Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Cocaine abuse during pregnancy continues to be a serious problem in the inner cities of the United States (Dudish and Hatsukami, 1996; Richardson et al., 1999; Scher et al., 2000). Consequently, there is a need to understand the pharmacokinetics of cocaine in pregnant mothers and their fetuses. For obvious reasons, the data on transplacental pharmacokinetics of cocaine in humans are very limited and confined to a few post-mortem case reports (Mittleman et al., 1989; Apple and Roe, 1990; Klein et al., 1992). Also, little is known about cocaine pharmacokinetics in fetuses of other primates. We were only able to find a single report describing the changes in the maternal/fetal concentration of cocaine injected intramuscularly to anesthetized near term macaque monkeys (Binienda et al., 1993).

Over the last several years, we have been involved in the examination of the consequences of the prenatal cocaine exposure in rhesus monkeys born from mothers receiving oral drug treatment at mid-gestation (Lidow, 1995, 1998; Lidow and Song, 2001; Lidow et al., 2001). It has been demonstrated that oral cocaine administration can serve as a good laboratory model of cocaine snorting (Van Dyke et al., 1977; Wilkinson et al., 1980; Jufer et al., 1998). Indeed, administration of similar doses of cocaine by each of these routes results in virtually identical levels of cocaine in plasma (Van Dyke et al., 1977; Wilkinson et al., 1980; Fattinger et al., 2000). The kinetics of postpeak decline in plasma cocaine levels are also the same for both routes of administration (Van Dyke et al., 1977; Wilkinson et al., 1980). Finally, both intranasal and oral cocaine administrations produce relatively high blood levels of cocaine metabolites (Cone et al., 1994; Jufer et al., 1998). This is due to the fact that a significant portion of the snorted cocaine reaches the gastrointestinal tract and, thus, is processed by the organism as an orally administered drug (Cone et al., 1994; Fattinger et al., 2000). The only difference in the pharmacokinetics of cocaine taken by the intranasal and oral routes is that in the case of the former route of administration the peak plasma levels of this drug occur about 30 min earlier than in the case of the administration by the latter route (Inaba, 1989). Our studies revealed that cocaine administered to pregnant monkeys in accordance with our model can induce significant alterations in cerebral cortical development (Lidow, 1995, 1998; He et al., 1999; Lidow and Song, 2001; Lidow et al., 2001).

The present article describes the maternal and fetal pharmacokinetics of cocaine and its major metabolite, benzoylecgonine, during the oral administration of cocaine to mid-term pregnant monkeys used in our model of prenatal drug exposure. A separate analysis was performed for a single oral dose and the last dose of cocaine in the chronic oral daily drug treatment. We also compared the pharmacokinetics of cocaine given orally and by a single intravenous injection. In addition, our study included analysis of the pharmacokinetics of cocaine administered by multiple daily intravenous cocaine injections. The latter mode of the treatment was designed 1) to provide for pharmacokinetics of cocaine close to those produced by smoking of crack (Jeffcoat et al., 1989; Isenschmit et al., 1992; Cone et al., 1994; Cone, 1995), which is the most prevalent form of cocaine administration today (Hays et al., 1999; Richardson et al., 1999; Scher et al., 2000); and 2) to replicate the average pattern of cocaine abuse by female drug addicts (Gossop et al., 1994; Richardson and Day, 1998).

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals. Eight healthy time-pregnant rhesus monkeys (Macaca mulatta), 5 to 7 years of age, were purchased from the commercial sources. The monkeys arrived at the University of Maryland Animal Facilities between pregnancy days 25 and 30 (E25-E30). Throughout the study, the animals were housed in individual cages and were fed High Protein Monkey Chow (Ralston Purina Co., St. Louis, MO), with fresh fruits given twice a day. The water was available at libitum. The day following their arrival, the animals were sedated with ketamine and custom-fitted in nylon vests to accustom them to this protective gear. Two of the monkeys were sedated again at E40, and prepared for chronic intravenous administration of cocaine as described in Mello et al. (1993a,b). This involved the surgical implantation of a silicone catheter in the jugular vein. The catheter was exited at the mid-scapular region under protection of the aforementioned nylon vests. Upon return of the animal to its cage, the catheter was connected to flexible stainless steel cables and fluid swivels, which allowed for a free movement within the cage. Cocaine hydrochloride (Research Technology Branch, National Institute of Drug Abuse, Rockville, MD) was administered beginning on E41 with five daily injections of 1 mg/kg each at 8:00 AM, 9:00 AM, 10:00 AM, 11:00 AM, and 12:00 PM. The drug was injected in 3 ml of sterile saline within a period of 40 to 50 s. The dosing of cocaine was based on the amount of this drug self-administered daily by an average pregnant crack cocaine user during the second trimester of pregnancy (Richardson and Day, 1998; Richardson et al., 1999) and corrected for the bioavailability of smoked cocaine base (Cone, 1995). The number of daily injections was equivalent to the average number of cocaine administrations reported for female drug addicts during pregnancy (Gossop et al., 1994; Richardson and Day, 1998). The interval between administrations was similar to that used in previous human models of cocaine abuse (Jufer et al., 1998). Also beginning on E41, two additional pregnant monkeys were treated with cocaine orally administered (in fruit treats) at a dose of 10 mg/kg, twice a day, at 8:00 AM and 8:00 PM. This regimen was the same as the one used in our previous studies of the consequences of prenatal cocaine exposure (Lidow, 1995; He et al., 1999; Lidow and Song, 2001; Lidow et al., 2001). The remaining four animals received no chronic cocaine treatment. They were used for the examination of the pharmacokinetics of single oral and intravenous doses of cocaine administered on the day of the blood collection (see below).

