| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Carcinogenesis |
Eppley Institute for Research in Cancer [S. C. C., X. W., G. X., L. Z., S. S. M.], and Departments of Pharmaceutical Sciences [J. L. V., S. S. M.] and Biochemistry and Molecular Biology [S. S. M.], University of Nebraska Medical Center, Omaha, Nebraska 68198-6805, and National Cancer Institute, Bethesda, Maryland 20892 [F. G., H. V. G.]
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-Hydroxy-NAms (the longest-lived intermediates in NAm activation) have half-lives of only 1–10 s; therefore, most NAms are believed to be activated in the tissues where they induce tumors (4
, 7)
. The esophagus is the second most common site (after the liver) for tumor induction by NAms in rats (6)
, probably because the esophagus contains P450 isoforms that activate esophagus-specific NAms (4)
, although the ready diffusion of some NAms into the esophagus probably also helps determine the organ specificity (8)
.
Our previous studies found that RLM produced mainly 4-hydroxy-MNAN, HCHO, and PENT from MNAN (9)
. Formation of HCHO and PENT should be accompanied by the formation of, respectively, a pentylating and a methylating agent, both of which could alkylate DNA and initiate cancer (Fig. 1)
. Farrelly and Stewart (10)
reported that RLM demethylated and depentylated MNAN with Km values of 2.6 and 1.2 mM, respectively. We incubated RLM with MAbs that inhibit individual P450s and then with 6 mM MNAN, and determined the MNAN metabolites. The results indicated that: (a) 4-hydroxylation of MNAN was mainly due to CYP2C11; (b) demethylation (HCHO production) was due in about equal parts to CYP2E1, CYP2B1/2B2, and CYP2C11; and (c) 30% of depentylation (PENT production) was due to CYP2C11 (11)
. REM and human esophageal microsomes demethylated 6 mM MNAN 18–20 times more efficiently than they demethylated 5 mM dimethyl-Nam, whereas human liver microsomes showed similar demethylating activities for both MNAN and dimethyl-Nam (12)
. Esophageal and liver microsomes of both rats and humans also depentylated MNAN. Human esophageal microsomes were one-third (for MNAN demethylation) and one-tenth (for MNAN depentylation) as active as REM. These results helped explain why unsymmetrical dialkyl-NAms induce esophageal cancer in rats and suggested that such NAms could initiate this cancer in humans.
|
-hydroxylation by REM of N'-nitrosonornicotine, a tobacco-specific NAm that induces esophageal tumors in rats (13)
. These low Km values suggested that we increase the sensitivity of our method by using radiolabeled MNAN. The use of [3H-pentyl]MNAN enabled us to examine the depentylation of 5 µM MNAN using microsomes with only 50–100 µg protein, an important factor in view of the limited supply of REM.
In the present study, we examined the dealkylation of [3H-pentyl]MNAN to give [3H]PENT rather than that of [3H-methyl]MNAN to give [3H]HCHO because methylation, but not pentylation, of DNA guanine in MNAN has been detected in the rat esophagus (14
, 15)
and DNA methylation is associated with MNAN depentylation (Fig. 1)
. The depentylation of MNAN was also linked more closely than its demethylation with its bacterial mutagenicity in the presence of RLM (16)
. Pentyldiazohydroxide produced by MNAN demethylation may not pentylate DNA extensively because it forms a pentyldiazonium ion that could lose a proton to yield 1-pentene, which should not alkylate DNA. Similarly, ethylene is produced during diethyl-NAm metabolism by rabbit nasal microsomes (17)
. Hence, depentylation is probably more relevant than demethylation of MNAN to its carcinogenicity. Accordingly, we examined the depentylation of [3H-pentyl]MNAN by REM, RLM, and certain human and rat P450s (18
, 19)
.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Synthesis of [3H-4,5-pentyl]MNAN and [3H-2,3-pentyl]MNAN.
