Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Treatment of substance abuse is a particularly difficult problem, because sites of action are in the CNS and a wide range of structurally related drugs are often abused. For those drugs that act primarily at a single, specific CNS site, it is sometimes possible to use an agonist or antagonist to treat addiction or drug overdose. However, antagonists are not available for most drugs, and the medical use of an agonist or antagonist can potentially disrupt normal CNS homeostasis (Heishman et al., 1989; Howlett and Rees, 1986). To add to the problem, many drugs of abuse produce their effects through multiple mechanisms and binding sites in the CNS. This makes it even more difficult to develop antagonists.

PCP and other arylcyclohexylamines are examples of structurally related drugs of abuse that activate multiple systems in the brain and have no known antagonists. For example, PCP is a noncompetitive antagonist at the NMDA receptor complex (Lodge and Anis, 1982), but it also acts on the dopaminergic (Vignon et al., 1982; Chaudieu et al., 1989) and serotoninergic systems (Hori et al., 1996). Whereas some arylcylclohexylamines produce their effects primarily through the NMDA receptor complex (Vignon et al., 1983; Johnson et al., 1988) or the dopamine transporter (Vignon et al., 1988; Maurice et al., 1991), other members of this drug class produce their effects through both of these binding sites, as well as other sites in the CNS (e.g., PCP). Consequently, it would be advantageous to have a medical strategy for treating the multiple effects of these drugs. One such strategy is to develop an antibody-based medication capable of recognizing the common structural features of this broad class of drugs. This medical strategy targets the drugs rather than the receptors.

Previous studies have shown that an anti-PCP monoclonal Fab fragment (the antigen binding fragment of IgG) can remove PCP from the brain (Valentine and Owens, 1996) and can reverse PCP-induced locomotor activity in rats at low doses of PCP (Valentine et al., 1996). In these previous studies, however, the quantity of available antibody was a limiting factor. Consequently, we could not examine the efficacy of the therapy at high PCP doses. In addition, we were unable to test a full range of Fab doses to understand better the pharmacodynamics of less than completely neutralizing doses. A treatment for PCP abuse in humans is needed, because intoxicated individuals can endanger themselves as well as other persons during periods of drug-induced violent or irrational behavior (McCarron et al., 1981a). PCP use can also cause psychosis, catatonia, an acute brain syndrome or even death (McCarron et al., 1981a; McCarron et al., 1981b; Miller et al., 1988). Furthermore, the behavioral effects of PCP intoxication resemble a schizophrenic psychosis (Rosenbaum et al., 1959; Jentsch et al., 1997) and can last for several days.

The current studies were conducted in a rat model of human behavioral toxicity that allowed extensive examination of the pharmacological properties of an antibody-based therapy for PCP-like drugs. In addition, improvements in our ability to produce large quantities of antibody permitted us to perform more thorough pharmacodynamic studies. The in vitro experiments included structure-activity studies of the antibody using a large series of arylcyclohexylamines and other drugs. The in vivo experiments included determination of the minimal effective Fab dose, the ED₅₀ value for anti-PCP Fab inhibition of PCP-induced locomotor effects and the effectiveness of the therapy against several potent arylcyclohexylamines. Finally, these studies also suggest that a single carefully selected mAb can be used to reverse the effects of several structurally related drugs of abuse.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Drugs and reagents. [³H]PCP (50 Ci/mmol) was purchased from New England Nuclear (Boston, MA). The Upjohn Co. (Kalamazoo, MI) generously donated Dexoxadrol. Dizocilpine [(+)MK-801 hydrogen maleate] was obtained from Research Biochemicals International (Natick, MA). All other drugs were obtained from the National Institute on Drug Abuse (Rockville, MD). All drug concentrations were calculated as the free base. All reagents were obtained from Sigma Chemical Co. (St. Louis, MO) unless otherwise stated.

