Development of an Effective Monoclonal Antibody against Heroin and Its Metabolites Reveals Therapies Have Mistargeted 6-Monoacetylmorphine and Morphine over Heroin

The opioid epidemic is a global public health crisis that has failed to abate with current pharmaceutical treatments. Moreover, these FDA-approved drugs possess numerous problems such as adverse side effects, short half-lives, abuse potential, and recidivism after discontinued use. An alternative treatment model for opioid use disorders is immunopharmacotherapy, where antibodies are produced to inhibit illicit substances by sequestering the drug in the periphery. Immunopharmacotherapeutics against heroin have engaged both active and passive vaccines targeting heroin’s metabolites, 6-monoacetylmorphine (6-AM) and morphine, since decades of research have stated that heroin’s psychoactive and lethal effects are mainly attributed to these compounds. However, concerted efforts to develop effective immunopharmacotherapies against heroin abuse have faced little clinical advancement, suggesting a need for reassessing drug target selection. To address this issue, four unique monoclonal antibodies were procured with distinct affinity to either heroin, 6-AM, or morphine. Examination of these antibodies through in vitro and in vivo tests revealed monoclonal antibody 11D12 as the optimal therapeutic and provided crucial insights into the key chemical species to target for blunting heroin’s psychoactive and lethal effects. These findings offer clarification into the problematic attempts of therapeutics targeting heroin’s metabolites and provide a path forward for future heroin immunopharmacotherapy development.


Safety statement
No unexpected or unusually high safety hazards were encountered.

Synthesis and bioconjugation of the deuterated heroin hapten
The deuterated heroin hapten (H dAc ) was synthesized according to previously published procedures. 1 Conjugation to keyhole limpet hemocyanin (KLH) and bovine serum albumin (BSA) were completed following previous studies. 1

Biochemical and in vivo procedures
Animals 6-week-old male Swiss Webster mice were obtained from Taconic Farms (Germantown, NY, USA) and allowed to acclimate for approximately 1 week before antinociception, pharmacokinetic, and overdose studies. All animal studies were performed in compliance with the Scripps Institutional Animal Care and Use Committee (Protocol #08-0127) and were in accordance with the National Institute of Health Guide for the Care and Use of Laboratory Animals. Mice were group-housed in an AAALAC-accredited vivarium containing temperature-and humidity-controlled rooms, with mice kept on a reverse light cycle (lights on: 9PM to 9AM). All experiments were performed during the dark phase. General health was monitored by both the scientists and veterinary staff at The Scripps Research Institute.

Drugs
Heroin hydrochloride (HCl) was obtained from the National Institute on Drug Abuse as a solid.
Opioids for cross-reactivity screening and deuterated drug standards were obtained from Cerilliant as solutions in organic solvent. Heroin HCl was dissolved and diluted into saline for animal administration.

Vaccinations and hybridoma generation
Each female A/J mouse received 100 μL of a vaccine formulated with 50 µg of H dAc -KLH, 50 µg of CpG ODN 1826 (Eurofins MWG Operon), and 50 μg of alum (Alhydrogel®, Invivogen) in PBS pH 7.4. Vaccines were made fresh and mixed for at least 1 hour, prior to injection. The suspension (100 µL per mouse) was administered intraperitoneally (IP) on weeks 0, 2, 5, and 9. No adverse reactions were observed. Blood sampling was performed on weeks 4, 6, and 10. A final intravenous (IV) infusion of conjugate was performed on week 13. Three days later, one animal was sacrificed, and its spleen was extracted, homogenized and splenocytes were washed with RPMI medium. The resulting B-cells were fused with X63Ag8.653 nonproducing myeloma cells using PEG1500 and selected by culturing in HAT medium. Of the plated hybridomas, 47 were positive for hapten binding by ELISA and further screened by Biacore. Of the 47 hybridomas screened, 17 were frozen and the best were selected for two rounds of subcloning by limiting dilution (4G12, 6B11, 6E1, 11D12). All monoclonal antibodies (mAbs) were isotyped.