Aqueous methylamine (30 ml of 40%, 380 mmol) was added over a period of 1 h dropwise with stirring to 10 g (65 mmol) of 5-bromo-1-pentene in a flask fitted with a dry-ice condenser containing a salt-ice mixture. The reaction was continued for 2–3 h at 0°C and for 18 h at room temperature. The mixture was adjusted to pH 2 with HCl and extracted repeatedly with CH2Cl2 until TLC of the aqueous phase indicated that nearly all of the methyldipentenylamine had been removed. The TLC used fluorescent silica gel plates (60-F254, Curtis-Matheson Scientific, Houston, TX), which were developed with CH2Cl2:methanol 85:15 saturated with NH4OH and showed Rf 0.64 for metnylpentenylamine and Rf 0.92 for methyldipentenylamine (detected under UV light). The aqueous phase was made basic with NaOH and extracted with 5x 80 ml of CH2Cl2. The extract was dried over Na2SO4 and evaporated to give 2.98 g (46%) of N-methyl-N-4,5-pentenylamine as a colorless oil that polymerized on heating or prolonged storage. 1H-NMR in CDCl3: 1.60 (q, CH2CH2N, 2H), 2.09 (q, CHCH2, 2H), 2.44 (s, CH3N, 3H), 2.60 (t, CH2N, 2H), 5.00 (m, CH2
CH, 2H), and 5.81 ppm (m, CHCH2, 1H). Mass spectrum: m/z 99.1 (19%, molecular ion, C6H13N), 84.08 (22%, C5H10N), 70.06 (59%, C4H8N), and 57.05 (100%, C3H7N).
[3H-2,3-pentyl]MNAN was synthesized similarly, except that 2 M methylamine (90 ml, 180 mmol) in methanol was reacted with 5 g (32 mmol) of 1-bromo-2-pentene. After the amine mixture was adjusted to pH 2, it was evaporated to remove the methanol, and 20 ml of water were added. N-Methyl-N-2,3-pentenylamine (0.45 g, 14% yield) was obtained as a colorless oil. 1H-NMR in CDCl3: 1.00 (t, CH3CH2, 3H), 2.06 (m, CH3CH2, 2H), 2.46 (s, CH3N, 3H), 3.23 (d, CH2N, 2H), and 5.57(m, CHCH, 2H).
Samples (50–100 mg) of methyl-4,5-pentenylamine in ethyl acetate were hydrogenated with tritium and then nitrosated at SRI International (Menlo Park, CA) to give crude [4,5-3H]MNAN, which was stored in toluene at -15°C. GC-TEA (22) of the undiluted [3H]MNAN showed a prominent peak with the retention time of MNAN as well as other peaks in some batches. Samples of methyl-2,3-pentenylamine in ethanol were hydrogenated with tritium at New England Nuclear Life Science Products (Boston, MA) to give [2,3-3H]methylpentylamine, which was stored in ethanol at -15°C. As required, we nitrosated 60-mCi samples of this amine by slow addition of nitrite to a solution of the amine and HCl (20) and CH2Cl2 extraction of the product to give [2,3-3H]MNAN.
Purification of [3H]MNAN.
To measure radioactivity, samples were mixed with 4 ml of Ecolume cocktail (ICN Inc., Costa Mesa, CA) and assayed in a liquid scintillation counter (Beckman Instrument Co., Fullerton, CA). We synthesized [4,5-3H]MNAN three times and [2,3-3H]MNAN once (batches 1–4, respectively, numbered in chronological order). Batch 1 showed 19 Ci/mol starting amine. TLC by "system 1" (silica gel 60-F254 plates developed with hexane:ether l:l; Rf of MNAN, 0.6) of samples of these four batches indicated that 77, 11, 11 and 71%, respectively, of the radioactivity was due to [3H]MNAN. Bands were scraped off the plates, mixed with 4 ml of Ecolume, and assayed for tritium. Because of their low purity, 60-mCi samples of batches 2 and 3 were mixed with 100 µg of unlabeled MNAN/100 µl CH2Cl2 and applied as strips to alumina 60-F254 TLC plates (200 x 200 x 0.25 mm), which were developed with hexane:ether:acetic acid 50:46:4 ("system 2"). The [3H]MNAN band (Rf, 0.6) was indicated by UV detection of cold MNAN applied as spots at each side of the plate and was eluted with 20 ml of CH2Cl2. On TLC of eluate samples by system 1, the [3H]MNAN band contained 80–90% of the eluted counts.
