Elimination of antipyrine and benzo[a]pyrene metabolism in cultured human lymphocytes A strong correlation was found in a carefully selected homogenous population (n = 57) between antipyrine plasma half-life and the percent induction of aryl hydrocarbon hydroxylase by 3-methylcholanthrene in mitogen-stimulated lymphocytes from the same individual. The correlation coefficient of r = -0.923 indicates that antipyrine and benzo[aJpyrene share one or several common determinants that are responsible for the observed interindividual variation in the oxidation rates of the two compounds. When a heterogenous population (n = 80) was studied. the above correlation was not found (r = -0.425).
Gottfried Kellermann, Ph.D., Mieke Luyten-Kellermann, Marjorie G. Horning, Ph.D., and Michele Stafford Madison, Wis., and Houston, Texas Department of Human Oncology, University of Wisconsin Medical School, and Institute
for Lipid Research, Baylor College of Medicine
The existence of large interindividual differences in drug metabolism has been widely recognized. Two types of such variability have been observed: (1) separately recognizable genetic phenotypes or genotypes and (2) a unimodal variability. With the second type of variability there is either complex genetic control or multifactorial control, with both genetic and environmental components. Recent studies applying the methods of quantitative genetics have revealed that the relative size of the genetic component is frequently substantial. 2, 27, 39,42 The situation is further complicated by the fact that the different Supported by Grant No. POl-CA-14520, awarded by the National Cancer Institute, Department of Health, Education, and Welfare, and Grant No. GM-16216 from the National Institutes of Health. Received for publication Dec. IS, 1975. Accepted for publication April 7, 1976. Reprint requests to: Dr. G. H. Kellermann, Wisconsin Clinical Hospital Cancer Center, 701 C University Hospital, 1300 University Ave., Madison, Wis. 53706.
72
reactions participating in the metabolism of some drugs are little understood. In addition, most drug-metabolizing enzymes are highly inducible either by chronic administration of the drug itself,4 of other drugs, and by exposure to a variety of chemicals in the environment,23 including cigarette smoke. 28 Other reports indicate an effect of age, sex,26, 41 and physical stress 38 on human drug metabolism. It is well known that the thyroid status markedly influences metabolic reactions carried out by the liver mixed function oxidases. 8, 40 An in vivo test alone which involves the administration of a single test drug may therefore prevent the accurate assessment of genetic factors influencing drug matabolism. Such studies provide only information on the ability of the individual to metabolize the drug at the time it is given. It is not surprising that all studies on plasma elimination rates reported in the literature show a more or less unimodal
Volume 20 Number 1
distribution, without indication of separate groups that might suggest a different responsible genetic type. Aryl hydrocarbon hydroxylase (AHH) , or benzpyrene hydroxylase, is a microsomal mixed-function oxidase that is involved in the metabolism of polycyclic aromatic hydrocarbons (PAH). We have established a relationship between the levels of inducibility of AHH and the rates of benzo[aJpyrene metabolism to water-soluble compounds. 21 In cultures of mitogen-stimulated lymphocytes from a population of normal subjects, we found three distinct groups, low, intermediate, and high, of AHH inducibility.20 Family studies indicated that these differences were under genetic control, and a model of multiple alleles at single locus was proposed to explain the variations. For reasons to be discussed later, it was not clear from these studies whether the observed genetic variation was a variation of the inducibility, or of the basal activities of the mixed-function oxidases. It was of interest to determine whether the individual genotypes recognized in vitro by the lymphocyte AHH technique would also show comparable differences in whole body kinetics in vivo toward drugs whose biotransformation follows a pathway similar to that of benzo[aJpyrene. It has been established that the first step in the metabolism of benzo[ aJpyrene involves the formation of an epoxide 10 ; the epoxide then rearranges nonenzymatically to the phenol or is converted by enzymatic reaction into the dihydrodiol or the glutathione conjugate. Recently the metabolism of antipyrine (AP) was studied in the rat, guinea pig, and human. 