Ethical considerations

All experimental procedures involving animals were approved by the Columbia University, Memorial Sloan Kettering Cancer Center, Rutgers University or High Point University Institutional Animal Care and Use Committee (IACUC) and adhered to principles described in the National Institutes of Health Guide for the Care and Use of Laboratory Animals and abide by the ARRIVE guidelines.

All human hearts used for the present study were not suitable for transplantation and originated solely from deceased organ donors in the United States (US). Research was carried out by investigators not involved in donor recruitment and all human originating material was received de-identified. As such the study is classified as non-human subjects research (NHR) and does not require prior Institutional Review Board (IRB) approval. Written legal, study nonspecific consent was obtained by the Organ Procurement Organizations (OPOs) from the donor or next of kin wishes. No compensation was provided for the organ donations.

Statistics & reproducibility

Number of replicates and statistical analysis are reported in the figure and table legends, in the supplementary information and source data files. All comparisons were planned before performing each experiment based on literature reports and/or preliminary experiments. Statistical methods were not used to predetermine sample sizes, with the exception that power analyses were conducted to estimate the number of animal subjects required for in vivo experiments.

For self-administration studies sample size calculations were used to determine the number of subjects required for α = 0.05 and β = 0.9, assuming variance determined from previous studies for specific inferential statistical analyses.

Outlier data were not excluded from analysis. Subjects that did not complete the treatment regimens during self-administration studies were excluded from analysis.

Data were replicated using technical and independent replicates. See figure and table legends, the supplementary information and source data files for specific details.

For self-administration studies pseudo-randomization prior to initiation of self-administration training was used.

In vitro pharmacological characterization, MOR/KOR KO and control WT tail-flick tests, cardiotoxicity study, pharmacokinetic and protein/tissue binding experiments were performed and analyzed by experimenters blinded to the chemical identity of the test substance.

Antagonism tail-flick and open-field experiments were performed unblinded, due to the fact that the specific traits of the antagonists used (pre-administration time, receptor target) needed to be known to correctly perform the experiments.

The overt behavioral effects of the test compounds precluded blinding of drugs or doses during self-administration studies.

Biological systems

Cell lines

HEK293T (CRL-3216) was purchased on from American Type Culture Collection (Rockville, MD, USA). CHO-K1 cells were obtained from the American Type Culture Collection American Type Culture Collection (Rockville, MD, USA). No commonly misidentified cell lines were used in the study. Cells have not been authenticated after purchase nor tested for mycoplasma.

For assays performed by contracted research organizations, cells were procured by the specified service provider.

Primary human cardiomyocytes (ex vivo human tissue, AnaBios corporation)

AnaBios Corporation’s procurement network includes only US based Organ Procurement Organizations (OPOs) and Hospitals. Policies for donor screening and consent are the ones established by the United Network for Organ Sharing (UNOS). Organizations supplying human tissues to AnaBios follow the standards and procedures established by the US Centers for Disease Control (CDC), are inspected biannually by the Department of Health and Human Services (DHHS), are certified by Centers for Medicare & Medicaid Services (CMS) and must abide by CMS regulations. Tissue distribution by OPOs is governed by their internal Institutional Review Board (IRB) procedures in compliance with Health Insurance Portability and Accountability Act (HIPAA) regulations regarding patient privacy. All transfers of donor organs to AnaBios are fully traceable and periodically reviewed by US Federal authorities. Donor characteristics are reported on (Fig. S8F) and exclusion criteria were previously described⁶⁸. Only tissue material originating from adult male donor hearts was used within the study.

Animals

All animals received regular veterinary care (weekly by institutional veterinarians) including daily health monitoring (by experimenters) of the animals (observing home cage behaviors, nesting, and body weight). All procedures were designed to minimize any stress/distress.

Surgical procedures considerations

Post-surgical monitoring of animals occurred every 20 min from completion of surgery until sternal recumbency (on all four legs); every 60 min until mobile and then daily. For behavioral procedures, the experimenter monitored the subject in the chamber and the computer to ascertain the drug effects for each session. The investigator also removed the subject from the experimental chamber and transferred the subject to the respective home cage. The experimenter monitored activity, overall appearance, and signs of distress after returning the subject to the home cage. Signs of distress include lack of grooming, food left in the chamber, significant weight loss ( > 80% of free-feeding weight), and signs of infection (lethargy, vocalization upon handling). Veterinarians were consulted prior to initiating any antibiotics treatment for infection.