Large-scale production and purification of mAb Fab fragments. The procedure for production of the monoclonal anti-PCP IgG (k light chain) from the hybridoma cell line mAb-6B5 in a Cell-Pharm System II hollow fiber bioreactor (Unisyn Technologies, Inc., Hopkinton, MA) is described elsewhere (Valentine et al., 1996). The IgG was purified at room temperature from the bioreactor tissue culture media (Dulbecco's modified Eagle's medium, Mediatech, Herndon, VA) supplemented with 10% fetal calf serum (Hyclone Laboratories Inc., Logan, UT) by cation exchange chromatography using an INdEX-100 column (Pharmacia Biotech, Piscataway, NJ) packed with 1.0 liter of SP Sepharose Big Beads (Pharmacia Biotech). The unpurified antibody in bioreactor tissue culture media (typically 20-30 g of IgG in 20 liters of tissue culture media) was diluted 1:5 (v/v) in deionized H₂O, and the pH was adjusted to 6.0 using concentrated HCl. This solution (approximately 100 liters) was then pumped through the cation exchange media at a flow rate of about 400 ml/min. An in-line absorbance detector (Pharmacia Biotech) was connected downstream of the column to monitor the absorbance of the eluent at 280 nm. After the application of the IgG, the column was washed with 50 mM 2-[N-morpholino]ethanesulfonic acid (Mes) buffer (pH 6.0) until the absorbance returned to base line. The IgG was then eluted from the column using 0.15 M NaCl in 50 mM Mes (pH 6.0) buffer at a flow rate of approximately 400 ml/min. The final concentration of anti-PCP IgG was about 7 mg/ml in a volume of 3 to 4 liters.

Characterization of the anti-PCP mAb. The ligand binding studies of the mAb IgG (in tissue culture media) were conducted in a [³H]PCP RIA similar to the method of Owens et al. (1988). A 100-µl aliquot of [³H]PCP (30,000-40,000 decays per min) in RIA buffer (50 mM Tris adjusted to pH 7.6, 0.15 M NaCl, 0.1% BSA, 0.2% NaN₃) was added to duplicate 50 × 14 mm polypropylene tubes, followed by the appropriate concentration of test ligand. Then a 100-µl aliquot of diluted mAb in tissue culture media was added. The mAb had been titered in RIA buffer to a concentration that would bind 15% to 20% of the [³H]PCP. To nonspecific binding tubes, 100 µl of tissue culture media was added (at the same dilution as the mAb-containing tubes). The reaction mixture was incubated overnight at 4°C to 8°C after vortex mixing. Next, 1 ml of cold RIA buffer containing 5% goat anti-mouse IgG serum (v/v; Pel-Freez Biologicals, Rogers, AR) and 5% polyethylene glycol 8000 (J.T. Baker Chemical Co., Phillipsburg, NJ) was added. The tubes were incubated for 15 min in the cold, followed by centrifugation at 4°C to 8°C for 15 min at 2000 × g to precipitate the antibody-bound radioactivity. The supernatant fluid was aspirated, and the pellet was resuspended in the same tube using 2 ml of scintillation fluid (Liquiscint, National Diagnostics, Manville, NJ). The tube was placed in a 7-ml scintillation vial, and the concentration of [³H]PCP was determined by liquid scintillation spectrometry.

Animals. All animal experiments were carried out with the approval of the Institutional Animal Care and Use Committee of the University of Arkansas for Medical Sciences and were in accordance with the Guide for the Care and Use of Laboratory Animals as adopted and promulgated by the National Institutes of Health. Adult male Sprague-Dawley rats were purchased with a single cannula (silastic 0.51 mm I.D. × 0.94 mm O.D.) implanted in their right jugular vein (Hilltop Lab Animals, Scottsdale, PA). For shipping purposes, the cannula was enclosed in a s.c. pocket. The day after arrival from the vendor, the cannulas were externalized, flushed with sterile saline and filled with heparinized saline (500 U/ml). The cannulas were flushed and filled 3 to 4 times a week to maintain patency. All drug and Fab treatments were given i.v. via the cannula. Each animal's food intake was monitored to maintain a constant weight of approximately 300 g. The animals were allowed at least 2 weeks to acclimate to their new environment before treatments. During the acclimation period, we thoroughly conditioned the animals to the experimental environment by handling them and placing them in the testing chambers on multiple occasions.

Protocols for behavioral experiments. Each experiment was conducted between 0700 and 1330 h (during the light phase of the light-dark cycle). Experiments were carried out in open-top polypropylene chambers (60 × 45 × 40 cm; l × w × h). Bedding of gray clay gravel was used to provide a nonreflective, contrasting background and a neutral odor. The experiments were recorded with a closed-circuit television camera located above the chambers. The camera was connected to a monitor and a S-VHS recorder. Two animals could be filmed in two separate chambers (one animal in each chamber) at the same time. For experiments, the animals were placed in the chambers 60 min before saline or drug administration. Saline or drug was administered at time 0 (zero) for all groups. Saline or anti-PCP Fab treatment was given 30 min after drug administration, except for the first series of experiments, which were designed to determine the dose-response relationship for PCP-induced locomotor effects.

