Immunotherapy against drugs of abuse is being studied as an alternative treatment option in addiction medicine and is based on antibodies sequestering the drug in the bloodstream and blocking its entry into the brain. Producing an efficient vaccine against heroin has been considered particularly challenging because of the rapid metabolism of heroin to multiple psychoactive molecules. We have previously reported that heroin’s first metabolite, 6-monoacetylmorphine (6-MAM), is the predominant mediator for heroin’s acute behavioral effects and that heroin is metabolized to 6-MAM primarily prior to brain entry. On this basis, we hypothesized that antibody sequestration of 6-MAM is sufficient to impair heroin-induced effects and therefore examined the effects of a monoclonal antibody (mAb) specific for 6-MAM. In vitro experiments in human and rat blood revealed that the antibody was able to bind 6-MAM and block the metabolism to morphine almost completely, whereas the conversion of heroin to 6-MAM remained unaffected. Mice pretreated with the mAb toward 6-MAM displayed a reduction in heroin-induced locomotor activity that corresponded closely to the reduction in brain 6-MAM levels. Intraperitoneal and intravenous administration of the anti–6-MAM mAb gave equivalent protection against heroin effects, and the mAb was estimated to have a functional half-life of 8 to 9 days in mice. Our study implies that an antibody against 6-MAM is effective in counteracting heroin effects.
The pharmacotherapies currently in use for the treatment of heroin abuse are agonists (methadone, buprenorphine) and antagonists (naltrexone, naloxone) to brain opioid receptors that either mimic or block the effects of heroin. The use of opioid maintenance therapy, including methadone and buprenorphine, has been successful in many aspects; however, these drugs are addictive and can contribute to overdose deaths (Simonsen et al., 2011; Hakkinen et al., 2012; Bernard et al., 2013). Concern may also be raised toward the reported negative effects of methadone on cognitive functions (Verdejo et al., 2005; Prosser et al., 2006, 2009; Andersen et al., 2011, 2012). An alternative approach with minor risk for central side effects is immunotherapy against drugs of abuse, which is based on antibodies sequestrating the drug in the bloodstream and blocking its entry to the brain, which can be achieved either by administration of preformed monoclonal antibodies (mAbs, passive immunization) or by using the patient’s own immune system to generate an antibody response (active immunization) (Kosten and Owens, 2005; Peterson and Owens, 2009; Shen et al., 2012).
Upon administration, heroin displays low brain concentration and low affinity for µ-opioid receptors and is therefore presumed to be a prodrug mainly acting through its metabolites (Inturrisi et al., 1983; Selley et al., 2001; Andersen et al., 2009; Gottas et al., 2013). In rodents and humans, heroin is rapidly metabolized by sequential deacetylation to 6-monoacetylmorphine (6-MAM) and morphine (Rook et al., 2006a; Andersen et al., 2009; Gottas et al., 2013), mainly by esterase enzymes (Owen and Nakatsu, 1983; Salmon et al., 1999). In humans, morphine is further transformed by glucuronidation to pharmacologically active morphine-6-glucuronide (M6G) and inactive morphine-3-glucuronide (M3G) (Fig. 1A) (Glare and Walsh, 1991; Milne et al., 1996; Rook et al., 2006b), whereas M3G is normally the only morphine glucuronide found in rodents (Zuccaro et al., 1997; Handal et al., 2002). For decades, morphine was considered the metabolite responsible for heroin’s pharmacological effects (Way et al., 1965); however, in the 1980s, it was suggested that 6-MAM also could be of importance (Umans and Inturrisi, 1981; Inturrisi et al., 1983). Previous studies in our laboratory demonstrated that the immediate heroin response in mice is mediated by 6-MAM (Andersen et al., 2009) and that heroin is metabolized to 6-MAM mainly in the periphery before its transfer to the brain (Boix et al., 2013). These findings are in contrast to the traditional assumption that heroin is particularly addictive because of its high lipophilicity, which allows it to easily pass the blood-brain barrier and to be metabolized to active metabolites in the brain (Oldendorf et al., 1972). The predominant role of 6-MAM for the immediate heroin response appears to be valid regardless whether heroin is injected subcutaneously or intravenously (Andersen et al., 2009; Gottas et al., 2013; Raleigh et al., 2013).
