Human Term Placental Lipoxygenase-Mediated N-Demethylation of Phenothiazines and Insecticides in the Presence of Linoleic Acid

Placenta (2000), 21, 646–653 doi:10.1053/plac.2000.0547, available online at http://www.idealibrary.com on

Human Term Placental Lipoxygenase-Mediated N-Demethylation of Phenothiazines and Insecticides in the Presence of Linoleic Acid C. G. Hover and A. P. Kulkarnia College of Public Health, University of South Florida, Tampa, Florida 33612, USA Paper accepted 24 February 2000

This study investigated the hypothesis that human term placental lipoxygenase (HTPLO) and soybean lipoxygenase (SLO) are capable of mediating N-demethylation of selected phenothiazines and insecticides in the presence of linoleic acid (LA). In addition to being LA dependent, the N-demethylation reaction mediated by HTPLO and SLO was limited by incubation time, pH of the medium, concentration of the enzyme and the substrate. Using Nash reagent to monitor formaldehyde production, the specific activity for LA-dependent N-demethylation of chlorpromazine, a model phenothiazine, was determined to be 1.70.3 nmoles/min/mg HTPLO. Besides chlorpromazine, N-demethylation of promazine, promethazine and trimeprazine was also observed. The insecticide, aminocarb, displayed a specific activity of 2.20.3 nmoles/min/mg HTPLO for N-demethylation. Other insecticides, namely chlordimeform, dicrotophos and zectran, were oxidized in a similar manner. As compared with HTPLO, the rates of N-demethylation of phenothiazines and insecticides mediated by SLO were higher. Classical inhibitors of lipoxygenase, as well as antioxidants and free radical scavengers, caused a dose-dependent reduction in the production of formaldehyde from chlorpromazine and aminocarb by HTPLO. These results clearly demonstrate the ability of polyunsaturated free fatty acids to support N-demethylation of xenobiotics via the lipoxygenase pathway.  2000 Harcourt Publishers Ltd Placenta (2000), 21, 646–653

INTRODUCTION N-demethylation is one of the major reactions by which many drugs, insecticides and other chemicals are metabolized. The removal of alkyl group(s), in general, increases the water solubility of the parent compound. Metabolic biotransformation of numerous xenobiotics by N-demethylation has been observed with several enzyme systems including cytochrome P450, prostaglandin synthase, peroxidases and, most recently, lipoxygenase (LO). The human term placenta from nonsmoker is unique in that cytochrome P450 (Juchau, 1995; Hakkola et al., 1998), flavin-containing mono-oxygenase and prostaglandin synthase (Kulkarni, 1996) are extremely low or undetectable. Phenothiazine derivatives are among the most commonly used and widely accepted antipsychotic (neuroleptic) drugs, having been used clinically for almost 50 years. In addition to the control of schizophrenia or other serious psychiatric illnesses, these drugs have been identified as having numerous other potential pharmacological applications including the use as antiemetics, antimicrobial and antitumour agents. In cona

To whom correspondence should be addrssed at: Florida Toxicology Research Center, Department of Environmental and Occupational Health, College of Public Health, MDC-56, University of South Florida, 13201 Bruce B. Downs Boulevard, Tampa, Florida 33612-3805, USA. Fax: +1 (813) 974-4986; E-mail: [email protected]

0143–4004/00/070646+08 $35.00/0

trast, reports on varying adverse effects, such as endocrine alterations, cardiac toxicity and reproductive toxicity associated with the phenothiazine exposure, have been reported in the literature (Potter and Hollister, 1998). Since phenothiazines also display antiemetic, antinauseatic and sedative properties, they are prescribed during pregnancy (Austin and Mitchell, 1998). Although these drugs appear to be relatively safe for use in pregnancy, conflicting reports on their teratogenic effects exist. Sobel (1960) suggested that chlorpromazine (CPZ) induces abortions, but causes no birth defects. Milkovich and van den Berg (1976) found no causal relationship between phenothiazine use during pregnancy and birth defects. However, Slone et al. (1977) reported a relationship between congenital malformations, perinatal mortality rate, birthweight, intelligence quotient score and antenatal exposure to phenothiazines. The results of more recent studies (Edlund and Craig, 1984; Altshuler et al., 1996) have concluded that in utero exposure to CPZ during 4–10 weeks of gestation leads to an increased incidence of congenital anomalies in infants. Mixed results have also been obtained from animal studies. For example, perphenazine was shown to cause cleft palate in mice and rats, but not in rabbits (Szabo and Brent, 1974). In another study, CPZ was also found to cause cleft palate in mice (Walker and Patterson, 1974). Currently, the biochemical mechanism(s) responsible for developmental toxicity of phenothiazines is unknown.  2000 Harcourt Publishers Ltd