Diet and feeding conditions

Reverse osmosis water (ultraviolet light treated) and food were provided ad libitum, with no restriction on feeding prior to experiments with exception of lever training and food operant responding (details provided in the corresponding sections).

Columbia University (mice): PicoLab® Rodent Diet #5053 or 5001 (protein ≥20.0/23.0%, fat ≥4.5%, fiber ≤6.0%, ash ≤7.0/8.0% and moisture ≤12.0%).

Rutgers University (mice): irradiated PicoLab® Purina #5058 (protein 20.0%, fat 9.0%, fiber 4.0%, ash ≤6.5% and moisture ≤12.0%).

High Point University (rats and mice): 2018 Tekland Global 18% rodent diet (Envigo), for lever training and food operant responding studies a custom manufactured food pellets were used 45 mg MLab Rodent Tablet 20 mg (TestDiet, Inc., #1811156 (5TUM)).

Sai Life Sciences (rats and mice): T.2014C.15 Teklad global Certified Rodent diet with 14% Protein (Envigo Research Private Ltd, Hyderabad).

Euthanasia practices

Columbia University (mice): CO₂ asphyxiation with death ensured after euthanasia by a secondary method of euthanasia: decapitation/cervical dislocation or bilateral thoracotomy.

Rutgers University (mice): CO₂ asphyxiation using a SmartBox automated CO₂ system (euthanex.com) as approved by Rutgers IACUC.

High Point University (rats and mice): Subjects were placed individually into an anesthesia induction chamber (VetEquip; 10 × 9.375 × 6.5 in (L x W x H)) containing an air tight lid, fresh gas inlet connected to a precision extended-capacity isoflurane vaporizer (Patterson Veterinary Tec-3 Vaporizer, #:07-870-3592), and an evacuation outlet connected to an activated charcoal absorption filter trap (VaporGuard, VetEquip; # 931401). O₂ was used as the carrier gas. The output was adjusted to 4.5 L/min, with an initial flow rate of 2% gradually increasing to 5% over 2 min. Rats lost consciousness within 3 min and the flow remained until respiration ceased and death ensued (5-6 min). Decapitation was used as the secondary (adjunctive) euthanasia method.

Sai Life Sciences (rats and mice): For tissue collection (PK and tissue homogenate binding) animals were sacrificed under deep Isoflurane anesthesia by cutting open the abdominal vena-cava and perfusing the whole body with saline from heart.

General euthanasia by asphyxiation in a CO₂ euthanasia chamber (up to 3 rats or 5 mice), not applicable for tissue collection. The chamber was filled with CO₂ (at a flow rate sufficient to fill 30–70% of the chamber volume per minute) until subjects fell unconscious. Once respiratory movement ceased, CO₂ flow was maintained for an additional minute and animals were kept in the chamber for two additional minutes after the flow was stopped. Death was confirmed after cessation of life signs (heartbeat, respiration, and pupil dilation) by a secondary method of euthanasia: bilateral thoracotomy or exsanguination.

Animal subjects used. Tail-flick, open-field and forced swim tests (Columbia and Rutgers University): Healthy adult C57BL/6 J mice male (8–15 weeks, 29-35 g) and female (8–15 weeks, 19–26 g) were purchased from the Jackson Laboratory (Bar Harbor, ME) and housed 5 mice per cage. Mice were maintained on a 12 h light/dark cycle (lights on [7:00]-[19:00]) and all testing was done in the light cycle. The temperature was kept constant at 22 ± 2 °C, and relative humidity was maintained at 50 ± 5%.

Conditioned placed preference (High Point University): Healthy adult male and female C57BL/6 mice (10–15 weeks, 25–32 g) were purchased from the Jackson Laboratory (Bar Harbor, ME) and housed in groups of 4 to 6 in AllerZone MicroIsolator cages placed in RAIR HD Enviro-Gard™ Ventilated racks with HEPA Filtered Air Units. Vivarium conditions were recorded daily, and the automated climate control ensured constant temperature in the range of 20–24 °C and humidity 40–60%. Mice were maintained on a 12 h light/dark cycle with lights on at [18:00] and experimental sessions took place during the dark phase of the cycle.

Self-administration and neurotropic factors expression studies: Adult male Fisher F-344 rats (CDF Strain, 90–150 days, 230–280 g) were purchased from Charles River Laboratories (Wilmington, MA) and housed in groups of 2 in AllerZone MicroIsolator cages placed in RAIR HD Enviro-Gard™ Ventilated racks with HEPA Filtered Air Units prior to surgery. Following surgery rats were housed individually in acrylic cages. Vivarium conditions were recorded daily, and the automated climate control ensured constant temperature in the range of 20–24 °C and humidity 40-60%. Rats were maintained on a 12 h light/dark cycle with lights on at [18:00], and experimental sessions took place during the dark phase of the cycle.