The proof-of-concept for vaccines against drugs of abuse was first reported in the 1970s, demonstrating both active and passive immunization strategies (Berkowitz and Spector, 1972; Berkowitz et al., 1974; Bonese et al., 1974; Killian et al., 1978). Since then, several active vaccines (using a morphine conjugate) have been developed with effects toward heroin- or morphine-induced behaviors (Anton and Leff, 2006; Li et al., 2011; Stowe et al., 2011; Pravetoni et al., 2012; Raleigh et al., 2013; Schlosburg et al., 2013). Passive immunotherapy toward opioids has received minor attention. This immunization strategy has the advantages of being independent of the interindividual differences in immune response, the protection is immediate, the duration of action is predictable, and mAb or antibody fragments can be designed to fit the therapeutic application (Peterson et al., 2006; Peterson and Owens, 2009).
Based on the increasing amount of evidence that 6-MAM is the metabolite responsible for the immediate rewarding effects of heroin, we hypothesized that sequestration of this specific metabolite would be sufficient to markedly impair heroin-induced effects. Because most of the previously reported active vaccines toward heroin display a broad specificity toward heroin and its metabolites, further characterization of the effects of specific 6-MAM sequestration appeared warranted. The mAb examined in this study was generated using a 6-MAM derivatized hapten containing a linker at the 6-MAM N-bridge (Fig. 1B) (Moghaddam et al., 2003). We compared the effects of the anti–6-MAM mAb on brain levels of heroin and metabolites with locomotor activation, which share some of the brain structures implicated in drug reward (Wise and Bozarth, 1987) and can therefore be used as a measure of opioid’s psychostimulatory effects (Morland et al., 1994). We also studied the duration of the protective effect of the antibody and compared different routes of administration.
Materials and Methods
Male C57BL/6J-Bom mice (7–8 weeks old, 20–25 g; Taconic, Ejby, Denmark) were housed six to eight per cage in the animal facility at the Norwegian Institute of Public Health (22 ± 1°C, 50% ± 10% humidity, light period 7:00 AM–7:00 PM). The animals arrived at least 5 days before the experiments. Commercial mouse pellets and water were available ad libitum. The experimental protocol of the study was approved by the Norwegian Animal Research Authority.
Rat blood was collected from sacrificed male Sprague Dawley rats (200–250 g; Taconic, Ejby, Denmark) and human blood from healthy adult volunteers. To inhibit clotting, heparin (14–17 IU/ml) was used. The experiments were initiated within 30 minutes after blood collection.
The anti–6-MAM mAb was developed and synthesized by Affitech Research AS (Oslo, Norway) using a biotin-polyethylene glycol-6-MAM conjugate (Fig. 1B). The mAb was based on a previously reported single chain fragment variable (scFv) antibody fragment (6-MAM-214; Moghaddam et al., 2003). For the purpose of this study, an antibody format with a longer half-life (IgG) was customized by using the Fv regions of 6-MAM-214. The anti–6-MAM mAb (human IgG1, λ-light chain) was produced in stable transfected Chinese hamster ovary cells, purified via protein A (MabSelect) before anion-exchange chromatography. Thereafter, it was dialyzed against the formulation buffer (phosphate buffer). The antibody was checked for endotoxins (≤0.5 EU/mg) and stored at 4°C.
Heroin-HCl (mol. wt. 421.91), 6-MAM-HCl (mol. wt. 408.21), M6G-HCl (mol. wt. 498.61), and M3G-HCl (mol. wt. 488.53) were purchased from Lipomed AG (Arlesheim, Switzerland). Morphine-HCl (mol. wt. 372) was from Norsk Medisinaldepot AS (Oslo, Norway), and naltrexone (mol. wt. 377.86) was delivered by Sigma-Aldrich (Oslo, Norway). Heroin and its metabolites were dissolved in 0.9% saline, and naltrexone was dissolved in methanol before dilution in 0.9% saline (final methanol concentration <0.01%). Anti–6-MAM mAb was injected intravenously (35 µl/10 g) or intraperitoneally (70 µl/10 g). Heroin and morphine were injected subcutaneously (100 µl/10 g).
Binding Test of the Anti–6-MAM mAb.