Hover and Kulkarni: Lipoxygenase-Mediated N-Demethylation of Xenobiotics

Literature reports indicate that phenothiazines undergo extensive oxidative metabolism by cytochrome P450 (Valoti et al., 1998), methaemoglobin (Kelder et al., 1991a) and peroxidases (Kelder et al., 1991b; Yang and Kulkarni, 1997). Recently, the ability of soybean LO (SLO) to catalyze the oxidation of phenothiazines in the presence of H2O2 was investigated (Perez-Gilabert, Sanchez-Ferrer and Garcia-Carmona, 1994a, 1994b). The first step in both metabolism and pharmacological activity of phenothiazines is generally accepted to involve oxidation of the drug to a cation free radical (Levy et al., 1972). It is believed that the cation radical serves as an intermediate in the further oxidation of the phenothiazine derivatives to their sulphoxides and other metabolites. Insecticides are extensively used in residential and commercial applications. Human exposure to insecticides in today’s environment is more a matter of extent than a possibility. The acute toxicity of numerous insecticides is well documented. With increasing awareness of human exposure to insecticide via food and water, there is increasing concern regarding their possible deleterious effects on the developing conceptus. Many studies, conducted with animal models, have identified several insecticides as teratogens. An investigation of the effect of organophosphorus insecticides in pregnant rats revealed a positive correlation between dosage levels and malformations (Kimbrough and Gaines, 1968). Casida and co-workers (Proctor, Moscioni and Casida, 1976; Moscioni, Engel and Casida, 1977; Seifert and Casida, 1978) studied over 70 pesticides and found teratogenic effects induced in chicken embryos following exposure to several organophosphorus and N-methylcarbamate insecticides. Additional reports exist on the teratogenecity of insecticides in animal models (Henderson and Kitos, 1982; Byrne and Kitos, 1983). Though cases of accidental and intentional insecticide poisoning are reported every year, the necessary scientific data do not exist to conclusively establish the cause and the effect relationship on the teratogenecity of insecticides in humans. The existing information is also extremely limited regarding the pathways involved in the metabolism of insecticides by the human placenta or other tissues. LOs, a family of non-haem, iron-containing proteins capable of multiple activities, are ubiquitously found in plant and animal tissues. LOs perform dioxygenation of polyunsaturated fatty acids containing a cis, cis-1,4-pentadiene moiety to yield the corresponding hydroperoxides, the first step in the biosynthesis of several biologically active metabolites. Additionally, the role of the LO pathway in the oxidative metabolism of xenobiotics has been extensively investigated (Kulkarni, 1996). Purified human term placental LO (HTPLO) has been employed in many studies (Joseph, Srinivasan and Kulkarni, 1993; Joseph et al., 1994; Datta and Kulkarni, 1994; Datta, Sherblom and Kulkarni, 1997; Rajadhyaksha et al., 1999) to understand better the bioactivation of various animal/human teratogens and other chemicals. Since the available information on purification and characteristics of LOs from major human organs is scant, HTPLO also serves as a model to demonstrate the ability of LOs from different human tissues to metabolize xenobiotics.

647

Earlier studies demonstrated that SLO is capable of desulphuration and dearylation of the organophosphorus insecticide parathion (Naidu, Naidu and Kulkarni, 1991a) and epoxidation of the organochlorine insecticide aldrin (Naidu, Naidu and Kulkarni, 1991b). Hu and Kulkarni (1998) published the results of a study focused on SLO-mediated N-demethylation of aminocarb and other insecticides in the presence of H2O2. Most recently, N-demethylation of phenothiazines by both SLO and HTPLO was observed in the presence of H2O2 (Rajadhyaksha et al., 1999). Since the in vivo concentration of H2O2 in the placenta and other tissues is expected to be limited due to the activities of catalase and GSH peroxidases, it is essential to investigate the extent of LO-mediated N-demethylation of xenobiotics in the presence of polyunsaturated fatty acids such as linoleic acid (LA), arachidonic acid and
-linolenic acid. In this study, we tested the hypothesis for the first time that LA and other endogenous fatty acids may support the HTPLO-mediated N-demethylation of selected phenothiazines and insecticides. For comparison, similar data for the SLO-catalyzed reaction are also presented.

MATERIALS AND METHODS Chemicals SLO, Type V (630 000 Sigma units per mg protein), Concanavalin A-Sepharose 4B (Con-A), LA 99 per cent pure, CPZ, promazine, promethazine, trimeprazine, aminocarb (4-dimethylamino-m-tolyl-methyl-carbamate), nor-dihydroguaiaretic acid (NDGA), gossypol, butylated hydroxyanisole (BHA), butylated hydroxytoluene (BHT), ascorbate, reduced glutathione (GSH) and dithiothreitol (DTT) were purchased from Sigma Chemical Co., St Louis, MO, USA. Zectran (4-dimethylamino-3,5-xylyl methylcarbamate) (98 per cent), chlordimeform (N,N-dimethyl-N u -[2-methyl-4-chlorophenyl]formamidine) (99.5 per cent), and dicrotophos (2dimethylcarbamoyl-1-methylvinyl dimethylphosphate) (98 per cent) were purchased from ChemService, Inc., West Chester, PA, USA. Arachidonic acid and
-linolenic acid were purchased from Calbiochem (La Jolla, CA, USA. All other chemicals used in this study were of reagent grade.

Enzyme preparation Human term placentae (n=9) obtained from healthy women with no known history of smoking, drug or alcohol abuse, or pathological/physiological problems were used in this study. Placentae, obtained from a local hospital immediately after normal delivery, were kept on ice and processed within an hour following delivery. The University of South Florida’s Institutional Review Board governing the policies and procedures for research involving human subjects approved the use of human term placentae in this study. The affinity chromatography procedure described by Joseph, Srinivasan and Kulkarni

Placenta (2000), Vol. 21

648

0.2 A Absorbance

(1993), which yields electrophoretically homogenous HTPLO, was employed to obtain preparations of the enzyme. Stated briefly, a sample of placental cytosol obtained by differential centrifugation was applied to a Con-A column. After elution, the active fractions containing HTPLO were pooled and dialyzed overnight against 50 m Tris buffer, pH 6.5 with two changes of buffer. The dioxygenase activity of the HTPLO preparations was assayed as described previously (Joseph, Srinivasan and Kulkarni, 1993) using LA as the substrate. Protein content of enzyme preparations was determined according to Bradford (1976).