Blood Collection. Blood collection for analysis was performed on E91, 2 days after the surgery to allow the animals to recuperate and to eliminate the possible effects of anesthetic and analgesic drugs on cocaine pharmacokinetics. In the two animals receiving daily oral cocaine treatments blood collections were performed at 1, 5, 10, 20, 30, 60, 120, 240, 360, 480, and 720 min following the morning drug administration. The same blood collection times were used for the two animals receiving a single morning oral administration of 10 mg/kg cocaine and the two animals receiving a single morning intravenous injection of 1 mg/kg cocaine. For the two animals receiving multiple daily intravenous cocaine injections, blood collections were performed at 0.1, 1.1, 2.1, 2.5, 3.1, 3.5, 4.1, 4.5, 5.5, 5.5, 6.0, 8.0, and 12.0 h after the first injection of the day. During each sampling, 300 µl of maternal blood and 150 µl of fetal blood were collected in heparinized Vacutainer tubes containing NaF and acetic acid. The samples were centrifuged at 5000 rpm for 15 min at 4°C. The supernatant was collected and stored at 70°C.

Analysis of Concentrations of Cocaine and Benzoylecgonine. The samples were analyzed for cocaine and benzoylecgonine according to the methodology of Cone et al. (1994) and Wang et al. (1994), with slight modifications. Specimens were mixed with internal standard solutions [trideuterated analogs of cocaine and benzoylecgonine (Sigma Chemical Co., St. Louis, MO)], diluted with acetate buffer (pH 6), filtered through fritted 9RFV02F4P reservoirs (United Chemical Technologies Co., Bristol, PA), and extracted by solid-phase extraction using Clean Screen DAU, 200 mg-10-ml filtration columns (United Chemical Technologies Co., Bristol, PA). Cocaine and benzoylecgonine were eluted with freshly prepared solvent (methylene chloride/2-propanol/ammonium hydroxide, 80:20:2, v/v/v). The eluent was evaporated under nitrogen in a 40°C water bath and reconstituted in 20 µl of acetonitrile. The samples were then incubated for 30 min at 80°C with 20 µl of derivatized reagent [N,O-bis-(trimethylsilyl)trifluoroacetamine, containing1% trimethylchlorosilane]. Chromatography-mass spectrometry was performed using 1-µl aliquots of the derivatized extract. The analysis was performed on a Hewlett-Packard 5971 mass selective detector interfaced with a Hewlett-Packard 5890A gas chromatograph (Hewlett-Packard, Meriden, CT). The splitless injection mode with purge-off time of 0.5 min was used for all analyses. Ultrapure grade helium was used as the carrier gas at a flow rate of 1 ml/min. The initial oven temperature was 70°C held for 1 min, programmed to 220°C at 35°C/min held for 0.25 min, programmed to 250°C at 10°C/min and held for 3 min. The injection port and transfer temperature were maintained at 250 and 280°C, respectively. The mass-selective detector was operated in the selected-ion monitoring mode. The ions for each compound were monitored in the following elution order (quantitative ion indicated in parentheses): [²H₃]cocaine, m/z (185), 85; cocaine, m/z (182), 82, 303; and [²H₃]benzoylecgonine, m/z (243), 85; benzoylecgonine, m/z (240), 82, 361. Quantification of cocaine and benzoylecgonine was based upon ratios of peak areas to the corresponding deuterated internal standards. Duplicate matrix-matched calibration curves for both cocaine and benzoylecgonine were processed with each batch of specimens. Curves were constructed across the concentration range of 3 to 4000 ng/ml for both analytes. Control samples, containing the analytes at concentrations of 50, 250, and 1000 ng/ml, were also processed in duplicate with each run. The concentrations of cocaine and benzoylecgonine were calculated as micrograms per liter of plasma.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Single Intravenous Cocaine Injection. After a single intravenous injection of cocaine, the plasma levels of this drug in the pregnant rhesus monkeys reached maximum levels within 1 to 2 min (Fig. 1; Table 1 the fetus, the peak levels were reached 1 to 2 min later (Fig. 1; Table 1). The maximal fetal plasma levels of cocaine after a single intravenous injection were more than 5 times lower than the peak levels in the mothers (p = 0.005; Fig. 1; Table 1). The AUC of the fetuses was also more than 4 times smaller than the maternal AUC (p = 0.006; Table 1). The t_1/2, K_el, and MRT values in the fetuses were comparable to those in the mothers (p > 0.05; Table 1). In both mothers and fetuses, the levels of cocaine in blood declined to virtually undetectable levels within 8 h following the injection (Fig. 1).