The [3H]MNAN contained small amounts of [3H]PENT, which seemed to be generated during storage and was mostly removed by semicarbazide treatment. On the day of the metabolic experiment, a CH2Cl2 solution of 150–200 µCi [3H]MNAN (unpurified batches 1 or 4, or TLC-purified batches 2 or 3) and (for batches 1 and 4) up to 25 µg of unlabeled MNAN were added to 1 ml of water, and the CH2Cl2 was evaporated at room temperature with a N2 stream over the surface of the water. The aqueous solution was mixed with 4 ml of 10% semicarbazide·HCl in 1.1 M Na acetate in water (pH 4–5) and heated at 65°C for 10 min. The [3H]MNAN was extracted with 3x 2 ml of CH2Cl2. The extract was dried over Na2SO4, concentrated to 0.7 ml, and subjected to TLC on eight 60-F254 alumina plates (65 x 50 x 0.25 mm, Curtis Matheson Scientific) developed with hexane:ether 4:6 ("system 3"). Spots of unlabeled MNAN at each side of the plates were used to indicate the [3H]MNAN bands (Rf, 0.6), which were scraped off, combined, eluted with 10 ml of CH2Cl2, and transferred (see above) to 1.0 ml of water. [On TLC by system 3, PENT and PENT semicarbazone traveled with Rf 0.60 and 0.75, respectively, close to the Rf of MNAN. PENT and its semicarbazone were revealed as blue spots when plates were sprayed with 5% phosphomolybdic acid in ethanol and heated for 10 min at 100°C.] We determined the concentration of [3H]MNAN by GC-TEA (22 , 23) and its radioactivity in the 1-ml aqueous solution. This contained 50–150 µCi [3H]MNAN.
Isolation of Microsomes.
Microsomes were prepared as described previously (11)
from the livers of 6- to 8-week-old adult male Sprague Dawley rats (Sassco Inc., Omaha, NE) that were untreated or induced with PB, 3MC, or isoniazid (9
, 11
, 12)
. The livers were homogenized in 3 ml/g tissue of 100 mM potassium phosphate buffer (pH 7.4) containing 0.1 mM DTT and 0.14 mM phenylmethylsulfonyl fluoride. RLM were obtained by differential centrifugation in the same buffer, suspended in 100 mM potassium phosphate buffer (pH 7.4) containing 20% glycerol, analyzed for protein by the Lowry method, and stored in 1.5-ml Eppendorf tubes at -70°C.
Microsomes were prepared from rat esophagi each weighing 50–80 mg (see "Materials") by our previous method (12) involving homogenization of the thawed esophagi in a Potter-Elvejhem homogenizer or by the following modification of Murphy’s method (13) : esophagi (five or six at a time) were each cut into four or five pieces while frozen, crushed with a Bessman tissue pulverizer precooled with liquid N2, transferred to an ice-cold glass Tenbroech homogenizer, and gently homogenized with six passes of the pestle each way in 6–7 ml of 50 mM Na PPi buffer (pH 7.4), containing 1 mM EDTA, 1 mM DTT, and 5 mM phenylmethylsulfonyl fluoride. The combined homogenate from three such procedures was differentially centrifuged. The microsome fraction was analyzed for protein and stored as for liver microsomes. Three such isolations, from 17, 16, and 50 esophagi (the last combined from three batches), yielded REM with 7.2, 2.6, and 7.4 mg of protein.
Use of Individual P450s.
These were stored in Eppendorf tubes at -70°C. On the day of use, 1 ml of a suspension containing P450s from the National Cancer Institute was mixed with 1 ml of 100 mM potassium phosphate buffer (pH 7.4) containing 20% glycerol, ultrasonicated twice for 5 s, and centrifuged at 50,000 rpm for 30 min. The pellet was resuspended in 1.0 ml of the same buffer with a Potter-Elvejhem homogenizer. P450s from Gentest Corporation were supplied in 100 mM potassium phosphate buffer (pH 7.4) and were gently shaken by hand before use.
Metabolic Experiments.
In Method A (used unless mentioned otherwise), each experiment included 12–16 tubes, each with 500 µl of medium containing 100 mM potassium phosphate buffer (pH 8.0), 10 mM MgCl2, 5 mM semicarbazide·HCl (9
, 11)
, [3H]MNAN (3–10 x 106 cpm), unlabeled MNAN (amount calculated after allowing for MNAN in the [3H]MNAN sample), microsomes with 50 or 100 µg protein, and (added last to start the reaction) NADPH-generating mixture containing 2 mM NADP, 10 mM glucose-6-phosphate, and 2 units of glucose-6-phosphate dehydrogenase (final pH 7.4). The experiments used [3H-4,5-pentyl]MNAN except for those where [3H-2,3-pentyl]MNAN is specified. Tubes 1 and 2 were blanks with 20 µM MNAN. For Km measurements, the remaining tubes contained microsomes or a P450 and 5–7 concentrations of MNAN, each run in duplicate. The tubes were incubated for 20 min at 37°C. In Method B, incubations were performed as in Method A but for 60 min and with 15 mM semicarbazide·HCl.