36 In addition to AP and 4-hydroxyantipyrine, a hydroxymethylantipyrine and a dihydrodiol were found in the unconjugated fraction isolated from urine. The epoxide was not detected in the urine of the 3 species; however, since the dihydrodiol and 4-hydroxyantipyrine were present as metabolites, the epoxide was undoubtedly an intermediate in their formation. These similarities in the metabolism of benzo[aJpyrene and AP prompted the use of AP for our in vivo experiments. It was assumed that plasma elimination rates
Antipyrine and benzo[a J pyrene
73
of AP could be equated with its metabolism in the liver. The recent demonstration by Huffman, Shoeman, and Azarnoff17 of a significant correlation (r = 0.89) between the plasma decay of AP and the appearance of a major metabolite, 4-hydroxyantipyrine, validates previous studies using the rate of plasma elimination of AP as a measure of drug metabolism in man. Subjects
Since our goal in this study was to determine whether there is a relation between the benzo[aJpyrene metabolism in the cultured lymphocytes of individuals and their plasma elimination rates of AP, we endeavored to select as nearly ideal subjects as possible. Factors reported to affect normal drug metabolism were taken into account in selecting the subjects. Forty-one males and 16 females participated (with 2 exceptions, none of them had participated in the previous drug study. 19 They ranged in age between 18 and 35 yr, were in good physical condition, without signs of physical or psychological stress, and were engaged in light work. The criteria for selection was established and the actual selection of subjects in the "selected" group was done before AP halflives and AHH values had been determined. A pilot study on factors influencing drug metabolism confirmed the existence of the sex difference first reported by Malley and associates. 26 The mean AP half-life was significantly shorter in females. In an attempt to correlate this phenomenon with the menstrual cycle, it was found that around the day of ovulation there was an increase in the metabolic rate of AP, reducing the half-life in some as much as 50%. Since it was not possible to establish whether this increased metabolism was restricted to the period of ovulation or whether there were other periods of the menstrual cycle during which the oxidation rate of a drug could be influenced by steroids or other endogenous factors, only those women having a fairly regular cycle were accepted, and tests were performed on them only between days 4 to 10 and 19 to 25 of the cycle. Since little is known of the effect of cigarette smoking on normal drug metabolism, heavy smokers
74
Kellermann et al.
(more than 20 cigarettes per day) were excluded. In addition to the group" selected subjects," a second group of "unselected subjects," 46 males and 34 females, was studied for AHH in their cultured lymphocytes and for AP plasma halflives. This group included 45 volunteers of the first group and in addition 35 volunteers who were chosen at random. The only requirements for their participation were good health and currently being on no medication. No age limitations were set and in female subjects the AP test was performed without regard to the stage of the menstrual cycle. Antipyrine was given as a single oral dose of 18 mg/kg in the morning. Subjects were fasting for 12 hr before taking the drug. Blood was drawn after 4,6.5, 9, and 11.5 hr. Plasma AP half-lives were measured as described in the next section. Two or three months later the AP half-lives were determined again in 11 individuals. For the lymphocyte culture, 20 ml of venous blood was drawn either before administration of antipyrine or several days after. Methods