Behavioral analysis. Analysis of the videotaped experiments was conducted off-line after completion of each experiment using the EthoVision system (Noldus Information Technology, Inc., Sterling, VA). The EthoVision software (versions 1.7-1.8) was run on a personal computer equipped with frame-grabber technology (TARGA+, Release 4.0, Truevision, Inc., Indianapolis, IN) and connected to a S-VHS recorder. A video image-sampling rate of 3.0 samples/s resulted in 48,600 samples during the 4.5-h period of data collection. For each sample, the x- and y-coordinates of the animal in the chamber and the size of the animal's image were recorded. Although several behavioral parameters (such as rearing, meandering, sinuosity and angular velocity) were considered for use during preliminary behavioral studies, the parameters distance traveled and total movement were found to be the most useful measures for arylcyclohexylamine (PCP, TCP and PCE) intoxication in rats. In the experiments with methamphetamine, animal rearing was also used as an indicator of drug effects. The results of all behavioral analyses were reported in 2-min cumulative intervals.

Data and statistical analysis. For each of the drugs in table 1, a IC₅₀ value was calculated from a 4 to 5-point standard curve after a logit-log transformation of the RIA data (Rodbard, 1974). Each IC₅₀ value was reported as the average value from two separate determinations. This average IC₅₀ value for each drug was used to calculate its relative potency to PCP.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Large-scale production, purification and characterization of the anti-PCP mAb Fab fragments. The overall recovery of anti-PCP Fab binding sites from the IgG (two Fab binding sites per IgG molecule) in tissue culture media was approximately 80%. The purity of the anti-PCP Fab was approximately 90% as determined by SDS-PAGE and densitometry. In a previous study (McClurkan et al., 1993), it was determined that the anti-PCP Fab K_d value (1.8 nM) is unaffected by the purification process and is essentially the same K_d value as for the native IgG (1.3 nM). Furthermore, we have found no decrease in binding activity or solubility after long-term storage of the IgG or Fab at 80°C.

Dose-response relationship of PCP-induced locomotor effects. These experiments were conducted to determine whether the behavioral parameters we had chosen (distance traveled and movement) were dose-dependent for PCP. There was a dose-dependent increase in PCP-induced locomotor effects as measured by these parameters (figs. 1 and 2). For this series of experiments, the AUC was calculated from 6 min (the time when the animals were returned to the chambers after saline or drug administration) until 226 min (the longest duration of effects for an animal in this experimental group). The duration of PCP-induced locomotor effects lasted 50.5 ± 8.1 min after the 1 mg/kg dose, 115.3 ± 13.3 min after the 3 mg/kg dose and 184.0 ± 34.0 min after the 6 mg/kg dose. After i.v. administration of the 1 mg/kg PCP dose, the animals became immediately hyperactive. In contrast, the animals were completely immobile immediately after the 3 and 6 mg/kg doses of PCP, except for tremors and head weaving. In a preliminary study, it was determined that this period of immobility was characterized by the inability of the animals to respond to a painful stimulus (results not shown). The painful stimulus was an ear pinch administered every 2 min for 1 h. This informal experiment suggested that the animals were under the anesthetic effects of PCP. This immobility stage lasted 18.7 ± 10.3 min for the 3 mg/kg PCP dose and 36.3 ± 16.3 min for the 6 mg/kg PCP dose. After emerging from the immobility stage, the animals were severely ataxic. As the ataxia decreased, the animals became extremely hyperactive. Eventually, the hyperactivity decreased until the activity returned to base line.