Pharmacokinetic studies: Healthy adult male C57BL/6 mice (8–12 weeks, 18–36 g) or healthy adult male Wistar rats (8–12 weeks, 250–280 g) were procured from Global (India). Three mice or rats were housed in each cage. Temperature and humidity were maintained at 22 ± 3 °C and 30–70%, respectively and illumination was controlled to give a sequence of 12 h light and 12 h dark cycle. Temperature and humidity were recorded by auto-controlled data logger system.

In vitro studies

Broad receptor panel assay

Binding constants (K_i) at the selected human receptors, ion channels and transporters were generously determined using radioligand displacement experiments by the National Institute of Mental Health’s Psychoactive Drug Screening Program, Contract # HHSN-271-2008-00025-C (NIMH PDSP)⁶⁹. The NIMH PDSP is Directed by Bryan L. Roth MD, PhD at the University of North Carolina at Chapel Hill and Project Officer Jamie Driscoll at NIMH, Bethesda, MD, USA. For experimental details please refer to the PDSP website http://pdsp.med.unc.edu/.

Commercial in vitro assays

KOR [³⁵S]GTPγS agonist assay (for noribogaine, item 333200), rat NMDA (Glutamate, [³H]MK-801, item 233010), human NOP (Orphanin ORL1, item 260600) and hERG ([³H]Dofetilide; item 4094) radioligand displacements and human nicotinic acetylcholine receptor (nAChR α₃β₄, item CYL8057IF2) inhibition studies were performed by contracted research organization. Experiments were conducted as described on https://www.eurofinsdiscoveryservices.com/.

hSERT inhibition assay

Stably transfected hSERT-HEK cellular cultures were maintained in Dulbecco’s Minimal Essential Medium (DMEM) with GlutaMAX (Gibco-Thermo Fisher Scientific, Waltham, MA) with the following additions: 10% (v/v) Fetal Bovine Serum (FBS, Atlanta Biologicals, Flowery Branch, GA), 100 U/mL Penicillin 10 μg/mL Streptomycin (Gibco-Thermo Fisher Scientific, Waltham, MA), and 500 μg/mL Geneticin (G418) (Sigma Aldrich, St. Louis, MO). Singly transfected cells were seeded at a density of 0.09 \(\times\) 10⁶ cells/well in poly-D-Lysine (Sigma Aldrich, St. Louis, MO) coated white solid-bottom 96-well plates (Greiner Bio-One, Kremsmünster, Austria). Cells were grown for 44 h in aqueous media at 37 °C under 5% CO₂ atmosphere. At the beginning of the experiment, the cellular growth solution was aspirated, and individual cells were rinsed with 150 μL of 1 × Dulbecco’s Phosphate Buffered Saline (PBS; HyClone Laboratories, Logan, UT). 63 μL of Experimental Media (DMEM without phenol red but with 4.5 g/L of D-Glucose (Gibco-Thermo Fisher Scientific, Waltham, MA), 1% (v/v) FBS, 100 U/mL Penicillin, and 10 μg/mL Streptomycin) with tiered concentrations of inhibitor (vehicle: DMSO and control inhibitor: Imipramine (Sigma Aldrich, St. Louis, MO)) were added⁷⁰. After pre-incubation period (one hour), 63 μL of Experimental Media containing various concentrations of tested inhibitor (or vehicle) along with 2 × a specified amount of fluorescent substrate APP⁺ (final concentration: 1.1 μM, Sigma Aldrich, St. Louis, MO)⁷¹ was added to the wells. After fluorescent probe was allowed to uptake for 30 min, the contents of each well were aspirated and consequently, rinsed twice with 120 μL of PBS. A final solution of 120 μL of PBS was finally added to all corresponding wells for cell maintenance before measuring fluorescence uptake (BioTek H1MF plate reader (BioTek Instruments, Winooski, VT), APP+ excitation and emission wavelengths 436 and 500 nm). Recorded inhibitor values were first subtracted from vehicular values to quantify the respective fluorescence uptake. Data were analyzed using the dose-response-inhibitor nonlinear curve fitting model (log[inhibitor] vs response (four parameters) in Graphpad Prism 10 (Graphpad Software, La Jolla, CA).