Plasma AP was measured in quadruplicates by the spectrophotometric method of Brodie and associates 3 and in duplicates by selective ion detection using a gas chromatograph-mass spectrometer computer (Finnegan 1015-PDP 8/1) system in a chemical ionization mode. The intensity of the protonated molecular ion (MH+ = 189) was monitored, and a calibration curve using a standard solution of AP was used for quantification. AP was isolated quantitatively from plasma using the ammonium carbonate-ethyl acetate procedure. 15 It was found with the spectrophotometric method that good results and a high degree of reproducibility were obtained when the plasma samples were fresh and unfrozen, and were measured for AP within 48 hr by the same technician throughout the study. The methodological error ranged between 3% and 8%. Lymphocyte cultures were set up by diluting the heparinized blood (20 ml) with 50 ml of 0.85% saline. The sample was divided into two 50-ml screw-capped centrifuge tubes, and 10 ml of a Hypaque-Ficoll gradient solution (specific
Clinical Pharmacology and Therapeutics
gravity: 1.080) was layered under the diluted blood. After centrifugation at 400 g in a PR-J centrifuge (head IEC No. 253 or No. 269) for 30 min, the supernatant was discarded and the lymphocyte-rich bands were collected, combined, washed with 20 ml Hanks balanced salt solution, and then incubated for 10 to 15 min at 37° in a Tris-buffered ammonium chloride solution. After centrifugation at 300 g for 10 min, the cell pellet was suspended in a small amount of culture medium and divided into 6 culture tubes (3 controls and 3 tubes for treatment with 3-methy1cholanthrene [3-MC]) containing 4 ml of culture medium. 2i Cultures were incubated in a 5% CO 2 incubator at 37° for 72 hr, and for another 24 hr after receiving 1.6 fLM 3-MC. Cells were harvested by centrifugation, washed with culture medium, and then assayed for AHH. Whole cells were used for the assay, and no sonification or homogenizing was required. Phosphate buffer, 0.9 ml, was added to each tube (0.08 M KH 2P0 4 , 0.005 M MgCI 2 , 0.03 M nicotinamide, pH 7.55) containing between 1.5 and 3 x 106 cells. NADH 0.70 mg and NADPH 0.70 mg were added in a volume of 0.1 ml buffer making a final volume of 1.0 m!. The incubation was started by addition of 25 fLg of benzo[aJpyrene in 50 fLl of acetone. After 30 min the reaction was stopped with 4 ml of 25% acetone in hexane. The tubes were Vortexmixed for 1 min, and the phases were separated by centrifugation; the organic phase was then transferred to another tube containing 0.5 ml of I N NaOH. After additional mixing for I min followed by centrifugation, the fluorescence of the lower phase was determined in an Aminco-Bowman spectrophotofluorometer with excitation at 396 nm and emission at 522 nm. The extent of induction is expressed as the ratio of AHH activity after induction with 3-MC over the activity before induction. The most critical factor in the lymphocyte culture system is the fetal calf serum (FCS). Since FCS varies from lot to lot with respect to its inducing ability for AHH, 9 a lot is selected and used for the experiment only if with its use all three AHH groups can be detected with certainty. A FCS that is older than 3 to 5 mo cannot be used for the AHH test. Use after that
Volume 20 Number 1
Antipyrine and benzo[a] pyrene
75
UJ
0
:::l
"C
>
"C
E
....0 ~
Q)
.c
E :::l
Z
0
Fig. 1. Distribution of plasma antipyrine half-lives in 57 "selected subjects." 18.0 17.0 16.0
••
15.0
.
.L::
14.0 13.0
.... N
i-=" 12.0 I
Q.
Gottfried Kellermann, Ph.D., Mieke Luyten-Kellermann, Marjorie G. Horning, Ph.D., and Michele Stafford Madison, Wis., and Houston, Texas Department of Human Oncology, University of Wisconsin Medical School, and Institute
for Lipid Research, Baylor College of Medicine
The existence of large interindividual differences in drug metabolism has been widely recognized. Two types of such variability have been observed: (1) separately recognizable genetic phenotypes or genotypes and (2) a unimodal variability. With the second type of variability there is either complex genetic control or multifactorial control, with both genetic and environmental components. Recent studies applying the methods of quantitative genetics have revealed that the relative size of the genetic component is frequently substantial. 2, 27, 39,42 The situation is further complicated by the fact that the different Supported by Grant No. POl-CA-14520, awarded by the National Cancer Institute, Department of Health, Education, and Welfare, and Grant No. GM-16216 from the National Institutes of Health. Received for publication Dec. IS, 1975. Accepted for publication April 7, 1976. Reprint requests to: Dr. G. H. Kellermann, Wisconsin Clinical Hospital Cancer Center, 701 C University Hospital, 1300 University Ave., Madison, Wis. 53706.