0.0 400.0

RESULTS CPZ was selected as a prototypic phenothiazine drug to examine the formation of the cation radical. Under acidic conditions (50 m acetate buffer, pH 3.5), a peak with an absorption maximum approx 525 nm [Figure 1(A and B)] was observed when active HTPLO or SLO was incubated with 5 m CPZ and LA. Previous studies have established that the approx 525 nm peak is due to the CPZ cation radical formed during CPZ oxidation by methaemoglobin (Kelder et al., 1991a), horseradish peroxidase (Vazquez et al., 1992; Kelder et al., 1994), myeloperoxidase (van Zyl et al., 1990), and human placental peroxidase (Yang and Kulkarni, 1997) as well as by SLO (Perez-Gilabert et al., 1994a, 1994b; Rajadhyaksha et al., 1999) and HTPLO (Rajadhyaksha et al., 1999). Due to instability, the cation radical species was undetectable when

500.0

600.0

Wavelength (nm) 0.8

Enzyme assay

B 0.6 Absorbance

Except as otherwise noted, 100 g HTPLO in 50 m phosphate buffer, pH 6.5, was mixed with 5 m desired substrate. After 2 min pre-incubation, the reaction was initiated by the addition of 3 m LA (1.0 ml final volume). A control (containing all components except enzyme) was run in each set of experiments. After incubation at 37C for 30 min, the reaction was terminated by the addition of 1 ml cold 10 per cent trichloroacetic acid. Formaldehyde accumulation in the reaction medium as a result of N-demethylation of CPZ (or other test chemical) was assayed spectrophotometrically at 412 nm using the double strength Nash reagent (Nash, 1953). An identical experimental protocol was used in the studies with SLO. The data reported are corrected for non-enzymatic reaction. The experimental protocol described above was employed with slight modification to assess the effect of inhibitors of LO (NDGA, gossypol), free radical scavengers (BHA, BHT) and reducing agents (ascorbate, GSH, DTT) on the Ndemethylation of CPZ and aminocarb by HTPLO. In each case, HTPLO (100 g) was pre-incubated in the buffer for 2–3 min with the specified concentration of the desired modifier before the addition of 5.0 m CPZ or aminocarb and 3.0 m LA. Each experiment was repeated at least three times with a different enzyme preparation and the data are presented as mean.

0.1

0.4 0.2 0.0 400.0

500.0

600.0

Wavelength (nm) Figure 1. Optical difference spectra obtained during human term placental lipoxygenase (HTPLO)-mediated N-demethylation of chlorpromazine (CPZ). The reaction media contained either (A) 100 g HTPLO (absent in the reference cuvette), 5 m chlorpromazine, and 3.0 m linoleic acid (LA) in 50 m citrate buffer, pH 3.5 or (B) 20 g SLO (absent in the reference cuvette), 5 m chlorpromazine, and 3.0 m LA in 50 m citrate buffer, pH 3.5. The spectrum was recorded 5 min after initiation of the reaction by the addition of LA.

experiments were conducted in 50 m phosphate buffer, pH 6.5, the optimal pH for the HTPLO-mediated N-demethylation of CPZ. Pilot experiments (data not shown), designed to establish optimal N-demethylation assay conditions, clearly indicated that formaldehyde accumulates in the reaction medium when active HTPLO was incubated with a selected phenothiazine/ insecticide and LA. Subsequently, xenobiotic Ndemethylation was studied under a variety of experimental conditions. As shown in Figure 2(A), the maximal rate of formaldehyde generation from CPZ was observed when the pH was approximately 6.5. An increase in the pH resulted in a decline in the rate of formaldehyde production. At the physiological pH of 7.4, approximately 50 per cent of the maximal reaction rate was noted. Formaldehyde production was also dependent on the incubation time. The reaction was linear for over 40 min [Figure 3(A)]. In view of this, all other experiments were terminated at 30 min. As shown in Figure 3(B), the rate of formaldehyde accumulation increased proportionally with an increase in enzyme. When experiments were conducted in the presence of varying concentrations of LA, a concentration of 3 m LA yielded the highest reaction rate [Figure 3(C)]. Increasing the LA concentration (>3 m) further resulted in no further increase in the formaldehyde production. The data presented in Figure 3(D) indicate the

nmoles HCHO/min/mg HTPLO

Hover and Kulkarni: Lipoxygenase-Mediated N-Demethylation of Xenobiotics

2.0 A

1.0

0.0

5

6

7

8

7

8

nmoles HCHO/min/mg HTPLO

pH

B 2.0

1.0

0.0

5

6 pH

Figure 2. (A) Effect of pH on human term placental lipoxygenase (HTPLO)-mediated N-demethylation of chlorpromazine (CPZ). The reaction media contained 100 mg protein (absent in the control) 5 m CPZ, and 3.0 m linoleic acid (LA) in 50 m phosphate buffer at the indicated pH. See Materials and Methods for further details. (B) Effect of pH on HTPLOmediated N-demethylation of aminocarb (AC). The reaction media contained 100 g protein (absent in the control) 5 m aminocarb, and 3.0 m linoleic acid (LA) in 50 m phosphate buffer of the indicated pH. See Materials and Methods for further details.