The incubation mixtures were worked up as described previously (9 , 11) . In brief, reactions were stopped with Ba(OH)2 and ZnSO4. After centrifugation, the supernatants were reacted for 1 h with 2,4-dinitrophenylhydrazine in HCl. The hydrazones were extracted into iso-octane and back-extracted into CH3CN, which was evaporated. The residues were dissolved in 200 µl of CH3CN, and 100 µl samples were subjected to HPLC on a C-18 column eluted with ethanol-water (2 min at 50% ethanol, 5 min gradient to 70% ethanol, 17 min at 70% ethanol, and 6 min at 50% ethanol). Retention times were 7 min for dinitrophenylhydrazine and 21 min for PENT dinitrophenylhydrazone. Eluates were collected as 1-min fractions from 17 to 26 min. Each fraction was evaporated to dryness with N2 at 65°C, mixed with 4 ml of Ecolume, and counted. Because counting efficiency did not vary, all of the results are expressed as cpm. Conversion of unlabeled PENT to its dinitrophenylhydrazone was 23 ± 0.6% (mean ± SE for 28 tests) when 20 nmol/tube was subjected to the conditions of the metabolic experiment with various types of microsomes and analyzed by HPLC with detection at 355 nm (12) . Results were not corrected for these losses. After subtracting the mean counts for the same fractions of the HPLC of blank tubes 1 and 2, the sum of the cpm eluted at about 21–23 min (depending on peak position) was used to calculate PENT yields from MNAN. Yields were graphed with a Prism Program (Graph Pad, San Diego, CA) for nonlinear regression curves. Km and Vmax values were obtained from the best-fitting straight lines for Lineweaver-Burke plots using the Excel 4.0 program (Microsoft, Redmond, WA).
Coumarin 7-Hydroxylation.
7-Hydroxycoumarin formation from coumarin was determined fluorimetrically (24
, 25)
. Microsomes (with 50 or 100 µg protein/tube) were incubated for 30 min at 37°C with coumarin in 0.5 ml of a mixture containing 100 mM potassium phosphate buffer (pH 7); 10 mM MgCl2; REM, RLM, or CYP2A6; and NADPH-generating system (see "Metabolic Experiments" section). Reactions were stopped by adding 0.5 ml of 0.31 M trichloracetic acid and 4 ml of 1.6 M glycine-NaOH buffer (pH 10.3). The concentration of 7-hydroxycoumarin (
ex, 371 nm; em, 454 nm) was determined with an RF 5000U spectrofluorimeter (Shimadzu Corp., Kyoto, Japan).
Western Blots.
Microsomes were solubilized in solutions of cholic acid and 2-mercaptoethenol with heating for 2 min at 100°C (26)
. Samples with 35–50 µg of protein/well were subjected to gel electrophoresis. The gels were developed with MAbs to individual P450s (MAbs 1-98-1 and 1-91-3 for CYP2E1, 2-66-3 and 4-7-1 for CYP2B1/2, 1-68-11 for CYP2C11/12, and 1-7-1 for CYP1A1/2; Ref. 21
) and were visualized with peroxidase-labeled sheep antimouse antibody (27)
. Positive results were indicated by the staining of bands at 50–57 kDa depending on the P450.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). Inclusion of 5 mM semicarbazide in the metabolic incubations increased PENT yield by mean values of 118% for REM and 61% for CYP2A6 and was essential when RLM were used but had no effect with rat CYP2E1 (Table 2)
. PENT production was linear with time for 30 min when PB-induced RLM were incubated with 100 µM [3H]MNAN. Therefore, in earlier experiments of this study, reaction mixtures were incubated for 20 min in the presence of 5 mM semicarbazide (Method A). Using this method, PENT yield from 50 and 200 µM MNAN increased as the amount of REM was raised from 20 to 100 µg of protein/tube (Fig. 2A)
.
|
|
|
. Hence, experiments performed with Method A that showed PENT yields less than twice the background level were repeated using Method B. Method B was also used in all later studies with REM. The apparent Km for MNAN depentylation by REM was 64 µM with a Vmax of 20 pmol PENT/min/mg (Table 1
and Fig. 3
).
|
MNAN metabolism by uninduced and PB-, 3MC-, and isoniazid-induced RLM showed classic dose-response curves for PENT yield versus MNAN concentration, with apparent Km values of 170–610 µM (Table 1)
. When PB- and isoniazid-induced RLM were used, 100 µM MNAN was depentylated 7.4 and 1.6 times faster, respectively, than uninduced RLM, and the apparent Km values were about one-half of the Km for uninduced RLM. Although 3MC-induced and uninduced RLM depentylated 100 µM MNAN at similar rates, the apparent Km was 3.6 times lower for 3MC-induced than for uninduced RLM.