72
reactions participating in the metabolism of some drugs are little understood. In addition, most drug-metabolizing enzymes are highly inducible either by chronic administration of the drug itself,4 of other drugs, and by exposure to a variety of chemicals in the environment,23 including cigarette smoke. 28 Other reports indicate an effect of age, sex,26, 41 and physical stress 38 on human drug metabolism. It is well known that the thyroid status markedly influences metabolic reactions carried out by the liver mixed function oxidases. 8, 40 An in vivo test alone which involves the administration of a single test drug may therefore prevent the accurate assessment of genetic factors influencing drug matabolism. Such studies provide only information on the ability of the individual to metabolize the drug at the time it is given. It is not surprising that all studies on plasma elimination rates reported in the literature show a more or less unimodal
Volume 20 Number 1
distribution, without indication of separate groups that might suggest a different responsible genetic type. Aryl hydrocarbon hydroxylase (AHH) , or benzpyrene hydroxylase, is a microsomal mixed-function oxidase that is involved in the metabolism of polycyclic aromatic hydrocarbons (PAH). We have established a relationship between the levels of inducibility of AHH and the rates of benzo[aJpyrene metabolism to water-soluble compounds. 21 In cultures of mitogen-stimulated lymphocytes from a population of normal subjects, we found three distinct groups, low, intermediate, and high, of AHH inducibility.20 Family studies indicated that these differences were under genetic control, and a model of multiple alleles at single locus was proposed to explain the variations. For reasons to be discussed later, it was not clear from these studies whether the observed genetic variation was a variation of the inducibility, or of the basal activities of the mixed-function oxidases. It was of interest to determine whether the individual genotypes recognized in vitro by the lymphocyte AHH technique would also show comparable differences in whole body kinetics in vivo toward drugs whose biotransformation follows a pathway similar to that of benzo[aJpyrene. It has been established that the first step in the metabolism of benzo[ aJpyrene involves the formation of an epoxide 10 ; the epoxide then rearranges nonenzymatically to the phenol or is converted by enzymatic reaction into the dihydrodiol or the glutathione conjugate. Recently the metabolism of antipyrine (AP) was studied in the rat, guinea pig, and human. 36 In addition to AP and 4-hydroxyantipyrine, a hydroxymethylantipyrine and a dihydrodiol were found in the unconjugated fraction isolated from urine. The epoxide was not detected in the urine of the 3 species; however, since the dihydrodiol and 4-hydroxyantipyrine were present as metabolites, the epoxide was undoubtedly an intermediate in their formation. These similarities in the metabolism of benzo[aJpyrene and AP prompted the use of AP for our in vivo experiments. It was assumed that plasma elimination rates
Antipyrine and benzo[a J pyrene
73
of AP could be equated with its metabolism in the liver. The recent demonstration by Huffman, Shoeman, and Azarnoff17 of a significant correlation (r = 0.89) between the plasma decay of AP and the appearance of a major metabolite, 4-hydroxyantipyrine, validates previous studies using the rate of plasma elimination of AP as a measure of drug metabolism in man. Subjects
Since our goal in this study was to determine whether there is a relation between the benzo[aJpyrene metabolism in the cultured lymphocytes of individuals and their plasma elimination rates of AP, we endeavored to select as nearly ideal subjects as possible. Factors reported to affect normal drug metabolism were taken into account in selecting the subjects. Forty-one males and 16 females participated (with 2 exceptions, none of them had participated in the previous drug study. 19 They ranged in age between 18 and 35 yr, were in good physical condition, without signs of physical or psychological stress, and were engaged in light work. The criteria for selection was established and the actual selection of subjects in the "selected" group was done before AP halflives and AHH values had been determined. A pilot study on factors influencing drug metabolism confirmed the existence of the sex difference first reported by Malley and associates. 26 The mean AP half-life was significantly shorter in females. In an attempt to correlate this phenomenon with the menstrual cycle, it was found that around the day of ovulation there was an increase in the metabolic rate of AP, reducing the half-life in some as much as 50%. Since it was not possible to establish whether this increased metabolism was restricted to the period of ovulation or whether there were other periods of the menstrual cycle during which the oxidation rate of a drug could be influenced by steroids or other endogenous factors, only those women having a fairly regular cycle were accepted, and tests were performed on them only between days 4 to 10 and 19 to 25 of the cycle. Since little is known of the effect of cigarette smoking on normal drug metabolism, heavy smokers
74
Kellermann et al.