formaldehyde accumulation in the reaction medium as the function of the CPZ concentration. The calculation of enzyme kinetic data by Lineweaver–Burk plots yielded an apparent KM value of 2.4 m for CPZ and an apparent Vmax value of 2.2 nmoles of formaldehyde/min/mg HTPLO. The estimated KM for LA in this reaction was determined to be 1.6 m. Essentially similar results were observed with SLO (data not shown). A significant concentration-dependent decrease in the CPZ N-demethylase activity was observed when NDGA, a classical LO inhibitor, was added to the incubation medium (Table 1). A similar inhibitory response was also noted when gossypol, another known LO inhibitor, was included in the reaction medium. The incorporation of free radical scavengers, BHA or BHT, in the incubation mixture also suppressed the production of formaldehyde from CPZ in a concentration-dependent manner, suggesting an involvement of free radicals in the reaction. Similarly, the inclusion of a reducing agent, such as ascorbate, GSH, or DTT in the reaction medium caused significant suppression of the reaction. Other phenothiazines tested under identical incubation conditions were also similarly oxidized by HTPLO in the presence of LA (Table 2). The data demonstrating SLO-mediated

649

N-demethylation of CPZ and other phenothiazines in the presence of LA are also included for comparison (Table 2). The observations indicate promethazine serves as the best substrate for N-demethylation by both SLO and HTPLO while other phenothiazines exhibit relatively low specific activities when compared with CPZ. The carbamate insecticide, aminocarb, was used as the prototype to examine HTPLO-mediated N-demethylation of insecticides. Using formaldehyde production as the end point, several pilot studies were conducted to optimize the assay conditions for N-demethylation of aminocarb (data not shown). The results of these experiments indicated the presence of an active enzyme is necessary to observe the reaction. No significant amount of formaldehyde was detected when either LA was absent or a boiled enzyme preparation was used. The enzymatic N-demethylation of aminocarb was evaluated under a variety of experimental conditions. A maximal rate of formaldehyde production was observed when the pH of the incubation medium was 6.5 [Figure 2(B)]. At the physiological pH of 7.4, the relative activity was approximately 80 per cent of the maximal rate. The formaldehyde production was dependent upon the incubation time, with a linear rate exceeding 30 min [Figure 4(A)]. In view of this, the reaction was terminated at 30 min in all subsequent experiments. As shown in Figure 4(B), the amount of formaldehyde formed increased with an increase in enzyme concentration. The reaction exhibited a dependence on the concentration of LA [Figure 4(C)] as well as aminocarb [Figure 4(D)] in the incubation medium. Analysis of the data by Lineweaver–Burk plots yielded an apparent KM value of 2.7 m for aminocarb and an apparent Vmax value of 2.8 nmoles of formaldehyde/min/mg HTPLO. From similar experiments conducted in the presence of varying concentrations of LA, an apparent KM value of 1.2 m was obtained for LA. Taken together these results clearly suggest the enzymatic nature of the reaction. The data presented in Table 1 indicate a concentrationdependent decline in the enzymatic formaldehyde formation from aminocarb when known inhibitors of LO, namely NDGA and gossypol, were added to the reaction media. The free radical scavengers, BHA and BHT also reduced the rate of formaldehyde production in a dose dependent manner, suggesting the involvement of free radicals. Similar concentrationdependent reduction in the rate of reaction was observed when a reducing agent, such as ascorbate, GSH or DTT, was included in the reaction medium. The ability of HTPLO to mediate the N-demethylation of other insecticides in the presence of LA was also studied using the optimal conditions determined for aminocarb (Table 2). The results of SLO-mediated N-demethylation of AC and other insecticides in the presence of LA are also included for comparison (Table 2). The specific activity data presented in Table 2 indicate aminocarb is the best substrate for HTPLO and SLO-mediated N-demethylation. Zectran was N-demethylated at approximately half the rate of aminocarb, while the rate of formaldehyde release from either chlordimeform or dicrotophos was comparatively low.

Placenta (2000), Vol. 21

0.4

80

B

A nmoles HCHO/min

nmoles HCHO/mg HTPLO

650

60 40 20

10

0

20

30

0.3 0.2 0.1

0

40

50

1

2

3

4

5

Linoleic acid (mM)

nmoles HCHO/min/mg HTPLO

nmoles HCHO/min/mg HTPLO

C 2

1

150

200

HTPLO (µg/ml)

Time (min)

0

100

2

D

1

0

5

10

Chlorpromazine (mM)

Figure 3. Formaldehyde production during human term placental lipoxygenase (HTPLO)-mediated N-demethylation of chlorpromazine (CPZ) under different experimental conditions. The assay conditions are specified in the inserts. Effect of (A) incubation time; (B) HTPLO concentration; (C) linoleic acid (LA) concentration; and (D) CPZ concentration. See Materials and Methods for further details.

Other endogenous fatty acids, such as arachidonic and
-linolenic acid were substituted for LA in the incubation medium to determine their potential to support the HTPLO catalyzed N-demethylation of the CPZ and aminocarb. The data (Table 3) indicated that the rates of N-demethylation CPZ and aminocarb by HTPLO and SLO are relatively low, approximately 18 per cent as compared to the LA supported reaction.