View larger version (37K): [in this window] [in a new window]   Fig. 1.   Animal movement vs. time after saline or one of three doses of PCP in a representative animal. These experiments were designed to determine the dose-response relationship of PCP-induced locomotor effects. Movement (the time the animal spent moving during each 2-min interval) was recorded continuously. The animal was placed in the chamber 1 h before drug administration (60 min). At time zero (0), the animal received an i.v. bolus dose of saline, of 1 mg/kg PCP, of 3 mg/kg PCP or of 6 mg/kg PCP. The time when the animal was removed from the chamber for saline or drug administration (0-6 min) was not plotted. The arrow indicates the time of saline or drug administration. For ease of comparison between doses, the AUC6226 (the longest duration of effects in one of the animals at 6 mg/kg) is shaded. The AUC6226 for each animal after each treatment was used to calculate the PCP dose-response curves shown in figure 2. The behavioral parameter "distance traveled" (not shown) produced similar results.

View larger version (15K): [in this window] [in a new window]   Fig. 2.   PCP dose-response curves for total distance traveled (top panel), total movement (middle panel), and duration of effects (bottom panel). These experiments provided the preliminary PCP dose-response data for the validation of our behavioral model of arylcyclohexylamine-induced effects. Values are mean ± S.D. of the AUC6226 for total distance traveled and total movement (see fig. 1). Duration of effects (mean ± S.D.) was calculated on the basis of the time required for the movement to return to base-line levels ("Materials and Methods"). The animals received four treatments in a mixed-sequence, repeated-measures design (n = 4, except for the 3 mg/kg dose, where n = 3). The control (saline) results are represented as the hollow circle, and the PCP treatments are represented as solid circles. A logistical dose-response curve (solid line) was fit to each data set. A one-way, repeated-measures ANOVA followed by a Student-Newman-Keuls test was used to evaluate the difference between treatments (*P < .05 compared with saline control). All doses of PCP were also different from each other.

Dose-response relationship for the anti-PCP Fab inhibition of PCP-induced locomotor effects. The goal of these experiments was to determine the dose of anti-PCP Fab required to reverse the effects of PCP-induced locomotor effects. For this series of experiments, the AUC was measured from 36 min (the time when the animals were returned to the chambers after saline or anti-PCP Fab treatment) until 150 min (the longest PCP-induced effect in the animals). The anti-PCP Fab decreased total distance traveled, total movement and duration of effects in a dose-dependent manner (fig. 3). The minimal effective dose for the anti-PCP Fab to reverse the measured behavioral effects was 0.18 mol-eq (P < .05).

View larger version (23K): [in this window] [in a new window]   Fig. 3.   Dose-response relationship for the anti-PCP Fab inhibition of PCP-induced locomotor effects. The animals (n = 4) received seven treatments in a mixed-sequence, repeated-measures design. PCP was administered at 3 mg/kg in all cases, but the anti-PCP Fab doses were varied as shown on the x-axis. The treatments are represented by the following symbols: saline followed by saline treatment (), PCP followed by saline treatment () and PCP followed by anti-PCP Fab (). All values were calculated as a percentage of each animal's response to the PCP followed by saline treatment (i.e., 100% response). Values for the total distance traveled (top panel) and total movement (middle panel) are plotted as the mean ± S.D. of the AUC36150. Values for duration of effects (bottom panel, mean ± S.D.) were calculated on the basis of the time needed for animal movement to return to base-line values. A one-way, repeated-measures ANOVA followed by a Student-Newman-Keuls test was used to evaluate the difference between treatments (*P < .05 compared with PCP followed by saline treatment). Curves were fit to the data using a sigmoidal inhibitory pharmacodynamic model. The Emax, ED50, and the  values (upper right corner of each plot) were calculated from the curves.

Effectiveness of the anti-PCP Fab against PCP, TCP and PCE. These experiments were used to determine whether the anti-PCP Fab was an effective treatment for several potent, prototypic members of the arylcyclohexylamine drug class (tables 1 and 2). For these experiments, the AUC was calculated from 36 min until 194 min. As expected, the effects from the three arylcyclohexylamines followed the known pharmacological potencies of these drugs (PCP < TCP < PCE) (table 1; fig. 5). The anti-PCP Fab treatment markedly decreased the locomotor activity induced by PCP, TCP and PCE to levels that were not statistically different from base-line control treatments (saline administration followed by saline treatment) (figs. 4 and 5). The duration of the locomotor effects induced by PCP, TCP and PCE were dramatically decreased by the anti-PCP Fab treatment (table 3). As previously described for PCP at 3 and 6 mg/kg, a period of immobility followed TCP (3 mg/kg) and PCE (3 mg/kg) administration. This immobility stage after administration of TCP and PCE lasted approximately 30 to 40 min. As figure 4 shows, there was no animal response during the initial periods after drug administration. A precise calculation of the duration of the immobility stage was not possible in these experiments, because the animals were handled during their removal from the chambers for saline or Fab treatments 30 to 36 min after drug administration. When the animals returned to base-line levels of activity after anti-PCP Fab treatment, they quickly exhibited normal behaviors such as calmly sitting in the chamber, grooming and sleeping.