Radioligand competition binding assays (Mouse opioid receptors)

Assays were performed as previously reported⁷². Briefly, IBNtxA and [¹²⁵I]BNtxA were synthesized at MSKCC as previously described^73,74,75. Na¹²⁵I was purchased from Perkin-Elmer (Waltham, MA). [¹²⁵I]BNtxA binding was carried out in membranes prepared from Chinese Hamster Ovary (CHO) cells stably expressing murine clones of MOR, DOR, and KOR, as previously described^73,74,76. Binding incubations were performed at 25 °C for 90 min in 50 mM potassium phosphate buffer, pH 7.4, containing 5 mM magnesium sulfate. After the incubation, the reaction was filtered through glass-fiber filters (Whatman Schleicher & Schuell, Keene, NH) and washed three times with 3 mL of ice-cold 50 mM Tris-HCl (pH 7.4) on a semiautomatic cell harvester. Nonspecific binding was defined by addition of levallorphan (8 µM) to matching samples and was subtracted from total binding to yield specific binding. Protein concentrations were determined using the Lowry method with BSA as the standard⁷⁷. Data were analyzed using the binding-competitive nonlinear curve fitting model (One site – Fit K_i) available in Graphpad Prism 10 (Graphpad Software, La Jolla, CA).

[³⁵S]GTPγS functional assay (mouse opioid receptors)

Assays were performed as previously reported⁷². Briefly, [³⁵S]GTPγS binding was performed on membranes prepared from transfected cells stably expressing opioid receptors in the presence and absence of the indicated compound for 60 min at 30 °C in the assay buffer (50 mM Tris-HCl (pH 7.4) 3 mM MgCl₂, 0.2 mM EGTA, and 10 mM NaCl) containing 0.05 nM [³⁵S]GTPγS; 2 μg/mL each leupeptin, pepstatin, aprotinin, and bestatin, and 30 μM GDP, as previously described⁷⁸. After the incubation, the reaction was filtered through glass fiber filters (Whatman Schleicher & Schuell, Keene, NH) and washed three times with 3 mL of ice-cold buffer (50 mM Tris-HCl, pH 7.4) on a semiautomatic cell harvester. Filters were transferred into vials with 3 mL of Liquiscint (National Diagnostics, Atlanta, GA), and the radioactivity in vials was determined by scintillation spectroscopy in a Tri-Carb 2900TR counter (PerkinElmer Life and Analytical Sciences). Basal binding was determined in the presence of GDP and the absence of drug. Data were normalized to 1000 nM DAMGO, DPDPE, and U50,488 for MOR, DOR, and KOR binding, respectively. Data were analyzed using the dose-response-stimulation nonlinear curve fitting model (log[agonist] vs. response (three parameters)) in Graphpad Prism 10 (Graphpad Software, La Jolla, CA).

BRET functional assays (G-protein, β-arrestin and Nb33 recruitment)

HEK-293T cells were obtained from the American Type Culture Collection (Rockville, MD) and were cultured in a 5% CO₂ atmosphere at 37 °C in Dulbecco’s Modified Eagle Medium (high glucose #11965; Life Technologies Corp.; Grand Island, NY) supplemented with 10% Fetal Bovine Serum (FBS, #35-010-CV, Corning, Corning, NY, USA), 100 U·mL⁻¹ penicillin (#30-002-CI, Corning, Corning, NY, USA), and 100 µg·mL⁻¹ streptomycin (#30-002-CI; Corning, Corning, NY, USA). The following chemicals were used without further modification: coelenterazine H (#DC-001437, Dalton Pharma Services, Toronto, ON, Canada), PEI (#NC1014320, Polysciences, Warrington, PA, USA), DAMGO (Abcam, Cambridge, United Kingdom) and (±)-U-50488 HCl (Tocris Biosciences, Minneapolis, MN, USA).