View larger version (28K): [in this window] [in a new window]   Fig. 1.   Concentration of cocaine and benzoylecgonine in the mid-gestation pregnant rhesus monkeys and their fetuses after a single intravenous dose of cocaine (1 mg/kg), a single oral dose of cocaine (10 mg/kg), and the last dose of the 50-day-long chronic oral cocaine administration (10 mg/kg per administration). Each point represents a mean of n = 2 ± S.D.

Single p.o. Cocaine Administration. The maximal levels of cocaine in the circulation of pregnant monkeys after a single oral administration were reached within about an hour (Fig. 1; Table 1). Cocaine levels in plasma then declined with t_1/2 values comparable to those seen in the animals receiving a single intravenous injection of the drug (Table 1). As in the latter animals, the cocaine levels were below detection by 8 h following administration (Fig. 1). The bioavailability of orally administered cocaine was calculated as 25% (Table 1). In the fetuses of the monkeys receiving a single oral administration of cocaine, the peak levels of this drug were reached within 15 to 30 min after such levels were reached in the maternal circulation (Fig. 1; Table 1). These levels were nearly 5 times lower than those in the mothers (p = 0.004; Fig. 1; Table 1). The fetal AUC was also nearly 4 times smaller than the maternal AUC (p = 0.006; Table 1). There were no significant differences between mothers and fetuses in the values of t_1/2, K_el, and MRT of cocaine (p > 0.05; Table 1). These values were also similar to those in the animals receiving a single intravenous injection of the drug (p > 0.05; Table 1).

Chronic p.o. Cocaine Administration. In the circulation of pregnant monkeys, the peak levels and AUC of cocaine after an oral dose concluding a 50-day-long chronic treatment was only slightly larger that those observed after a single oral administration of this drug (p > 0.05; Fig. 1; Table 1). Also, as in the case of a single drug administration, the peak levels and daily AUC of cocaine for the chronically exposed fetuses were several magnitudes lower than for their mothers (p = 0.003 and 0.013 for peak levels and AUC, respectively; Fig. 1; Table 1). However, these parameters in the plasma of the fetuses chronically exposed to cocaine were nearly twice as large as in the fetuses subjected to only a single cocaine treatment (p = 0.039 and 0.047 for peak levels and AUC, respectively; Table 1).

Chronic Intravenous Cocaine Administration. The peak levels of cocaine in plasma of pregnant monkeys subjected to five daily intravenous injections of cocaine for over 50 days were more than twice the peak levels of cocaine in pregnant animals receiving a single injection of this drug (p = 0.042; Figs. 1 and 2; Tables 1 and 3). These levels were similar to the maximal levels of cocaine in the animals receiving chronic oral cocaine treatment (p = 0.086; Figs. 1 and 2; Tables 1 and 3). The AUC of cocaine in the intravenously chronically treated monkeys, however, was more than twice as large as that for the morning cocaine administration in the animals chronically receiving the drug by the oral route (p = 0.026; Figs. 1 and 2; Tables 1 and 3). This was because the multiple daily intravenous injections kept high plasma levels of cocaine 6 times longer than a single administration within the chronic oral treatment (Fig. 2). It should be remembered that in our studies cocaine was given orally twice daily. Consequently, the total daily AUC for the chronic oral treatment was much closer to that recorded for the chronic intravenous treatment.

View larger version (23K): [in this window] [in a new window]   Fig. 2.   Concentration of cocaine and benzoylecgonine in the mid-gestation pregnant rhesus monkeys and their fetuses on the 50th day of chronic treatment with five daily intravenous injections of cocaine, 1 mg/kg each. The arrows mark the injection times. Each point represents a mean of n = 2 ± S.D.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

This study supports the conclusion of previous reports (Binienda et al., 1993; Duhart et al., 1993; Saady et al., 1995) demonstrating a significant similarity between the pharmacokinetics of cocaine in nonhuman primates and human drug addicts. All the pharmacokinetic parameters calculated in the course of the present investigation were well within the range of these parameters obtained from human volunteers (Wilkinson et al., 1980; Barnett et al., 1981; Chow et al., 1985; Jeffcoat et al., 1989; Isenschmit et al., 1992; Cone et al., 1994; Cone, 1995). Even a relatively low bioavailability of the orally administered cocaine was close to that obtained in human studies of this route of drug administration (Wilkinson et al., 1980; Fattinger et al., 2000).