(more than 20 cigarettes per day) were excluded. In addition to the group" selected subjects," a second group of "unselected subjects," 46 males and 34 females, was studied for AHH in their cultured lymphocytes and for AP plasma halflives. This group included 45 volunteers of the first group and in addition 35 volunteers who were chosen at random. The only requirements for their participation were good health and currently being on no medication. No age limitations were set and in female subjects the AP test was performed without regard to the stage of the menstrual cycle. Antipyrine was given as a single oral dose of 18 mg/kg in the morning. Subjects were fasting for 12 hr before taking the drug. Blood was drawn after 4,6.5, 9, and 11.5 hr. Plasma AP half-lives were measured as described in the next section. Two or three months later the AP half-lives were determined again in 11 individuals. For the lymphocyte culture, 20 ml of venous blood was drawn either before administration of antipyrine or several days after. Methods
Plasma AP was measured in quadruplicates by the spectrophotometric method of Brodie and associates 3 and in duplicates by selective ion detection using a gas chromatograph-mass spectrometer computer (Finnegan 1015-PDP 8/1) system in a chemical ionization mode. The intensity of the protonated molecular ion (MH+ = 189) was monitored, and a calibration curve using a standard solution of AP was used for quantification. AP was isolated quantitatively from plasma using the ammonium carbonate-ethyl acetate procedure. 15 It was found with the spectrophotometric method that good results and a high degree of reproducibility were obtained when the plasma samples were fresh and unfrozen, and were measured for AP within 48 hr by the same technician throughout the study. The methodological error ranged between 3% and 8%. Lymphocyte cultures were set up by diluting the heparinized blood (20 ml) with 50 ml of 0.85% saline. The sample was divided into two 50-ml screw-capped centrifuge tubes, and 10 ml of a Hypaque-Ficoll gradient solution (specific
Clinical Pharmacology and Therapeutics
gravity: 1.080) was layered under the diluted blood. After centrifugation at 400 g in a PR-J centrifuge (head IEC No. 253 or No. 269) for 30 min, the supernatant was discarded and the lymphocyte-rich bands were collected, combined, washed with 20 ml Hanks balanced salt solution, and then incubated for 10 to 15 min at 37° in a Tris-buffered ammonium chloride solution. After centrifugation at 300 g for 10 min, the cell pellet was suspended in a small amount of culture medium and divided into 6 culture tubes (3 controls and 3 tubes for treatment with 3-methy1cholanthrene [3-MC]) containing 4 ml of culture medium. 2i Cultures were incubated in a 5% CO 2 incubator at 37° for 72 hr, and for another 24 hr after receiving 1.6 fLM 3-MC. Cells were harvested by centrifugation, washed with culture medium, and then assayed for AHH. Whole cells were used for the assay, and no sonification or homogenizing was required. Phosphate buffer, 0.9 ml, was added to each tube (0.08 M KH 2P0 4 , 0.005 M MgCI 2 , 0.03 M nicotinamide, pH 7.55) containing between 1.5 and 3 x 106 cells. NADH 0.70 mg and NADPH 0.70 mg were added in a volume of 0.1 ml buffer making a final volume of 1.0 m!. The incubation was started by addition of 25 fLg of benzo[aJpyrene in 50 fLl of acetone. After 30 min the reaction was stopped with 4 ml of 25% acetone in hexane. The tubes were Vortexmixed for 1 min, and the phases were separated by centrifugation; the organic phase was then transferred to another tube containing 0.5 ml of I N NaOH. After additional mixing for I min followed by centrifugation, the fluorescence of the lower phase was determined in an Aminco-Bowman spectrophotofluorometer with excitation at 396 nm and emission at 522 nm. The extent of induction is expressed as the ratio of AHH activity after induction with 3-MC over the activity before induction. The most critical factor in the lymphocyte culture system is the fetal calf serum (FCS). Since FCS varies from lot to lot with respect to its inducing ability for AHH, 9 a lot is selected and used for the experiment only if with its use all three AHH groups can be detected with certainty. A FCS that is older than 3 to 5 mo cannot be used for the AHH test. Use after that
Volume 20 Number 1
Antipyrine and benzo[a] pyrene
75
UJ
0
:::l
"C
>
"C
E
....0 ~
Q)
.c
E :::l
Z
0
Fig. 1. Distribution of plasma antipyrine half-lives in 57 "selected subjects." 18.0 17.0 16.0
••
15.0
.
.L::
14.0 13.0
.... N
i-=" 12.0 I
Q.