Binding of the anti–6-MAM mAb to different opioids was investigated by size-based separation of mAb-bound and free opiate fractions. Heroin, heroin metabolites, and naltrexone (0.1 µM) were incubated separately in 0.9% saline at room temperature with or without the anti–6-MAM mAb (final concentration 2 µM). After 15 minutes, with gentle mixing every 3rd minute, the samples were transferred to microcentrifuge tubes containing 30-K spin filters (Nanosep Centrifugal devices, Pall Life Science, Lund, Sweden) and centrifuged (14,000g, 10 minutes). The resulting filtrates were transferred without further sample preparation to autosampler vials and analyzed in duplicates by liquid chromatography-tandem mass spectrometry (LC-MS/MS) or ultraperformance liquid chromatography mass spectrometry (UPLC-MS) the same day. Degradation of heroin to metabolites during the 15 minutes assay was measured in each sample and found to be negligible.
Effects of the Anti–6-MAM mAb on Heroin Metabolism In Vitro.
Anti–6-MAM mAb (final concentration 0.5–6 µM) or saline was added to rat or human blood and preincubated in a water bath (37°C) for 10 to 15 minutes. At t = 0, heroin was added to each tube to a final concentration of 0.4 µM. After 90 minutes, ice-cold ammonium formate buffer (13 mM, pH 3.1; final concentration 5 mM) with sodium fluoride (final concentration 2 mg/ml) was added; the samples briefly mixed and immediately frozen in liquid nitrogen. The samples were stored at −80°C until LC-MS/MS analyses were performed the same day.
Sodium fluoride (final concentration 2 mg/ml) was added to both in vitro and ex vivo samples to inhibit esterase activity and thereby avoid postsampling degradation of heroin and 6-MAM (Brogan et al., 1992). Ice-cold acidic buffer was used to dilute blood samples and to homogenize brain tissue (see below) since heroin has been shown to be most stable at low temperatures and low pH (Barrett et al., 1992).
The acute locomotor activity was measured in chambers using Versamax optical animal activity monitoring system (AccuScan Instruments Inc., Columbus, OH) as previously described in detail (Andersen et al., 2009). In short, the mice were placed individually for 60 minutes in an activity chamber (habituation). After habituation, each mouse was gently removed to another room, injected with heroin (2.5 or 5 µmol/kg, equivalent to 1.05 or 2.10 mg/kg s.c.) or morphine (2.5 µmol/kg, equivalent to 0.93 mg/kg s.c.) (t = 0), and immediately returned to its respective activity chamber. Locomotor activity was measured for either 20 or 120 minutes and expressed either as the distance traveled per 5-minute intervals or the total run distance during the session (Grung et al., 1998). The anti–6-MAM mAb (10–100 mg/kg) or saline was injected immediately before the habituation session, except in the study comparing intravenous versus intraperitoneal administration of the antibody where the mAb was administered 3 hours before habituation to ensure distribution to the blood and in the study investigating the capacity of the antibody to sequester 6-MAM over time where mAb was administered 1 or 2 weeks before the activity test. We have previously reported that injection of saline does not induce changes in run response in C57BL mice (Andersen et al., 2009).
Sample Preparation of Brain Tissue for Opioid Concentration Analysis.
Immediately after the locomotor activity test, the mice were removed from the activity chambers and anesthetized with isoflurane. After blood sampling by heart puncture and cervical dislocation at t = 25 or 125 minutes, the cerebrum was quickly removed, washed in ice-cold 0.9% saline, blotted on a filter paper, and homogenized (2 ml/g tissue) in ice-cold ammonium formate buffer (5 mM, pH 3.1) with sodium fluoride (4 mg/ml). The homogenate was diluted further 1:1 in ammonium formate buffer and immediately frozen in liquid nitrogen. The samples were stored at −80°C until analyzed by LC-MS/MS within 24 hours.