DISCUSSION The placenta permits bidirectional transfer of endogenous and exogenous chemicals between maternal and fetal compartments. The possible metabolic involvement of this tissue is of interest to understand the biochemical mechanism of teratogens and developmental toxicants. After years of research, Juchau (1995) concluded that cytochrome P450 capable of xenobiotic oxidation was absent in the placentae of nonsmoking women. A similar opinion was recently expressed by Hakkola et al. (1998). Other microsomal enzymes, viz. flavincontaining mono-oxygenase and prostaglandin synthase, are also undetectable in the placentae of non-smokers (Kulkarni, 1996). In contrast, peroxidase (Yang and Kulkarni, 1997) and LO are abundant in the human term placenta (Joseph, Srinivasan and Kulkarni, 1993) and these enzymes also occur

in the human conceptual tissues at early and mid-gestation (Joseph, Srinivasan, and Kulkarni, 1994; Joseph et al., 1994; Datta and Kulkarni, 1994; Datta et al., 1995; Kulkarni and Murthy, 1995). Additionally, the studies with laboratory animals have documented that LO activity capable of xenobiotic co-oxidation occurs in the rat embryos (Roy et al., 1993) and it plays a significant role in the phenytoin teratogenesis in the mouse (Yu and Wells, 1995). Two distinct modes of operation have been described for the LO-mediated xenobiotic oxidation. One involves a fatty acid peroxyl radical-mediated xenobiotic co-oxidation (Byczkowski and Kulkarni, 1989) while the other depends on the hydroperoxidase activity of LO supported by either H2O2 (Kulkarni and Cook, 1988) or fatty acid hydroperoxides (Clapp, Banerjee and Rotenberg, 1985; Mansuy et al., 1988). Under the experimental conditions employed, it is difficult to discern quantitatively which pathway(s) operates during N-demethylation of the phenothiazines and insecticides tested. In any case, since the steady state in vivo concentrations of H2O2 and polyunsaturated fatty acid hydroperoxides in the placenta are expected to be relatively low, the contribution of the hydroperoxidase activity of LO toward xenobiotic oxidation may be somewhat limited. On the other hand, LA is a dietary essential fatty acid from which arachidonic and
-linolenic acids are biosynthesized in the human body. Arachidonic acid is an integral part of membrane lipids and is abundant (210 ng/mg)

Hover and Kulkarni: Lipoxygenase-Mediated N-Demethylation of Xenobiotics

Table 1. Effect of modifiers on N-demethylation of chlorpromazine and aminocarb by human term placental lipoxygenase (HTPLO) in the presence of linoleic acid (LA)

Modifier — NDGA Gossypol BHA BHT Ascorbate Glutathione DTT

Concentration () — 50 100 250 5 10 20 100 250 100 250 100 250 500 100 250 500 50 100 200

651

Table 2. N-demethylation of phenothiazines and insecticides by soybean lipoxygenase (SLO) and human term placental lipoxygenase (HTPLO) in the presence of linoleic acid (LA)

N-demethylation relative activity

Specific activity (nmoles HCHO/min)

Chlorpromazine

Aminocarb

Substrate

100 69 52 21 70 66 45 89 56 77 51 69 39 21 68 61 44 92 79 66

100 64 50 20 63 59 42 91 58 78 55 66 41 21 67 60 45 93 74 65

Phenothiazine Chlorpromazine Promazine Promethazine Trimeprazine Insecticide Aminocarb Chlordimeform Dicrotophos Zectran

The incubation media (1.0 ml final volume) contained 100 g HTPLO (enzyme absent in the control), 5 m chlorpromazine or aminocarb, 3.0 m LA and the indicated concentration of modifier in 50 m phosphate buffer pH 6.5. The specified modifier and HTPLO were pre-incubated for 2 min prior to the addition of other components. All incubations were performed at 37C for 30 min. Values reported are mean of at least three observations. The specific activities noted in the control experiments without a modifier were 1.70.3 and 2.20.3 nmoles HCHO/min/mg HTPLO for chlorpromazine and aminocarb, respectively. NDGA: nordihydroguaiaretic acid; BHA: butylated hydroxyanisole; BHT: butylated hydroxytoluene; DTT: dithiothreitol.

in human placental tissue (Ogburn et al., 1988). In light of this, the polyunsaturated fatty acid-coupled co-oxidation of xenobiotics in human placenta may be viewed as both important and relevant. The available experimental evidence gathered with different protein catalysts (Levy et al., 1972; De Mol and Busker, 1984; De Mol et al., 1986; Kelder et al., 1991a, 1991b) has established that the formation of the phenothiazine cation radical represents a bioactivation process that leads to the covalent binding of phenothiazines to proteins and nucleic acids. Yang and Kulkarni (1997) demonstrated the ability of the human term placental peroxidase to oxidize phenothiazines to the corresponding cation radicals in the presence of H2O2. SLO was also shown to mediate the formation of cation free radicals as the initial metabolites during oxidation of phenothiazines (Perez-Gilabert, Sanchez-Ferrer and Garcia-Carmona, 1994a, 1994b; Rajadhyaksha et al., 1999) in the presence of H2O2. The spectral data shown in Figures 1(A and B) indicate that similar formation of CPZ cation radical occurs when reaction media containing HTPLO or SLO are supplemented with LA. If the

Per mg SLO

Per mg HTPLO

16 1.4 9.11.0 110 9.9 28 3.1

1.70.3 1.00.2 11.31.2 3.10.5

21 2.0 1.60.2 1.70.2 12 1.6

2.20.3 0.30.06 0.40.07 1.90.2

The incubation media (1.0 ml final volume) contained 20 g SLO and 1.0 m LA or 100 g HTPLO and 3.0 m LA, 5.0 m specified substrate in 50 m phosphate buffer pH 6.5. The enzyme was absent in the control incubation. All incubations were performed at 37C for 30 min. The amount of formaldehyde produced was estimated by Nash reagent. Values are mean (nd3).