We examined the effect of inhibitors on the depentylation of 100 µM MNAN by REM and by uninduced RLM. A 9:1 CO:air mixture inhibited MNAN metabolism by REM by a mean of 98% (Table 3)
. Coumarin (0.4 mM) produced only a 19% inhibition of MNAN depentylation by REM when coumarin, REM, and MNAN were added at the same time, but produced a 55% inhibition with an apparent Ki of 120 µM when coumarin and REM were preincubated for 15 min before adding MNAN (Tables 3
and 4)
. Preincubation for 30 min produced no additional effect. All of the subsequent studies with coumarin used preincubation for 15 min. Coumarin inhibition of MNAN metabolism by REM reached 65% when coumarin concentration was raised to 0.6 mM (Fig. 4)
.
|
|
|
. The depentylation of 100 µM MNAN was significantly inhibited (by a mean of 41%) by the MAb to CYP 2C11/2C12 and was significantly enhanced (by a mean of 24%) by the MAb to CYP2E1.
MNAN Metabolism by Individual P450s: Rat and Human CYP2E1.
MNAN depentylation by rat CYP2E1, human CYP2E1 from the National Cancer Institute, and human CYP2E1 from Gentest showed apparent Km values of 210, 150, and 170 µM, respectively (Table 5)
. MAb 1-91-3 to rat CYP2E1 (50 µg) was preincubated with 10 pmol of rat CYP2E1 as in the MAb tests in Table 3
, and the mixture was then incubated by Method A with 100 µM MNAN. The MAb inhibited MNAN production by 85, 89, and 92% in three tubes, demonstrating its strong activity. When 10, 20, and 40 pmol of human CYP2E1 from Gentest was incubated with 100 µM MNAN, PENT yield increased linearly with the amount of P450 (PENT yield in pmol/min: 0.41 and 0.71 for 10, 1.28 and 1.91 for 20, and 3.98 and 4.53 for 40 pmol of CYP2E1).
|
; Table 5
). Replacing the standard 100 mM phosphate buffer by 50 mM Tris buffer did not affect MNAN depentylation by CYP2A6, despite a contrary claim for metabolism by this P450 in the 1994 Gentest catalogue. Coumarin produced a 96% inhibition of the depentylation of 100 µM MNAN by 30 pmol/tube CYP2A6 [PENT yield: 68.5 and 53.9 (for CYP2A6) and 1.8 and 2.4 (for CYP2A6 preincubated with 0.6 mM coumarin) pmol/nmol P450/min], with an apparent Ki of 7.5 µM (Table 4)
.
|
. Human CYP3A4 from the National Cancer Institute showed a nearly linear dose-response curve in experiments with 20 and 40 pmol of P450/tube, with no saturation even at 12 mM MNAN and a rate for 100 µM MNAN of 48 pmol of PENT/nmol CYP3A4/min (Table 5)
. CYP3A4 from Gentest gave similar results. This P450 might have given low results because it was not fully dispersed in the medium. However, CYP3A4 metabolism of 100 µM MNAN was inhibited by 86% when 0.5 mg/ml of the detergent 3-(3-cholamidopropyldimethylamino)-1-propane sulfonate ("CHAPS") was included and was unaffected by including 1 mg/ml of BSA in the medium.
Coumarin Metabolism and the Effect thereon of MNAN.
For coumarin 7-hydroxylation, Table 4
and Fig. 6
demonstrate a rapid metabolism of coumarin with an apparent Km of 50 µM for CYP2A6, a slower metabolism with a higher apparent Km for RLM, and low but still measurable activity for REM. MNAN inhibited coumarin 7-hydroxylation by RLM and CYP2A6 with apparent Ki values of 3000 and 320 µM, respectively but did not seem to inhibit the low activity of REM for this reaction (Table 4)
. Table 4
also shows whether each inhibition of MNAN and coumarin metabolism seemed to be competitive, uncompetitive, or noncompetitive (30
, 31)
.
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
We used [3H-4,5-pentyl]MNAN for most of the studies and [3H-2,3-pentyl]MNAN for the more recent experiments. The results should not depend on which MNAN isomer was used because MNAN activation does not involve a compound with labile hydrogen at C-2 and because PENT, which could enolize and exchange T for H at C-2 under alkaline conditions, was kept at neutral or acidic pH or was combined with semicarbazide or dinitrophenylhydrazine. The two isomers of [3H]MNAN seemed to give similar results but the [3H]2,3-pentyl isomer seemed to be more readily synthesized and more stable than the [3H]4,5-pentyl isomer.