Heroin and heroin metabolites were analyzed by a LC-MS/MS method previously described in Andersen et al. (2009) and Karinen et al. (2009) with minor adjustments. In short, internal standard (50 µl, 0.5 µM) and ice-cold acetonitrile/methanol (500 µl, 85:15) were added to brain homogenate or blood samples (200 µl), immediately shaken, and frozen for at least 10 minutes. After centrifugation (4700g, 4°C, 10 minutes), the organic phase was transferred to a glass tube and evaporated to dryness at 40°C under a gentle stream of nitrogen. The dry residue was reconstituted with ice-cold mobile phase (100 µl, 3% acetonitrile/97% 5 mM ammonium formate buffer, pH 3.1) and centrifuged (3800g, 4°C, 10 minutes). The resulting supernatants were transferred to autosampler vials. Separation of the supernatants and the filtrates from the binding study was performed at 50°C on a XTerra MS C18 column (Waters Corp., Milford, MA) using gradient elution with mobile phase consisting of methanol and ammonium formate buffer (5 mM, pH 3.1). The flow rate was 0.2 ml/min. The calibration standards for heroin were handled separately from the metabolites to examine the stability of heroin during sample preparation and analysis. The limits of detection in brain tissue were 0.04 nmol/g for M3G, 0.004 nmol/g for M6G, 0.005 nmol/g for morphine, 0.003 nmol/g for 6-MAM, and 0.008 nmol/g for heroin. The interassay variability was lower than 15% for all compounds.
Naltrexone was analyzed by a UPLC-MS method. Internal standard (50 µl, 0.1 µM) and methanol (30 µl) were added to the filtrate (70 µl). Separation was performed at 65°C on a HSST3 column (Waters Corp.) using gradient elution with mobile phase consisting of methanol and ammonium formate buffer (10 mM, pH 3.1). The gradient initially contained 20% methanol, which increased to 100% over the next 2.8 minutes and was maintained for 0.3 minute before returning to its initial conditions. The flow rate was 0.5 ml/min. Masses monitored were 342.10 > 270.10 and 242.10 > 282.10 for naltrexone and 345.10 > 270.10 for naltrexone d3. Limit of detection was 0.001 µM. The interassay variability was lower than 15%.
Data are presented as mean ± S.E.M. unless otherwise stated. Statistical analyses of the locomotion data were carried out by General Linear Model for repeated measures with drug and dose as fixed factors and time as repeated factor. The Greenhouse-Geisser correction of the degrees of freedom was used when sphericity was violated (Fig. 3A). For all other locomotion data, each time point was tested separately using a two-tailed t test (Figs. 4,A and B, 5, A and B, and 6, A and B) or analysis of variance (ANOVA) (Fig. 3B). The drug concentration data were analyzed with a t test (Figs. 4C and 6C) or ANOVA (Figs. 2, A and B, 3C, and 5C). P values less than 0.05 were considered as statistically significant. Bonferroni correction was used for the post hoc tests after ANOVA. All statistical tests were performed using SPSS, version 20 (SPSS Inc., Chicago, IL).
In Vitro Studies of the Anti–6-MAM mAb.
In vitro, the anti–6-MAM mAb inhibited the metabolism of 6-MAM to morphine dose-dependently in both rat and human blood (Fig. 2). The highest antibody concentration (6 µM) showed an almost complete inhibition of 6-MAM deacetylation in rat blood. The presence of the anti–6-MAM mAb had no effect on the conversion of heroin to 6-MAM since the concentrations of heroin were unaffected (P > 0.05; Fig. 2B). Although the metabolism of 6-MAM was clearly affected by the presence of the anti–6-MAM mAb, the sum of heroin and heroin metabolites was not changed, strongly indicating that both free and bound 6-MAM were measured by the LC-MS/MS method used. This can be expected when using a sample preparation method with protein precipitation (acetonitrile/methanol).
The antibody used in this study was highly specific toward 6-MAM (>90% binding) compared with the other heroin metabolites morphine, M3G, and M6G (<2% binding). The mAb displayed some binding to heroin (23%), whereas minor binding was observed toward naltrexone (5%; Table 1).
Effects of the Anti–6-MAM mAb on Heroin-Induced Locomotor Activity and Opiate Brain Concentrations.
The locomotor activity induced by 5 µmol/kg heroin reached its maximum after approximately 25 minutes and declined to zero within 90 minutes (Fig. 3A). Administration of the anti–6-MAM mAb 1 hour before injection of heroin affected the run activity dose-dependently (P < 0.01; Fig. 3A), resulting in a reduction of the total run distance by 31%, 43%, and 66% after pretreatment with 10, 50, and 100 mg/kg, respectively (Fig. 3B). Morphine was the only opioid detected in brain tissue 125 minutes after heroin exposure, as expected from previous studies in our laboratory (Andersen et al., 2009). The reduction in the morphine concentration (50%–69%; Fig. 3C) reflected the mAb doses used.