DNA Constructs (G-protein, β-arrestin, and Nb33): The rat KOR (rKOR) and mouse MOR (mMOR) were provided by Dr. Lakshmi Devi at Mount Sinai School of Medicine. The human KOR (hKOR) and MOR (hMOR) were obtained from the Missouri S&T Resource Center. The GRK3, Gα_oB with Renilla luciferase 8 (RLuc8) inserted at position 91 (Gα_oB-RLuc8), and Gβ₁ (β₁) were provided by C. Galés^79,80. Venus-Arrestin2 and Gγ_2, which was fused to the full-length mVenus at its N-terminus via the amino acid linker GSAGT (mVenus-γ2), were constructed in house. The expression vectors coding for rat/human KOR and mouse/human MOR tagged at the C-terminus with Nanoluc (KOR-nluc or MOR-nluc) and were constructed using standard techniques in molecular biology and confirmed by DNA sequencing (Genewiz, South Plainfield, NJ, USA). Briefly, three DNA inserts were PCR amplified, one coding for the N-terminal signal peptide and flag tag, one coding for KOR or MOR, and one coding for nanoluc. The inserts were ligated and cloned into a pcDNA3.1 (+) vector (#V79020, ThermoFisher Scientific, Waltham, MA, USA). The plasmid coding for the nanobody33Venus (Nb33) construct⁴⁹ was a gift from Dr. Meritxell Canals at the University of Nottingham.

The following cDNA amounts were transfected into HEK-293T cells (4 × 10⁶ cells/plate) in 10 cm dishes using polyethylenimine (PEI) in a 1.5:1 ratio (diluted in DMEM, Life Technologies). G-protein β-γ release: 2.5 μg KOR, 0.1 μg Gα_oBRLuc8, 6.2 μg β₁, 6.2 μg mVenus-γ2. Arrestin recruitment: 0.2 μg KOR-nluc, 15 μg Arrestin-2-Venus, with/without 5 μg GRK3. Cells were maintained in the HEK-293T media described above. After 24 hours the media was changed, and the experiment was performed 48 h after transfection. Nb33 recruitment: A total of 5 μg of cDNA was transiently transfected into HEK-293T cells (2 × 10⁶ cells per plate) in 10 cm dishes (1 μg receptor-nluc, and 4 μg Nb-33-Venus), using PEI in a 6:1 ratio (diluted in DMEM). Cells were maintained in the HEK-293T media described above. Experiments were performed 48 h after transfection.

Transfected cells were dissociated and re-suspended in phosphate-buffered saline (PBS). Approximately 200,000 cells/well were added to a black-framed, white-well, 96-well plate (#60050; Perkin Elmer; Waltham, MA). At time zero, the luciferase substrate coelenterazine H (5 μM) was added to each well. Ligands were added after 5 min, then BRET signal was measured 5 min later for G-protein, and 10 min later for Nb33 and β-arrestin recruitment. BRET measurements were performed using a PHERAstar FS plate reader (BMG Labtech, Cary, NC, USA). The BRET signal was calculated as the ratio of the light emitted by the mVenus acceptor (510–540 nm) over the light emitted by the NanoLuc donor (475 nm). This drug-induced BRET signal was normalized using the E_max of U-50,488 (for KOR) and DAMGO (for MOR) as the maximal response at KOR or MOR. Data were analyzed using the dose-response-stimulation nonlinear curve fitting model (log[agonist] vs. response (four parameters)). All experiments were repeated in three independent trials, each with triplicate determinations.

TRUPATH BRET based assays

Assays were performed as previously reported^23,81. Briefly, cells were plated either in 6-well dishes at a density of 700,000–800,000 cells per well, or 10 cm dishes at 7–8 million cells per dish. Cells were transfected 2–4 h later, using a 1:1:1:1 DNA ratio of receptor: Gα-RLuc8:Gβ:Gγ-GFP2 (100 ng per construct for 6-well dishes, 750 ng per construct for 10 cm dishes), except for the Gγ-GFP2 screen, in which an ethanol coprecipitated mixture of Gβ1–4 was used at twice its normal ratio (1:1:2:1). Transit 2020 (Mirus Biosciences) was used to complex the DNA at a ratio of 3 µl Transit per µg DNA, in OptiMEM (Gibco-ThermoFisher) at a concentration of 10 ng DNA per µl OptiMEM. The next day, cells were harvested from the plate using Versene (0.1 M PBS  +  0.5 mM EDTA, pH 7.4) and plated in polyD-lysine-coated white, clear-bottom 96-well assay plates (Greiner Bio-One) at a density of 30,000–50,000 cells per well. One day after plating in 96-well assay plates, white backings (PerkinElmer) were applied to the plate bottoms, and growth medium was carefully aspirated and replaced immediately with 60 µl of assay buffer (1× Hank’s balanced salt solution (HBSS)  +  20 mM HEPES, pH 7.4), followed by a 10 µl addition of freshly prepared 50 µM coelenterazine 400a (Nanolight Technologies). After a 5 min equilibration period, cells were treated with 30 µl of drug for an additional 5 min. Plates were then read in an LB940 Mithras plate reader (Berthold Technologies) with 395 nm (RLuc8-coelenterazine 400a) and 510 nm (GFP2) emission filters, at integration times of 1 s per well. Plates were read serially six times, and measurements from the sixth read were used in all analyses. BRET2 ratios were computed as the ratio of the GFP2 emission to RLuc8 emission. Data were analyzed using the dose-response-stimulation nonlinear curve fitting model (log[agonist] vs. response (three parameters)). All experiments were repeated in three independent trials each with duplicate determinations.