View larger version (41K): [in this window] [in a new window]   Fig. 4.   Treatment of PCP-, TCP- and PCE-induced behavioral toxicity with anti-PCP Fab in a representative animal. The time the animal spent moving during each 2-min interval was recorded continuously. The saline followed by saline treatment is not shown, but it was similar to the saline treatment shown in figure 1. Each animal was placed in the chamber 1 h before drug administration. At time zero, the animal received saline or a 3 mg/kg dose of PCP, of TCP or of PCE (arrow). The saline (left panels) or 1.0 mol-eq dose of anti-PCP Fab (right panel) treatments were given at 30 min (arrowhead). The times when the animal was removed from the chamber for the saline or drug administration (0-6 min) and for the saline or anti-PCP Fab treatment (30-36 min) are not plotted. For ease of comparison, the AUC36194 (from the time the animal was placed back in the chamber after Fab treatment to the longest duration of effects in one of the animals in these experiments) is shaded. The behavioral parameter "distance traveled" produced similar results (not shown).

View larger version (36K): [in this window] [in a new window]   Fig. 5.   The effectiveness of the anti-PCP Fab against PCP, TCP and PCE as measured by total distance traveled and total movement. The rats (n = 4) received seven treatments in a mixed-sequence, repeated-measures design (see fig. 4 for a representative animal). The animals received saline or a 3 mg/kg dose of PCP, of TCP or of PCE followed 30 min later by saline (hatched bars) or a 1.0 mol-eq dose of anti-PCP Fab (solid bars). Values are mean ± S.D. of the AUC36194. A one-way, repeated-measures ANOVA followed by a Student-Newman-Keuls test was used to evaluate the difference between treatments (*P < .05 compared with saline control followed by saline treatment, left hatched bar). All drug followed by anti-PCP Fab treatments were different from drug followed by saline treatments. The results of the Fab treatments of drug effects (solid bars) were not statistically different from those of the saline followed by saline treatment (left hatched bar).

Effectiveness of the anti-PCP Fab against methamphetamine. These experiments were conducted to determine the selectivity of the anti-PCP Fab therapy. Methamphetamine was chosen as a control drug, because it belongs to a drug class different from PCP but produces some of the same locomotor effects. The dose of methamphetamine (1 mg/kg) used for these studies produced a period of hyperactivity that was similar in duration to the period of hyperactivity produced by a 3 mg/kg dose of PCP (results not shown). Animal rearing was used as an additional behavioral parameter for this experimental group, because in preliminary experiments it was determined to be a dose-dependent effect of methamphetamine. However, PCP did not produce an increase in rearing behavior (results not shown), as judged by comparing the rearing behavior in animals administered PCP (followed by a saline treatment) and in animals administered saline (followed by a saline treatment).

View larger version (40K): [in this window] [in a new window]   Fig. 6.   The effectiveness of the anti-PCP Fab against (+)methamphetamine as measured by total distance traveled, total movement and rearing. The rats (n = 3) received four treatments in a mixed-sequence, repeated-measures design. The animals received methamphetamine (1 mg/kg) followed 30 min later by saline (hatched bars) or a 1.0 mol-eq dose of anti-PCP Fab (solid bars). Values are mean ± S.D. of the AUC36184. A paired t test was used to evaluate the difference between treatments (*P < .05).

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

A major goal of these studies was to determine whether a mAb could be developed for treating the CNS toxicity due to a chemical class of structurally related drugs, instead of a single drug. To accomplish this goal, we needed to generate a mAb that recognized the pharmacologically active features needed for binding of arylcyclohexylamines to CNS sites of action. For most arylcyclohexylamines, the primary site of action is the PCP recognition site, which is in the ion channel associated with the NMDA receptor complex (Lodge and Anis, 1982; Vignon et al., 1983; Johnson et al., 1988).