In the subsequent experiments, the mice were given 10 mg/kg anti–6-MAM mAb and 2.5 µmol/kg heroin. Locomotor activity was measured for the first 20 minutes after heroin injection, and brain concentrations of heroin and its metabolites were analyzed in the same mice at 25 minutes. Pretreatment with the antibody 1 hour before injection of heroin reduced locomotor activity by 52% and 64% at 15 and 20 minutes, respectively (Fig. 4A), and reduced the total run distance by 50% (Fig. 4B). This decrease in heroin-induced behavior was accompanied by a 58% and 29% reduction in brain 6-MAM and morphine concentrations, respectively (Fig. 4C). Heroin was not detected in the brain samples.
Compared with the pronounced anti–6-MAM mAb effect seen when the mAb was given 1 hour before heroin injection, the reduction in heroin-induced run activity was no longer significant when the mAb had been in the circulation for 1 or 2 weeks before a single heroin injection (Fig. 5, A and B). One week after administration, the mAb was still able to reduce the brain concentration of 6-MAM upon heroin injection (35% reduction compared with saline-pretreated mice), but 2 weeks after mAb administration, this reduction was not significant (11% reduction, P > 0.05) (Fig. 5C). The data from mice injected with heroin 1 hour after administration of mAb (Fig. 4) are included in Fig. 5 for comparison.
Both intraperitoneal and intravenous injections of the anti–6-MAM mAb significantly reduced the heroin-induced locomotor activity and brain concentrations of 6-MAM and morphine compared with their respective controls (Fig. 6). There were no differences in the mAb effect between the two ways of administration (P > 0.05).
The anti–6-MAM mAb displayed no cross-reactivity toward morphine. This was confirmed in vivo as brain morphine concentrations upon administration of 2.5 µmol/kg s.c. morphine were not different between antibody-treated and control mice (0.054 ± 0.005 nmol/g in control versus 0.059 ± 0.003 nmol/g in antibody-treated mice). This dose of morphine did not induce locomotor activity, as shown previously by Andersen et al. (2009).
Producing an efficient vaccine against heroin has been considered particularly challenging because of the rapid metabolism of heroin to multiple psychoactive molecules. Based on the increasing amount of evidence that 6-MAM is the metabolite responsible for the immediate rewarding effects of heroin, we hypothesized that sequestration of this specific metabolite would be sufficient to markedly impair the acute behavioral effects of heroin. In this study, we therefore examined the effects of a mAb specific toward 6-MAM. We found a severe reduction in heroin-induced locomotor activity in mice pretreated with the anti–6-MAM mAb, which closely corresponded with the measured reductions in brain levels of active heroin metabolites.
The mAb used was based on a previously reported scFv antibody fragment recognizing 6-MAM with an affinity of 0.3 µM (6-MAM-214; Moghaddam et al., 2003). For the purpose of this study, an antibody format with a longer half-life (human IgG1) was customized. This mAb displayed no cross-reactivity to morphine, M3G, or M6G and only modest binding to heroin itself. The absence of cross-reactivity toward morphine (2.5 µmol/kg s.c.) was confirmed in vivo where no difference was found in morphine brain concentrations between control and mAb pretreated mice. In addition, a negligible cross-reactivity was found toward naltrexone, which is the drug in use for treatment of heroin addiction with the highest structure similarity to 6-MAM. The in vitro studies performed in blood confirmed the specificity of the mAb, showing an almost complete block of the metabolism of 6-MAM to morphine, whereas the conversion of heroin to 6-MAM remained unaffected.