CPZ cation radical formation by peroxidase and/or LO and subsequent covalent binding to macromolecules occurs in vivo, some impairment in the placental functions is expected. Previously, it has been proposed that the first step in the LO-mediated H2O2-supported N-demethylation of aminopyrine (Perez-Gilabert, Sanchez-Farrer, and Garcia-Carmona, 1997; Yang and Kulkarni, 1998), aminocarb (Hu and Kulkarni, 1998) and phenothiazines (Rajadhyaksha et al., 1999) involves one electron oxidation of the substrate to a nitrogen-centered cation radical that undergoes either deprotonation (-H + ) or loses a hydrogen atom (-H•) to form an iminium cation. Subsequent hydrolysis of the iminium cation yields the final stable metabolite and HCHO. A similar mechanism is believed to be responsible for the LA-dependent N-demethylation of phenothiazines and insecticides tested in this study. Product analysis was not conducted since besides enzymatic N-demethylation, it is known that a non-enzymatic formation of sulphoxide and sulphone (and a few other products) occurs from CPZ cation radical generated by SLO (Perez-Gilabert et al., 1994b). Considering the generation of multiple metabolites involving the enzymatic and non-enzymatic catalysis of the substrate and the possibility that both lipid peroxide and lipid hydroperoxide may support co-oxidation, the stoichiometry of CPZ N-demethylation was not studied. The relatively low specific activities of CPZ N-demethylation in the presence of arachidonic and
-linolenic acids may be the reflection of their relatively weak ability to support the dioxygenase activity of the HTPLO (Joseph, Srinivasan and Kulkarni, 1993). The previous comparative studies conducted with partially purified LO and peroxidase have established that the enzymes isolated from the early and mid-gestational human intrauterine conceptual tissues exhibit a remarkable qualitative similarity

Placenta (2000), Vol. 21

652

A

0.5 nmoles HCHO/min

nmoles HCHO/mg HTPLO

100

50

0

10

20

40

30

B

0.4 0.3 0.2 0.1 0

50

2

1

2

3

4

5

nmoles HCHO/min/mg HTPLO

nmoles HCHO/min/mg HTPLO

C

1

150

200

HTPLO (µg/ml)

Time (min)

0

100

3 D 2

1

0

5

10

Aminocarb (mM)

Linoleic acid (mM)

Figure 4. Formaldehyde production during human term placental lipoxygenase (HTPLO)-mediated N-demethylation of aminocarb (AC) under different experimental conditions. The assay conditions are specified in the inserts. Effect of (A) incubation time; (B) HTPLO concentration; (C) linoleic acid (LA) concentration; and (D) AC concentration. See Materials and Methods for further details.

Table 3. N-demethylation of chlorpromazine and aminocarb by human term placental lipoxygenase (HTPLO) in the presence of polyunsaturated fatty acids N-demethylation (nmoles HCHO/min/mg HTPLO)

Fatty acid

Concentration (m) Chlorpromazine

Linoleic acid 3.0 Arachidonic acid 3.0
-Linolenic acid 3.0

1.70.29 0.30.08 0.20.06

Aminocarb 2.20.28 0.40.07 0.40.09

The incubation media (1.0 ml final volume) contained 100 g HTPLO (absent in the control), 5 m chlorpromazine or aminocarb, and the indicated concentration of fatty acid in 50 m phosphate buffer pH 6.5. All incubations were performed at 37C for 30 min. The amount of formaldehyde produced was estimated by Nash reagent. Values are mean (nd3).

N-demehylation reaction catalyzed by these enzymes may offer some protection to the human conceptus from the toxicity of the phenothiazines and insecticides. In summary, this study tested a hypothesis that LO may be involved in the enzymatic N-demethylation of phenothiazines and insecticides in the presence of polyunsaturated fatty acids. The evidence presented here suggests, for the first time, that SLO and HTPLO are capable of catalyzing N-demethylation of phenothiazines and insecticides in the presence of LA and other fatty acids. The presence of high titers of peroxidase and HTPLO capable of catalyzing xenobiotic oxidation in the human term placenta and the existence of a similar potential in the human intrauterine conceptual tissues may explain, at least in part, the possible developmental toxicity associated with the exposure to the phenothiazines and insecticides. REFERENCES

with those obtained from the term placenta in terms of their ability to oxidize xenobiotics (Joseph et al., 1994; Datta and Kulkarni, 1994; Datta et al., 1995). Although the ability of the human embryonic/fetal tissue LO and peroxidase was not tested in this study, considering these reports, it can be presumed that the phenothiazines and insecticides will serve as excellent substrates for oxidation by these enzymes. Thus, on one hand, the resulting substrate-derived free radicals and covalent binding may directly contribute to the developmental toxicity of these chemicals; while on the other hand, the

Altshuler LL, Cohen L, Szuba MP, Burt VK, Gitlin M & Mintz J (1996) Pharmacologic management of psychiatric illness during pregnancy: dilemmas and guidelines. Am J Psychiatry, 153, 592–606. Austin M-PV & Mitchell PB (1998) Psychotropic medications in pregnant women: treatment dilemmas. Med J Australia, 169, 228–431. Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram protein utilizing the principle of protein-dye binding. Anal Biochem, 72, 248–254. Byczkowski JZ & Kulkarni AP (1989) Lipoxygenase-catalyzed epoxidation of benzo(a)pyrene-7,8-dihydrodiol. Biochem Biophys Res Commun, 159, 1199–1205. Byrne HB & Kitos PA (1983) Teratogenic effects of cholinergic insecticides in chick embryos-IV. Biochem Pharmacol, 32, 2881–2890.