Semicarbazide was included in all of the incubations with MNAN. It increased the yield of HCHO from dimethyl-NAm 2.5-fold in a 1979 study on mouse liver microsomes (33)
and has since been used in many similar investigations, e.g., those in references (9
, 11
, 12
, 29)
. Presumably, semicarbazide acts because it forms unstable semicarbazones of aldehydes that protect them from oxidation to carboxylic acids, a reaction catalyzed by rodent liver microsomes (34)
. After the microsomal incubation, the semicarbazone is converted to a more stable dinitrophenylhydrazone or other derivative for determination (33)
. Although semicarbazide competitively inhibited dimethyl-NAm demethylation by rat CYP2E1 in RLM (28)
, it did not affect MNAN depentylation by overexpressed rat CYP2E1 (Table 2)
, perhaps because semicarbazide inhibition of CYP2E1 activity was counterbalanced by an inhibition of PENT oxidation. The addition of 5 mM semicarbazide increased by >60% the depentylation of MNAN by REM, RLM, and CYP2A6 (Table 2)
. The use of higher levels of semicarbazide did not have an additional effect (see "Results, MNAN Metabolism by REM and RLM"). Therefore, semicarbazide should continue to be used in dealkylation studies not involving CYP2E1.
MNAN Metabolism by REM and RLM.
REM showed an apparent Km of 64 µM for MNAN depentylation, with 10% of the Km for uninduced RLM (Table 1)
. Although this high-affinity activity of REM showed a Vmax that was only 9% of that for RLM, the low Km for MNAN metabolism by REM supports the view that NAm carcinogenesis in the rat esophagus is due to tissue-specific activation of these NAms. One reason for the low Vmax for REM is presumably that MNAN is mainly metabolized by basal cells of the esophageal mucosa (14)
, but REM were prepared from the entire mucosa and part of the submucosa. For comparison, methylbenzyl-NAm (a more potent esophageal carcinogen than MNAN on a molar basis) is debenzylated by REM with a Km of <10 µM (35)
.
The finding that MNAN activation by REM was 98% inhibited by CO (Table 3)
demonstrates that the reaction involved P450s. Coumarin (0.4 mM) inhibited CYP2A5 in mouse liver microsomes (36)
but this P450 apparently does not occur in rat liver, although rat nasal mucosa contains a P450, probably CYP2A3, that metabolizes dimethyl-NAm and NNK and is inhibited by coumarin (37)
. The observation that coumarin inhibited REM metabolism of MNAN by up to 65% with a Ki of 120 µM (Fig. 4
; Tables 3
and 4
) suggests that a CYP2A5-like enzyme makes a major contribution to the esophageal metabolism of NAms that induce esophageal cancer. The weak activity of REM for the 7-hydroxylation of coumarin (Table 4)
confirms a similar observation by Murphy et al. (35)
. Findings that none of the test MAbs inhibited MNAN metabolism by REM (Table 3)
and that immunoblots did not reveal any P450s other than traces of CYP2B1/2B2 in REM (50 µg/tube, see "Results") indicate that MNAN depentylation in the esophagus did not involve P450s 1A1, 1A2, 2C11, 2E1, or (probably) 2B1 or 2B2. The lack of inhibition of REM activity by the MAb to human CYP2A6 (Table 3)
may have occurred because rat CYP2A5 or the analogous rat esophageal P450 is not inhibited by this MAb (we found no information on this point).
The observation that MNAN depentylation by uninduced RLM was enhanced 24% by MAb 1-91-3 to CYP2E1 (Table 3)
indicates that CYP2E1 catalyzed a pathway of MNAN metabolism other than depentylation. This other pathway is presumably demethylation, one-third of which was due to CYP2E1 at a MNAN level of 6 mM (11)
. The finding that 41% of the depentylation of 100 µM MNAN by RLM was inhibited by the MAb to the constitutive male P450, CYP2C11, indicates that about 41% of this metabolism was due to CYP2C11 (Table 3)
, similar to the figure of 30% found for the depentylation of 6 mM MNAN (11)
. It is not known which enzymes catalyze the remaining 50–60% of MNAN depentylation by RLM.