Enzyme-linked immunosorbent assay (ELISA)
For initial screening of mouse serum and hybridoma supernatants, a similar procedure was used for end point titer determinations as previously reported. 2 For isotyping, goat antimouse antibodies recognizing kappa/lambda LC or gamma 1/2a/2b HC were used for plate coating.
Surface plasmon resonance (SPR) 1) Ranking antibody binding affinity and specificity from hybridoma culture supernatant by competitive SPR analyses 47 hybridomas were positive for hapten binding in ELISA and were further analyzed by SPR to rank drug affinities and specificities against 6-AM, heroin, and morphine. The analyses were conducted on a Biacore 3000 instrument equipped with a CM5 sensor chip (Cytiva). The ligand, deuterated heroin hapten, was immobilized onto the chip surface using Amine Coupling Kit After each sample analysis, all flow cells were regenerated with 30 seconds injection of 10 mM Gly-HCl, pH 1.5, before next cycle of analysis. From this analysis, 17 hybridomas were selected for more in-depth in vitro competitive analyses, and 6 hybridomas were selected for subcloning and mAb production at the end.

2) Determine IC50 of heroin, 6-AM and morphine for selected mAbs using SPR methodology
Six selected mAbs were subjected to more in-depth competitive analyses to determine the IC50 of heroin, 6-AM and morphine compounds. In brief, the chip was prepared vide supra. Each mAb was pre-titrated and pre-incubated with 12 different concentrations of compounds (twofold dilution from 2000 nM to 1.95 nM with no drug control for heroin, 6-AM and morphine) at room temperature for 1 hour, before the mixture was injected over the sensor chip surface to observe the binding. For each analysis cycle, one pre-incubated sample was injected for 300 sec over all flow cells (i.e., Fc1 through Fc2), followed by 150 sec of dissociation in running buffer. The relative response at the end of dissociation (Fc2 minus Fc1) was recorded in the sensorgram and was used for inhibitive binding evaluation. After each sample analysis, all flow cells were regenerated with 30 seconds injection of 10 mM Gly-HCl, pH 1.5, before next cycle of analysis. The IC50 was calculated using GraphPad Prism software (ver. 6.00).

3) Determine the binding kinetics for selected mAbs via SPR
The binding kinetics of heroin, 6-AM, and morphine towards selected mAbs were conducted on a Biacore S200 instrument equipped with a series S CM5 sensor chip (Cytiva). Each selected mAb was immobilized on the CM5 chip surface using Amine Coupling Kit (Cytiva) as follows: 1) The active flow cell Fc 2 or Fc4 surface was activated for 7 minutes with a 1:1 S4 mixture of 0.1 M NHS and 0.4 M EDC at a flow rate of 10 µL/min; 2) Individual mAb was resuspended in 10 mM sodium acetate (with pre-determined pH per pH scouting results) and injected over activated Fc2 or Fc4 at 10 µL/min aiming at immobilization level of 1500 RU; 3) The flow cell surfaces were blocked with a 7-minute injection of 1.0 M ethanolamine-HCl (pH 8.5) at a flow rate of 10 µL/min. As a reference flow cell, Fc1 or Fc3 was activated with NHS/EDC and blocked with ethanolamine as described above. The kinetics were determined via single cycle kinetics methodology with five analyte concentrations (ranging from 10 nM to 2000 nM), and the data was fitted using a 1:1 binding model. The assay was run in PBS-P+ buffer (Cytiva) at a flow rate of 50µL/min with a 10 Hz data collection rate as follows: 1) 3 x startup cycles (each cycle includes 30 sec of running buffer injection and 300 sec of dissociation, all at a flow rate of 50 µL/min, the chip surface was regenerated with Gly-HCl (pH 1.5) for 30 sec) were conducted before SCK analysis. 2) For binding kinetics, heroin, 6-AM or morphine were prepared in running buffer in five different concentrations. The diluted compound samples were then injected for 30 sec consecutively from low to high followed by 3600 sec of dissociation in running buffer.
3) The sensor chip surface was regenerated with 30 sec of injection of Gly-HCl (pH1.5) solution before the next cycle of SCK analysis. A blank running buffer injection was also conducted before each compound run using the exact same conditions for SCK analysis. All data, including responses from Fc2 minus Fc1 or Fc4 minus Fc3, were collected by Biacore control software in a result file. The run data sets stored in the result file were then analyzed by Biacore S200 evaluation software (ver. 1.1 build:27) using predefined LWM kinetics/affinity single-cycle method. Each run has passed quality control as determined by the Biacore S200 evaluation software, which includes: 1) whether or not the kinetic constants are within instrument specifications; 2) whether or not the kinetics constants appear to be uniquely determined; 3) whether or not significant bulk contribution is found.