Hover and Kulkarni: Lipoxygenase-Mediated N-Demethylation of Xenobiotics Clapp CH, Banerjee A & Rotenberg SA (1985) Inhibition of soybean lipoxygenase 1 by N-alkylhydroxylamines. Biochemistry, 24, 1826–1830. Datta K & Kulkarni AP (1994) Oxidative metabolism of aflatoxin B1 by lipoxygenase purified from human term placenta and intrauterine conceptual tissues. Teratology, 50, 311–317. Datta K, Joseph P, Roy S, Srinivasan S & Kulkarni AP (1995) Peroxidative xenobiotic oxidation by partially purified peroxidase and lipoxygenase from human fetal tissues at 10 weeks of gestation. Gen Pharmacol, 26, 107–112. Datta K, Sherblom PM & Kulkarni AP (1997) Co-oxidative metabolism of 4-aminobiphenyl by lipoxygenase from soybean and human term placenta. Drug Metab Dispos, 25, 196–205. De Mol NJ, Becht ABC, Koenen J & Lodder G (1986) Irreversible binding with biological macromolecules and effects in bacterial mutagenicity tests of the radical cation of promethazine and photoactivated promethazine. Comparison with chlorpromazine. Chem Biol Interact, 57, 73–83. De Mol NJ & Busker RW (1984) Irreversible binding of the chlorpromazine radical cation and of photoactivated chlorpromazine to biological macromolecules. Chem Biol Interact, 52, 79–92. Edlund MJ & Craig TJ (1984) Antipsychotic drug use and birth defects: an epidemiologic reassessment. Comparative Psychiatry, 25, 32–38. Hakkola J, Pelkonen O, Pasanen M & Raunio H (1998) Xenobioticmetabolizing cytochrome P450 enzymes in the human feto-placental unit: Role in intrauterine toxicity. Crit Rev Toxicol, 28, 35–72. Henderson M & Kitos PA (1982) Do organophosphate insecticides inhibit the conversion of tryotophan to NAD + in ovo? Teratology, 26, 173–181. Hu J & Kulkarni AP (1998) Lipoxygenase-mediated demethylation of pesticides. Pesticide Biochemistry and Physiology, 61, 145–153. Joseph P, Srinivasan NS & Kulkarni AP (1993) Purification and partial characterization of lipoxygenase with dual catalytic activities from human term placenta. Biochem J, 293, 3–91. Joseph P, Srinivasan SN & Kulkarni AP (1994) Xenobiotic oxidation during early pregnancy in man: peroxidase catalyzed chemical oxidation in conceptal tissues. Xenobiotica, 24, 83–590. Joseph P, Srinivasan NS, Byczkowski JZ & Kulkarni AP (1994) Bioactivation of benzo(a)pyrene-7,8-dihydrodiol catalyzed by lipoxygenase purified from human term placenta and conceptal tissues. Reprod Toxicol, 8, 307–313. Juchau MR (1995) Placental enzymes: Cytochrome P450s and their significance. In Placental Toxicology (Ed.) Rama Sastry BV, pp. 197–212. Boca Raton, FL: CRC Press. Kelder PP, Fischer MJE, De Mol NJ & Janssen LHM (1991a) Oxidation of chlorpromazine by methemoglobin in the presence of hydrogen peroxide: formation of chlorpromazine radical cation and its covalent binding to methemoglobin. Arch Biochem Biophys, 284, 313–319. Kelder PP, De Mol NJ, Hart BA & Janssen LHM (1991b) Metabolic activation of chlorpromazine by stimulated human polymorphonuclear leukocytes: induction of covalent binding of chlorpromazine to nucleic acids and proteins. Chem Biol Interact, 79, 15–30. Kelder PP, De Mol NJ, Fischer MJE & Janssen LHM (1994) Kinetic evaluation of the oxidation of phenothiazine derivatives by methemoglobin and horseradish peroxidase in the presence of hydrogen peroxide. Implications for the reaction mechanisms. Biochim Biophys Acta, 1205, 230–238. Kimbrough RD & Gaines TB (1968) Effect of organic phosphorus compounds and alkylating agents on the rat fetus. Arch Environ Health, 16, 805–808. Kulkarni AP (1996) Role of xenobiotic metabolism in developmental toxicity, Chapter 12. In Handbook of Developmental Toxicology (Ed.) Hood R, pp. 383–421. Boca Raton, FL: CRC Press. Kulkarni AP & Cook DC (1988) Hydroperoxidase activity of lipoxygenase: Hydrogen peroxide-dependent oxidation of xenobiotics. Biochem Biophys Res Commun, 155, 1075–1081. Kulkarni AP & Murthy KR (1995) Xenobiotic metabolism in humans during early pregnancy: Peroxidase-mediated oxidation and bioactivation of 2-aminofluorene. Xenobiotica, 25, 799–810. Levy L, Tozer TN, Tuck LD & Loveland DB (1972) Stability of some phenothiazine radicals. J Med Chem, 15, 898–905. Mansuy D, Cucurou C, Biatry B & Battioni JP (1988) Soybean lipoxygenase-catalyzed oxidations by linoleic acid hydroperoxide: Different reducing substrates and dehydrogenation of phenidone and BW 755C. Biochem Biophys Res Commun, 151, 339–346. Milkovich L & van den Berg BJ (1976) An evaluation of the teratogenicity of certain antinauseant drugs. Am J Obstet Gynecol, 125, 244–248.