Our results also support the earlier observations of Binienda et al. (1993) that cocaine administered to pregnant rhesus monkeys can penetrate the placenta and enter the fetal circulation. Furthermore, both studies have demonstrated that the maximal levels of cocaine in the circulation of the fetuses are several magnitudes lower than the peak levels of this drug in their mothers. However, although we found the maternal/fetal ratio of peak cocaine levels after a single drug administration to be close to 1:5, this ratio was 1:7 in studies of Binienda et al. (1993). These differences may relate to the fact that, in contrast to the latter studies that examined cocaine pharmacokinetics in anesthetized near term animals, we used fully awake mid-term monkeys.

The comparison of the relationship between the maternal and fetal plasma levels of cocaine seen in the present study with those in the pregnant human drug abusers is rather difficult. The only human information in this regard comes from the post-mortem study of the pregnant victim of a car accident (Mittleman et al., 1989). In that case, the levels of cocaine in the fetal plasma were nine times lower than in the plasma of the mother. The authors of the report, however, speculate that such a low proportion of cocaine in the fetal circulation may be due to the fact that "the mother's death occurred so rapidly following the absorption of snorted drug that the pharmacokinetic compartment represented by the fetus did not reach equilibrium with the maternal blood" (Mittleman et al., 1989). The relationship observed in this study between the maximal levels of cocaine in the fetal and maternal circulation of rhesus monkeys was very similar to that reported for rodents and sheep (DeVane et al., 1989, 1991; Spear et al., 1989; Collins et al., 1999; Ma et al., 1999). This may indicate that this relationship is characteristic of all mammalian species.

As can be expected based on human studies (Jeffcoat et al., 1989; Cone et al., 1994; Jufer et al., 1998), the maximal levels of benzoylecgonine in a maternal plasma after the intravenous cocaine injection were much lower than those of cocaine, whereas the peak plasma levels of this metabolite after an oral cocaine administration were significantly higher than those of the parent chemical compound. The latter is due to the significant nonenzymatic hydrolysis of cocaine in the stomach and the nonenzymatic and enzymatic hydrolysis of this drug during the first pass metabolism (Cone et al., 1994; Sandberg et al., 1995; Jufer et al., 1998).

There is a long-standing controversy as to whether benzoylecgonine can cross the placental barrier. Although some investigators believe that the placenta is virtually impermeable to benzoylecgonine (Schama et al., 1998), others have demonstrated that this metabolite can cross the placenta in both humans and rodents, although to a much lower extent than cocaine (Simone et al., 1994; Sandberg et al., 1995; Morishima et al., 1997). Examinations of aborted human fetuses have shown significant plasma levels of benzoylecgonine (Apple and Roe, 1990; Klein et al., 1992). However, this may be a result of cocaine hydrolysis in the fetal circulation. The present study demonstrates the presence of benzoylecgonine in the fetal plasma after both intravenous and oral cocaine administration. We also found that, after an intravenous injection, the fetal peak levels of benzoylecgonine were close to the levels of cocaine. In contrast, the maximal levels of this metabolite were much higher than those of cocaine in the fetuses of the mothers receiving a single oral cocaine administration. This indicates that at least some of the benzoylecgonine in the latter fetuses came from the maternal circulation. It must be stressed that in all fetuses the plasma levels of benzoylecgonine were several times lower than in the maternal circulation, and even after chronic oral cocaine treatment they never reached levels capable of inducing cytotoxicity (Lin and Leskawa, 1994).

The final objective of the present study was to compare the cocaine and benzoylecgonine exposures in the fetuses of the mothers receiving chronic oral cocaine treatment (used in our previous analysis of the effects of this drug on cerebral cortical development; Lidow, 1995; He et al., 1999; Lidow and Song, 2001; Lidow et al., 2001) and the fetuses of the mothers subjected to a more realistic treatment by multiple daily intravenous cocaine injections. We found that in both cases the levels of maternal cocaine were within the range of those reported for human drug addicts (Jatlow, 1988; Jufer et al., 1998). We also found that chronic treatment with two daily administrations of 10 mg/kg cocaine produced somewhat more significant overall exposure of the fetuses to both cocaine and benzoylecgonine than the treatment with multiple daily intravenous injections of 1 mg/kg of the drug. On the other hand, the latter injections resulted in a much longer exposure of the fetuses to near maximal levels of cocaine.