To determine whether this mAb recognized the most pharmacologically potent forms of arylcyclohexylamines, we studied the relative potencies of binding of PCP and other arylcyclohexylamines in a [³H]PCP RIA. In general, the mAb recognized the most pharmacologically potent forms of arylcyclohexylamines (PCP, TCP, PCE, PHP, THP and PCHP) (table 1). However, the actual antibody affinity constants (K_a or 1/K_d values) for these compounds were significantly higher (see representative values for PCP, TCP and PCE in table 2) than the actual affinity constant for the CNS receptor. We considered the higher-affinity binding a necessity because the mAb must redistribute these drugs from the CNS. In addition, the mAb showed virtually no recognition for other drugs. These in vitro data suggested that the antibody would be an effective antagonist for most of the potent arylcyclohexylamines, many of which are already Schedule I drugs with a high abuse potential. The one exception was ketamine, which was not recognized by the antibody. This lack of recognition was presumably due to interference by the chlorine molecule (at the ortho position of the aromatic ring). The other arylcyclohexylamines used in this study do not contain chlorine. Ketamine, a dissociative anesthetic agent, is the only arylcyclohexylamine with an approved medical use.

The first series of behavioral experiments determined the dose-response relationship for PCP-induced locomotor effects in a rat model of human behavioral toxicity (figs. 1 and 2). These measures were used because behavioral problems account for some of the major medical problems associated with PCP toxicity in humans (McCarron et al., 1981a; McCarron et al., 1981b) and because PCP effects could be monitored in freely moving animals in a noninvasive manner (Valentine et al., 1996; Sams-Dodd, 1995). After considering these data, we chose a PCP dose of 3 mg/kg for use in all other studies because it produced profound effects with sufficient duration for our purposes. Furthermore, on the basis of the actual PCP serum concentrations in emergency room patients (Walberg et al., 1983) and the average pharmacokinetics values for PCP in humans (Cook et al., 1982a; Cook et al., 1982b), we calculated that 99% of these patients had less than 3 mg/kg in their body. Therefore, a 3 mg/kg dose in the rat provided a good model of the maximal dose of PCP found in patients in an emergency room situation.

To understand better the pharmacological principles of the therapy, we conducted a full dose-response study of anti-PCP Fab reversal of CNS-mediated behavioral effects. We also determined the minimal effective dose of Fab and the ED₅₀ value for reversal of effects. As far as we know, this is the first reported full dose-response study of the pharmacodynamic relationship of antibody and drug interactions. The highest dose of anti-PCP Fab administered in these experiments was 1.0 mol-eq to the PCP dose, because in a previous study we showed that a 3.0 mol-eq dose of anti-PCP Fab was no more effective than a 1.0 mol-eq dose (Valentine et al., 1996). The minimal effective dose for the anti-PCP Fab inhibition of PCP-induced locomotor effects ranged from 0.1 to 0.18 mol-eq (fig. 3). The ED₅₀ values for the reversal of the behavioral effects ranged from 0.32 to 0.51 mol-eq. These data suggest that reductions in PCP-induced responses are directly related to the dose of anti-PCP Fab, and significant changes in PCP-induced behavioral effects can be achieved with less than 0.2 mol-eq of Fab. It is important to point out that this mAb Fab follows many of the established dose-response relationships for pharmacological antagonists.

To determine whether our anti-PCP Fab would be effective against a broad range of arylcyclohexylamines, we studied three potent, prototypic ligands (PCP, TCP and PCE). The anti-PCP Fab was equally effective at reversing the locomotor effects of PCP, TCP and PCE (table 3 and figs. 4 and 5), although there were significant differences between receptor and antibody affinities for the drugs (table 2). Consequently, both in vitro and in vivo data demonstrated that the anti-PCP Fab was an effective therapy for toxicity caused by PCP and other related arylcyclohexylamines.

In this and a previous study, we have examined aspects of the in vivo selectivity of the anti-PCP Fab. Although dizocilpine is a more potent NMDA antagonist than PCP, it is structurally unrelated to PCP, and the anti-PCP Fab is not effective at inhibiting dizocilpine-induced locomotor activity (Valentine et al., 1996). In the same study (Valentine et al., 1996), we found that a nonspecific Fab prepared from polyclonal human IgG had no effect on PCP-induced locomotor activity. In the current study, methamphetamine was chosen as an additional control, because it induces locomotor activity but is structurally unrelated to PCP. The antibody did not show a significant cross-reactivity with methamphetamine in the in vitro studies of the mAb specificity (table 1). Furthermore, the anti-PCP Fab did not have a statistically significant effect on methamphetamine-induced behavior (fig. 6). Nevertheless, the anti-PCP Fab appeared to produce a small decrease in methamphetamine-induced effects. Although we do not know the reason for these effects, the increased protein load resulting from administration of the anti-PCP Fab may have produced some nonspecific pharmacokinetic changes. However, the overall effects were negligible compared with the anti-PCP Fab's reversal of the effects of PCP-like drugs (fig. 5).