The effect of P450 inducers was examined for RLM but not REM because of the large number of induced rats that would be needed to prepare sufficient REM for such a study. The Km values for MNAN depentylation by PB-, 3MC-, and isoniazid-induced RLM were 28–54% of the Km for uninduced RLM, and Vmax for PB-induced RLM was 5.8 times that for uninduced RLM (Table 1)
. These results indicate that CYP2B1 or CYP2B2 (induced by PB), CYP1A1 or CYP1A2 (induced by 3MC), and CYP2E1 (induced by isoniazid; Ref. 11
) can all depentylate MNAN. The rates for the depentylation of 100 µM MNAN were 7.4 and 1.6 times faster for PB- and isoniazid-induced RLM, respectively, than for uninduced RLM (Table 1)
, similar to the corresponding relative rates of 5.9 and 1.7 for the depentylation of 6 mM MNAN (11)
.
MNAN Metabolism by Individual P450s.
The rat liver enzyme, CYP2E1, showed a Km of 210 µM for MNAN depentylation (Table 5)
. This relatively high Km suggests that large, but not small, doses of MNAN methylate rat liver DNA (14
, 15)
because of activation by CYP2E1. This Km value was higher than those of 15–40 µM for the dealkylation of dimethyl- and diethyl-NAm by CYP2E1 (38)
. Rat CYP2E1 also demethylates and debutylates methylbutyl-NAm (Km, 2–4 mM; Refs. 29
and 39
) and dealkylates dipropyl- but not dibutyl-NAm (40)
. The Km of 115–170 µM for MNAN depentylation by human CYP2E1 (Table 5)
was somewhat lower than that for rat CYP2E1, which suggests that CYP2E1 might play a role in activating unsymmetrical dialkyl-NAms in humans. Rat CYP2A1 did not depentylate MNAN (Table 5)
, although it is important for NNK activation by rat lung and nasal microsomes (41)
. The lack of a Km for human CYP3A4 and its weak activity for MNAN depentylation (Table 5)
suggest that this P450 is not important for MNAN activation, although its abundance in human liver (42)
could counterbalance these considerations.
The human liver and nasal P450, CYP2A6 (43
, 44)
, showed a very low Km of 17 µM for MNAN depentylation (Fig. 5
; Table 5
). Coumarin, a specific inhibitor of CYP2A isoforms (36
, 43)
, inhibited CYP2A6 metabolism of MNAN with an apparent Ki of 7.5 µM (Table 4)
. These findings suggest that MNAN depentylation by human esophageal microsomes, which showed a Km of 80–160 µM for this reaction (45)
, could be due to CYP2A6. CYP2A6 activated NNK to form a methylating mutagen with a Km of 120 µM (46)
and probably catalyzed NNK and diethyl-NAm metabolism by human liver microsomes (47
, 48)
. Coumarin and an antibody to CYP2A5 inhibited the dealkylation of dimethyl- and diethyl-NAm by mouse liver microsomes (47)
. Diethyl-NAm was mainly metabolized by CYP2A5 in mouse and hamster liver microsomes (47, 48, 49)
. CYP2A3 debenzylated MBZN with a Km of 3 µM (35)
. Rat nasal microsomes activated N'-nitrosonornicotine and methylbenzyl-NAm with Km values of 2–5 µM by reactions that were inhibited by coumarin (50)
. If MNAN is also activated by rat nasal CYP2A enzymes, this would probably explain why MNAN induces nasal as well as esophageal tumors in rats (3)
.
CYP2A5 is a rodent homologue of CYP2A6 and is also inhibited by coumarin (36)
. It occurs in mouse nasal mucosa and in mouse and hamster, but not rat, liver (44
, 51)
. CYP2A3 occurs in rat nasal mucosa (37
, 44)
. Our finding that coumarin strongly inhibited MNAN metabolism by REM (Fig. 4
; Table 4
) and our results for CYP2A6 metabolism of MNAN (Table 5)
suggest that a P450 of the 2A subfamily is responsible for most MNAN metabolism by REM. We found an apparent Km of 50 µM for coumarin 7-hydroxylation by CYP2A6 (Table 4)
, higher than the reported Km values for this P450 of 0.5–0.7 (24)
and 6 (43)
µM. This difference is probably due to the long (30 min) incubation time used in our tests. The low activity of REM for coumarin 7-hydroxylation (11% of that for RLM; see Table 4
) is consistent with the view that CYP2A3 rather than CYP2A5 occurs in REM because CYP2A3 shows low activity (10% of that for CYP2A6; Ref. 43
), whereas CYP2A5 shows high activity (36
, 51)
for coumarin metabolism.
Conclusions.