In the anti–6-MAM mAb pretreated mice (Figs. 4–6), the 60%–67% reduction in acute heroin-induced locomotor activity 20 minutes after heroin injection corresponded closely with the 58%–60% reduction in brain 6-MAM levels. This strongly supports our previous finding that the immediate heroin response is mediated by heroin’s first metabolite, 6-MAM (Andersen et al., 2009). Reduced brain levels of morphine (29%–44%) were also found in the anti–6-MAM mAb-immunized mice (Figs. 4 and 6). This finding can be explained by the combined effect of reduced brain entrance of the morphine precursor 6-MAM and reduced metabolism of 6-MAM to morphine in the blood, both caused by 6-MAM binding to the antibody. We have previously shown that the brain morphine concentrations found after injection of 2.5 µmol/kg heroin are too low to induce locomotor activity in mice (Andersen et al., 2009), and therefore, the reduction in locomotor activity cannot be assigned to a reduction in morphine levels. Concerning heroin itself, which displays low affinity for µ-opioid receptors (Inturrisi et al., 1983; Selley et al., 2001) and is present in low concentrations in brain after administration (Andersen et al., 2009), it seems unlikely that the modest binding by the antibody can explain the observed ∼60% reduction in the behavioral response.
In two recently published studies using an active immunization strategy with a morphine conjugate producing antibodies targeting heroin, 6-MAM, and morphine, the reduction in brain 6-MAM levels was reported to be 44% and 69% after exposure to heroin or 6-MAM, respectively (Pravetoni et al., 2012; Raleigh et al., 2013). When comparing these results using a broad immune response with the effects of our specific anti–6-MAM mAb, the reductions in brain 6-MAM levels accomplished were of equivalent size. Our in vitro and in vivo experiments indicate that the 6-MAM sequestration could be even further increased by using higher doses of anti–6-MAM mAb.
In theory, the most efficient sequestration strategy for drug abuse vaccines should be to bind the first substance present in blood, in this case heroin, to avoid brain entrance and central effects. Raleigh et al. (2013) used an immunogen known to elicit antibodies with high affinity for both heroin and downstream metabolites and found a pronounced sequestration of heroin in the blood of immunized rats without affecting the brain levels of heroin. This result indicates that only a small fraction of the injected heroin enters the brain so rapidly that it escapes both antibody binding and metabolism by esterases in the blood. Raleigh et al. (2013) concluded that the most likely explanation for the reduced 6-MAM levels in brain of vaccinated animals was extensive binding and retention of 6-MAM in serum. This supports our previous study showing that the metabolism of heroin to 6-MAM is very fast and occurs primarily in the periphery prior to brain entry (Boix et al., 2013). Stowe et al. (2011) examined an antibody with high affinity for heroin and morphine, but no affinity for 6-MAM, and reported it to be inefficient in preventing acquisition of heroin self-administration, emphasizing the importance of 6-MAM sequestration. There are major differences in the presence of heroin-metabolizing enzymes in human and rodent blood (Berry et al., 2009; Bahar et al., 2012), which may complicate the extrapolation of opioid metabolism data across species. Despite these differences, the in vivo half-lives of heroin in rodents and humans have been shown to be remarkably similar (t1/2 ∼2.5–4 minutes) (Way et al., 1960; Rook et al., 2006b; Boix et al., 2013; Gottas et al., 2013; Raleigh et al., 2013).
Independent of an active or passive immunization strategy, it is of major importance that the antibodies have high specificity. Otherwise, the antibodies will rapidly be occupied with inactive or downstream metabolites and the binding capacity readily overcome by taking repeated doses of the abused drug over a short period of time (Kosten and Domingo, 2013). A high antibody specificity also has the advantage of not interfering with endogenous opioids, classic opioid analgesics, or conventional addiction treatment, which again opens for the possibility to combine current opioid substitution therapy with novel immunotherapeutics. A combined active and passive immunization against nicotine has been reported to enhance the nicotine vaccine efficacy (Roiko et al., 2008; Cornish et al., 2011), whereas combination of vaccines toward different drugs of abuse has been shown to be possible without compromising with the efficacy of the individual components (Pravetoni et al., 2012). The main advantages of passive immunization compared with active immunization are that pharmaceutical grade antibodies can be given to every patient in a precise dose, the protection is immediate, the duration of action is more predictable, and customized antibody forms can be selected relative to the therapeutic applications (e.g., long-acting for relapse prevention and short-acting for overdose) (Peterson et al., 2006; Peterson and Owens, 2009). Another advantage using therapies acting through a pharmacokinetic rather than pharmacodynamic blockade is that potential heroin effects mediated through other mechanisms than opioid receptor binding would also be blocked, such as opioid-induced activation of the central immune system through Toll-like receptor 4, suggested to be involved in drug reinforcement (Hutchinson et al., 2012; Wang et al., 2012; Theberge et al., 2013).
