., Park, S. S., Bereziat, J.-C., Brun, G., Bartsch, H., and Gelboin, H. V. (1986) Cancer Res. 46, 524-531. 1. Myers, M. J., Liu, G., Robinson, R. C., Miller, H., Gelboin, H. V., and Friedman, F. K. (1990) Biochem. Biophys. Res. Commun. 169, 171-176. 8. Selander, H. G., Jerina, D. M., and Daly, J. W. (1974) Arch. Biochem. Biophys. 164, 241-246. 9. Guenthner, T. M., Negishi, M., and Nebert, D. W. (1979) Anal. Biochem. 96, 201-207. 10. Cheng, K.-C., Park, S. S., Krutzsch, H. C., Grantham, P. H., Gelboin, H. V., and Friedman, F. K. (1986) Biochemistry 25,2397-2402. 11. Battula, N., Sagara, J., and Gelboin, H. V. (1987) Proc. Natl. Acad. Sci. USA 84,4073-4077. 12. Aoyama, T., Gonzalez, F. J., and Gelboin, H. V. (1989) Mol. Carcinogen. 2, 192-198. 13. Aoyama, T., Gonzalez, F. J., and Gelboin, H. V. (1989) Mol. Carcinogen. 1, 253-259. 14. Aoyama, T., Yamano, S., Waxman, D. J., Lapenson, D. P., Meyer, U. R., Fischer, V., Tyndale, R., Inaba, T., Kalow, W., Gelboin, H. V., and Gonzalez, F. J. (1989) J. Biol. Chem. 26, 10,388-10,395. 15. Stanker, L. H., Vanderlaan, M., and Juarez-Salinas, H. (1985) J. Immunol. Methods 76, 157-169. 16. Nebert, D. W., and Gelboin, H. V. (1968) J. Biol. Chem. 243,62426249. 17. De Waziers, I., Cugnenc, P. H., Yang, C. S., Leroux, J. P., and Beaune, P. H. (1990) J. Pharmacol. Exp. Z’her. 253, 387-394. 18. Friedman, F. K., Miller, H., Park, S. S., Graham, S. A., Gelboin, H. V., and Carubelli, R. (1989) Biochem. Pharmacol. 38,3075-3081. 19. Guengerich, F. P., Dannan, G. A., Wright, S. T., Martin, M. V., and Kaminski, L. S. (1982) Biochemistry 21,6019-6031. 20. Robinson, R. C., Nagata, K., Gelboin, H. V., Rifkind, J., Gonzalez, F. d., and Friedman, F. K. (1990) Arch. Biochem. Biophys. 277,4246. 21. Furuya, H., Shinji, T., Hatano, M., and Fujii-Kuriyama, Y. (1989) Riochem. Biophys. Res. Commun. 160, 661-676.