Our results demonstrate that REM and RLM can depentylate low concentrations of MNAN. REM, rat CYP2E1, human CYP2E1, and human CYP2A6 activated MNAN with Km values of 
210 µM (Table 5)
. Our inhibition and metabolism studies indicate that an enzyme resembling CYP2A3 catalyzes most MNAN depentylation by REM and confirm our finding (11)
that CYP2C11 contributes to MNAN depentylation by RLM. Identification of the major NAm-metabolizing P450 in the rat esophagus and extension of this study to microsomes obtained from various human tissues (we are currently completing the latter project; Ref. 45
) should help disclose how MNAN and certain other NAms induce esophageal cancer in rats and help indicate whether NAm activation could be involved in the etiology of human esophageal cancer.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 Supported by Grant RO1-CA-35628 and Core Grant P30-CA-36727 from the National Cancer Institute, Grant 97B-125 from the American Institute for Cancer Research, and Core Grant SIG-16 from the American Cancer Society. Some of this work was reported at two meetings of the American Association for Cancer Research (1
, 2) . 
2 To whom requests for reprints should be addressed, at Eppley Institute for Research in Cancer, University of Nebraska Medical Center, 600 South 42nd Street, Omaha, NE 68198-6805.
3 The abbreviations used are: GC-TEA, gas chromatography with detection by thermal energy analysis; HCHO, formaldehyde; HPLC, high-performance liquid chromatography; 3MC, 3-methylcholanthrene; MAb, monoclonal antibody; NAm, nitrosamine; MNAN, methyl-n-amyl-NAm; NMR, nuclear magnetic resonance spectrum; NNK, 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone; P450 or CYP, cytochrome P450; PB, phenobarbital; PENT, pentaldehyde; REM, rat esophageal microsomes; Rf, retardation factor; RLM, rat liver microsomes; and (in NMR descriptors) d, doublet; m, multiplet; q, quartet; and s, singlet.
Received 7/13/98. Accepted 10/29/98.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-hydroxylates N'-nitrosonornicotine. Carcinogenesis (Lond.), 15: 2709-2713, 1994.-Hydroxylation of N-nitrosobenzylmethylamine (NBZMA) by rat esophageal microsomes and cytochrome P450 2A3. Abstracts Division Chemical Toxicology, 216th Am. Chem. Soc. Natl. Meet, : 1 American Chemical Society Washington, DC 1998.
This article has been cited by other articles:
|
|
 |
|
  L. Cao, X. Xu, L. L. Cao, R.-H. Wang, X. Coumoul, S. S. Kim, and C.-X. Deng Absence of full-length Brca1 sensitizes mice to oxidative stress and carcinogen-induced tumorigenesis in the esophagus and forestomach Carcinogenesis, July 1, 2007; 28(7): 1401 - 1407. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. L. Wong, S. E. Murphy, M. Wang, and S. S. Hecht Comparative metabolism of N-nitrosopiperidine and N-nitrosopyrrolidine by rat liver and esophageal microsomes and cytochrome P450 2A3 Carcinogenesis, February 1, 2003; 24(2): 291 - 300. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. Godoy, R. M. Albano, E. G. Moraes, P. R. A. Pinho, R. A. Nunes, E. H. Saito, C. Higa, I. M. Filho, C. D. P. Kruel, C. C. Schirmer, et al. CYP2A6/2A7 and CYP2E1 expression in human oesophageal mucosa: regional and inter-individual variation in expression and relevance to nitrosamine metabolism Carcinogenesis, April 1, 2002; 23(4): 611 - 616. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. F. Ribeiro Pinto, E. Moraes, R. M. Albano, M. C. Silva, W. Godoy, T. Glisovic, and M. A. Lang Rat oesophageal cytochrome P450 (CYP) monooxygenase system: comparison to the liver and relevance in N-nitrosodiethylamine carcinogenesis Carcinogenesis, November 1, 2001; 22(11): 1877 - 1883. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. C. Chen, L. Zhou, X. Ding, and S. S. Mirvish Depentylation of the Rat Esophageal Carcinogen, Methyl-n-pentylnitrosamine, by Microsomes from Various Human and Rat Tissues and by Cytochrome P450 2A3 Drug Metab. Dispos., September 1, 2001; 29(9): 1221 - 1228. |
hydroxylation to give stable hydroxy-MNANs.
) and 200 (
) µM MNAN using Method A [incubation with 5 mM semicarbazide for 20 min (A)] or Method B [incubation with 15 mM semicarbazide for 60 min (B)]. Each point, the results for an individual tube. The tests in A and B were conducted at different times with different batches of REM. This may explain why A and B show similar results although Method B generally produced more PENT than Method A.