We think these studies represent a logical step in our attempts to scale up the development of large-scale production and testing of antibody-based therapy for drug abuse. However, there are advantages and disadvantages that need to be considered for the optimal use of immunotherapeutic agents in humans. On the basis of our calculations, the quantities of antidrug antibodies that will be needed for this type of therapy are significantly greater than for other medical applications of antibody-based therapy, such as radioimaging of tumors and delivery of chemotherapeutic agents (Goldenberg, 1993). The biotechnology for producing large quantities of antibodies (such as the bioreactors used in these studies) continues to improve, and we think that in the near future it will be cost-effective to test this type of therapy in humans. Furthermore, the use of Fab fragments (as in the current study) is one way to reduce or prevent an antigenic response. In fact, in patients who receive ovine antidigoxin Fab (Digibind), allergic reactions to the antibody fragments are rare (Hickey et al., 1991; Kirkpatrick, 1991).

Another concern for using antibody-based therapy is the generation of antireceptor antibodies. Our mAb was generated against a hapten that was previously shown to mimic the binding properties of arylcyclohexylamines to the PCP recognition site in the NMDA receptor complex (Owens et al., 1988). With repeated or long-term administration of this type of antibody, PCP-like antiparatype antibodies could be generated. If these antiparatype antibodies were to bind to PCP binding sites in the CNS, they could potentially produce drug-like effects. To address this point, we have generated anti-idiotype and antiparatype antibodies against this mAb. However, we have not been able to inhibit [³H]PCP binding to the PCP receptor with these antibodies (unpublished observations, S.M. Owens). Nevertheless, even if PCP-like antiparatypic antibodies were generated in vivo, it is unlikely that these antibodies could cross the blood-brain barrier in sufficient quantities to produce effects.

There are significant potential advantages to developing antibody-based medications for treating the medical problems associated with drug abuse. For instance, it is a common practice for chemists in illicit drug laboratories to modify the chemical structures of popular drugs of abuse to avoid detection and to increase the potency of their drugs (e.g., fentanyl-like compounds). These clandestine drugs have been termed designer drugs. A carefully chosen class-selective mAb would likely be an effective treatment for these new drugs because, by analogy, it is a "designer mAb." Finally, it would be too costly to develop (and for hospitals to purchase) a highly specific antibody-based medication for each individual drug of abuse. Another advantage is that antibodies could potentially be a more effective therapy than a conventional receptor antagonist. This is because a receptor antagonist typically reverses the effects at only one site of action, whereas drugs like PCP have multiple sites of action within the CNS. By significantly increasing the protein binding in the vascular compartment and lowering the volume of distribution of the drug (Owens and Mayersohn, 1986; Valentine et al., 1994; Valentine and Owens, 1996), antibodies act as "pharmacokinetic antagonists" to remove the drug rapidly from the CNS. In other words, the antibodies antagonize the effects of the drug by completely altering the pharmacokinetic properties of the drug. Furthermore, because antibodies do not appear to cross the blood-brain barrier under normal conditions, and because they are usually highly selective for the drug or drug class, antibody therapy should not disrupt normal CNS homeostasis.

In conclusion, these studies demonstrated that a monoclonal Fab was extremely effective at reversing the CNS-mediated behavioral toxicity associated with a major class of abused drugs. These data also indicated that the underlying mechanism for Fab reversal of drug effects follows a predictable pharmacological dose-response relationship. Because PCP pharmacokinetics in humans follows the principles of a first-order process (Cook et al., 1982a; Cook et al., 1982b), these data suggest that the dose of Fab needed for reversal of drug effects could be accurately calculated on the basis of the PCP plasma concentration in patients. Finally, the pharmacological and immunological concepts in these studies could have broad applicability for use in treating adverse effects due to other drug classes.