4) SPR quantification of selected mAbs in sera for pharmacokinetics studies
To quantify the amount of administrated mAb left in the serum over time, the sera collected at various time points were analyzed via SPR binding assay, and the mAb concentration of each time point was interpolated from a standard mAb binding curve. In brief, the quantitative assay was conducted on a Biacore 3000 instrument equipped with a CM5 chip. A chip was immobilized with deuterated heroin hapten vide supra. For each tested mAb, a standard mAb S5 binding curve was created by injecting a known concentration sample (covering 50, 100, 200, 400, 800, 1600, 3200 and 6400 ng/mL) over all flow cells (i.e., Fc1 through Fc2), followed by 150 sec of dissociation in running buffer. The relative response at the end of dissociation (reference flow cell subtracted, Fc2 minus Fc1) was recorded in the sensorgram and was used for standard curve fitting. After each sample analysis, all flow cells were regenerated with 30 seconds injection of 10 mM Gly-HCl, pH 1.5, before next cycle of analysis. The unknown sample from each time point was prepared in running buffer using pre-determined dilution factor and was injected over both Fc1 and Fc2 as described above. The relative response at the end of dissociation was used to interpolate the serum mAb concentration from the fitted standard curve using GraphPad Prism software (ver. 6.00).

Antinociception
Nociception was measured in two behavioral tests, hot plate (supraspinal) and tail flick (spinal), as previously described. 3 In the hot plate test, mice were placed in an acrylic cylinder (14 cm in diameter x 22 cm in height) on a 55 °C surface, and the latency to perform one of the following nociception responses: hind paw licking, hind paw withdrawal/shaking, or jumping was timed with a 35 s maximum cutoff time to prevent tissue damage. The tail flick test was performed using an ITC Life Science Tail Flick Analgesia Meter to aim a high-intensity light beam (45% active intensity) at the tail. The latency to tail withdrawal from the beam was timed with a 10 s maximum cutoff time to prevent tissue damage. Since, tail flick is a more reflective behavior, the hot plate test was performed first, followed by the tail flick test. Mice were baselined once prior to recording baseline measurements to acclimate them to the testing environment. Baseline measurements for each test were recorded prior to drug injection. All antinociception tests used 7-week-old male Swiss Webster mice.
During cumulative dosing antinociception, mice (n = 6/group) received increasing amounts of drug over the course of the study. Prior to drug exposure, mice were inoculated with mAbs 30 minutes before baseline measurements were conducted. Immediately after obtaining baseline values for both tests, heroin HCl was administered IP and latency to nociception was measured 15 min post-injection. Testing was repeated in 15 min intervals until full antinociception was observed in both assays (i.e. maximum cutoff time was reached). Both drugs were tested at the following intervals to generate a full dose-response curve: 1, 2, 3, 5, and 9 mg/kg. Antinociception Pharmacokinetic analyses were conducted using the two-compartmental model since all mice received IV bolus injections. Elimination and distribution phases were determined using established two-compartmental analysis, where equations were utilized to determine half-lives (t 1/2 ), maximum concentrations (C max ), and area under the curve (AUC). [7][8][9] The initial concentration or C max was determined by adding the inverse natural log of the distribution (A) and elimination phase (B) y-intercepts together. The elimination rate constant (β) was calculated by determining the slope of the natural log of drug concentration over time during the elimination phase. The halflife was calculated using the formula: The AUC was determined using the following equation: where A is the slope of the natural log of drug concentration over time during the distribution phase and all other variables were used in equations explained previously.

Analyses of blood samples
Standard curve samples were prepared by following the same sample processing method as experimental blood samples vide supra, but sample preparation consisted of substituting the 14 μL    Table S4. Statistical analysis of heroin-induced symptoms.
To measure the effect of monoclonal antibody treatment on heroin-induced effects, the appearance of overdose symptoms (death or seizures) was assigned a value of incidence (present = 1, absent = 0) for each mouse regardless of time of symptom onset. Group data was analyzed via the χ² distribution. The theoretical frequencies for each mAb group for seizures and death were based on the drug only control group.  S14 Figure S1. Optimization of drug dose for rescue antinociception assay.
Naïve mice (n = 4/group) were injected with different doses of heroin HCl to observe potentials for full antinociception. Mice were tested every 15 minutes until they returned to baseline values.  Abbreviations: min, minutes.