The chief obstacles in implementing passive immunotherapy for drug abuse are the need for large amounts of purified mAb and the cost and inconvenience of repeated injections (Brimijoin et al., 2013). The functional elimination half-life of the anti–6-MAM mAb used in this study was tested by injecting mice with mAb and expose to a single dose of heroin at different time points. When heroin was given 1 hour or 1 week after pretreatment, mice given mAb had brain concentrations of 6-MAM that were reduced by 60% and 35% compared with mice pretreated with saline. From these results (Fig. 5), a functional half-life of approximately 8 to 9 days may be estimated for the anti–6-MAM mAb in mice, which is in accordance with other studies reporting IgG half-lives of 5–12 days in rodents (Bazin-Redureau et al., 1997; Norman et al., 2009; Cornish et al., 2011; Treweek et al., 2011). In humans, the half-life of IgG has been reported to be as much as 3–4 weeks (Lobo et al., 2004; Peterson and Owens, 2009), which makes prophylactic prevention of drug abuse feasible, at least for limited time periods in subgroups of patients such as drug-abusing women during pregnancy or detoxified heroin addicts during periods with intense craving.
In the present study, the mice were given mAb in doses of 10–100 mg/kg, which corresponds to approximately 0.5–5 µM IgG, based on a distribution volume about twice the blood volume (Bazin-Redureau et al., 1997) and a molecular mass of 150 kDa. Because IgG has two binding sites per antibody molecule, 1–10 µM drug-binding sites were available. This is within a clinical relevant concentration range as Cmax of 6-MAM has been measured to be 5.0–17.5 µM after intravenous injection of heroin in heroin users (Rentsch et al., 2001; Girardin et al., 2003; Rook et al., 2006b). One of the advantages of passive immunization is the possibility to easily increase the antibody levels in the blood by giving higher doses of antidrug mAb.
Because mAbs are large molecules with relatively poor membrane permeability and stability in the conditions of the gastrointestinal tract, parenteral administration has been the preferred route of mAb administration (Dostalek et al., 2013). Intravenous injections in mice are technically more challenging than intraperitoneal and subcutaneous injections, and therefore, the latter injection techniques are preferred in animal studies. We found that intraperitoneal administration of anti–6-MAM mAb in the mice gave equivalent protection against heroin effects compared with intravenous administration. To ensure that the mAb was absorbed into the blood before heroin exposure, a 4-hour delay was introduced between the intraperitoneal injection of mAb and heroin exposure.
In summary, we have tested a mAb specific toward 6-MAM with minor cross-reactivity toward heroin and no cross-reactivity to morphine or subsequent metabolites. Pretreatment with this mAb reduced the immediate psychomotor stimulating effects of heroin in mice, and this reduction corresponded closely with the reduction in brain 6-MAM levels. The efficacy of the specific passive anti–6-MAM mAb is equivalent with previously reported active vaccines with a broad specificity toward heroin and metabolites. These findings strengthen the view that 6-MAM is the key mediator of acute heroin effects and imply that a vaccine against heroin, either active or passive, needs to sequester 6-MAM in the blood to be efficient.
The authors thank T. K. Olsen and K. M. Olsen for technical assistance. They also thank Affitech Research AS for making the anti–6-MAM mAb available for this research and especially M. Braunagel and U. Scheffler for their important contributions.
Participated in research design: Bogen, Boix, Mørland, Andersen.
Conducted experiments: Bogen, Nerem, Andersen.
Performed data analysis: Bogen, Nerem, Boix, Andersen.
Wrote or contributed to the writing of the manuscript: Bogen, Nerem, Boix, Mørland, Andersen.
- Received December 6, 2013.
- Accepted April 2, 2014.
This work was supported by the Research Council of Norway [Grant 213751].
- analysis of variance
- immunoglobulin G
- liquid chromatography
- monoclonal antibody
- mass spectrometry
- single-chain fragment variable
- ultraperformance liquid chromatography
- Copyright © 2014 by The American Society for Pharmacology and Experimental Therapeutics