653 Moscioni DA, Engel JL & Casida JE (1977) Kynurenine formamidase inhibition as a possible mechanism for certain teratogenic effects of organophosphorus and methylcarbamate insecticides in chicken embryos. Biochem Pharmacol, 26, 2251–2258. Naidu AK, Naidu AK & Kulkarni AP (1991a) Role of lipoxygenase in xenobiotic oxidation: Parathion metabolism catalyzed by highly purified soybean lipoxygenase. Pesticide Biochemistry and Physiology, 40, 150–158. Naidu AK, Naidu AK & Kulkarni AP (1991b) Aldrin epoxidation: Catalytic potential of lipoxygenase coupled with linoleic oxidation. Drug Metabol Dispos, 19, 758–763. Nash T (1953) The colorimetric estimation of formaldehyde by means of the Hantzsch reaction. Biochem J, 55, 416–421. Ogburn PL, Roberts JD, Dassoupoulos T & Walker JL (1988) Leukotriene B4, 6-keto-prostaglandin F1, thromboxane B2 and free arachidonic acid levels in human placental tissues. Ann NY Acad Sci, 524, 434. Perez-Gilabert M, Sanchez-Ferrer A & Garcia-Carmona F (1994a) Enzymatic oxidation of phenothiazines by lipoxygenase/H2O2 system. Biochem Pharmacol, 47, 2227–2232. Perez-Gilabert M, Sanchez-Ferrer A & Garcia-Carmona F (1994b) Lipoxygenase catalyzed oxidation of chlorpromazine by hydrogen peroxide at acidic pH. Biochim Biophys Acta, 1214, 203–208. Perez-Gilabert M, Sanchez-Ferrer A & Garcia-Carmona F (1997) Oxidation of aminopyrine by the hydroperoxidase activity of lipoxygenase: a new proposed mechanism of N-demethylation. Free Radic Biol Med, 23, 548–555. Potter WZ & Hollister LE (1998) Antipsychotic agents and lithium. In Basic and Clinical Pharmacology (Ed.) Katzung BG, pp. 464–495. Stamford, Connecticut: Appleton & Lang. Proctor NH, Moscioni DA & Casida JE (1976) Chicken embryo NAD levels lowered by teratogenic organophosphorus and methylcarbamate insecticides. Biochem Pharmacol, 25, 757–762. Rajadhyaksha AV, Reddy V, Hover CG & Kulkarni AP (1999) N-demthylation of phenothiazines by lipoxygenase from soybean and human term placenta in the presence of hydrogen peroxide. Teratog, Carcinog Mutagen, 19, 211–222. Roy SK, Mitra AK, Hilbelink DR, Dwornik JJ & Kulkarni AP (1993) Lipoxygenase activity in rat embryos and its potential for xenobiotic oxidation. Biol Neonate, 63, 297–302,. Seifert J & Casida JE (1978) Relation of yolk sac membrane kynurenine formamidase inhibition to certain teratogenic effects of organophosphorus insecticides and of carbaryl and eserine in chicken embryos. Biochem Pharmacol, 27, 2611–2615. Slone D, Siskind V, Heinonen OP, Manson RR, Kaufman DW & Shapiro S (1977) Antenatal exposure to the phenothiazines in relation to congenital malformations, perinatal mortality rate, birth weight and intelligence quotient score. Am J Obstet Gynecol, 128, 486–488. Sobel DE (1960) Fetal damage due to ECT, insulin coma, chlorpromazine, or reserpine. Arch Gen Psychiatry, 2, 606–611. Szabo KT & Brent RL (1974) Species differences in experimental teratogenesis by tranquilizing agents. Lancet, 1, 565. Valoti M, Frosini M, Palmi M, De Matteis F & Sgaragli G (1998) N-dealkylation of chlorimipramine and chlorpromazine by rat liver microsomal cytochrome P450 isozymes. J Pharm Pharmacol, 50, 1005–1011. Van Zyl JM, Basson K, Kriegler A & van der Walt B (1990) Activation of chlorpromazine by the myeloperoxidase system of the human neutrophil. Biochem Pharmacol, 40, 947–954. Vazquez A, Tudela J, Varon R, Garcia-Canovas F (1992) Determination of the molar absorptivities of phenothiazine cation radicals generated by oxidation with hydrogen peroxide/peroxidase. Anal Biochem, 202, 245–248. Walker BE & Patterson A (1974) Induction of cleft palate in mice by tranquilizers and barbituates. Teratology, 10, 159–164. Yang X & Kulkarni AP (1997) Oxidation of phenothiazines by human term placental peroxidase in non-smokers. Teratog, Carcinog Mutagen, 17, 139– 151. Yang X & Kulkarni AP (1998) N-dealkylation of aminopyrine catalyzed by soybean lipoxygenase in the presence of hydrogen peroxide. J Biochem Mol Toxicol, 12, 175–183. Yu WK & Wells PG (1995) Evidence for lipoxygenase-catalyzed bioactivation of phenytoin to a teratogenic reactive intermediate: in vitro studies using linoleic acid-dependent soybean lipoxygenase, and in vivo studies using pregnant CD-1 mice. Toxicol Appl Pharmacol, 131, 1–12.