Cardiotoxicity assay in adult human primary cardiomyocytes (AnaBios corporation)

Cardiotoxicity of noribogaine and oxa-iboga analogs was assessed according to published procedure^40,41. Briefly, adult human primary ventricular myocytes were isolated from ethically consented donor hearts that were enzymatically digested using a proprietary protocol. Cardiomyocytes were placed in a perfusion chamber mounted on the stage of inverted Motic AE31E (IonOptix) or Olympus IX83P1ZF (MyoBLAZER) microscope and continuously perfused at approximately 2 mL/min with recording buffer heated to 35 ± 1 °C using an in-line heater from Warner Instruments (IonOptix & MyoBLAZER) and allowed to equilibrate for 5 min under constant perfusion. The cells were field stimulated with supra-threshold voltage at a 1 Hz pacing frequency, with a bipolar pulse of 3 ms duration, using a pair of platinum wires placed on opposite sides of the chamber connected to a MyoPacer stimulator. Starting at 1 V, the amplitude of the stimulating pulse was increased until the cardiomyocytes started generating contractility transients, and a value 1.5× threshold was used throughout the experiment. Cardiomyocytes were then imaged at 240 Hz using an IonOptix MyoCam-S CCD camera (IonOptix) or at 148 Hz using an Optronis CP70-16-M/C-148 (MyoBLAZER) camera. Digitized images were displayed within the IonWizard acquisition software (IonOptix) or MyoBLAZER acquisition software. The longitudinal axis of the selected cardiomyocyte was aligned parallel to the video raster line, by means of a cell framing adapter. Optical intensity data was collected from a user-defined rectangular region placed over the cardiomyocyte image. The optical intensity data represented the bright and dark bands corresponding to the Z-lines of the cardiomyocyte. The IonWizard software or MyoBLAZER Analysis software analyzed the periodicity in the optical density of these bands by means of a fast Fourier transform algorithm.

Compound test solutions were formulated from stock solutions within 30 min prior to experimental application to the cells. Test solutions were applied after vehicle control (120 s interval, 1 Hz stimulation) in an increasing concentration order (in 300 s intervals, 1 Hz stimulation) and experiment was terminated after wash control (300 s interval, 1 Hz stimulation).

Positive control 30 nM ATX-II (toxin from anemonia sulcate) was applied after vehicle control (120 s interval, 1 Hz stimulation) and the data were recorded (300 s interval, 1 Hz stimulation).

An aftercontraction (AC) was visually identified as spontaneous secondary change in the slope of the contractility transient that occurred before the next stimulus-induced contraction and that produced an abnormal and unsynchronized contraction. Contraction Failure (CF) was also visually identified when an electrical stimulus was unable to induce a contraction. Alternans and Short-Term Variability (STV) are visualized in Poincaré plots of Contraction Amplitude variability. STV (STV = Σ|CA_n+1 − CA_n | (20×\(\sqrt{2}\))⁻¹) was calculated with the last 20 transients of each control and test article concentration period (CA_n – contractility amplitude, nth in sequence). Alternans were identified as repetitive alternating short and long contractility amplitude transients. STV values were normalized to the vehicle control value of each cell. AC, CF and Alternans were plotted and expressed as % of incidence of cells exhibiting each of the signals normalized by the total number of cardiomyocytes.

The concentration range was selected on the basis of these considerations: the free plasma noribogaine C_max values were estimated from the reported clinical data of total plasma concentrations³⁹ and our own human plasma protein binding data (Fig. S7g); approximately 100 nM noribogaine (60 mg oral dose) produced a mild QT effect and represents a relatively safe plasma level, while 300–400 nM (180 mg oral dose) induced concerning levels of QT prolongation ( > 500 ms), whereas low micromolar concentrations can be reached after detox therapeutic doses of ibogaine ( > 8 mg/kg) which have been associated with risk of arrhythmias.

In vivo pharmacology and behavioral experiments

Test compounds and formulations

Drugs for in vivo experiments were synthesized in the Sames laboratory as described in the synthesis section (noribogaine hydrochloride and oxa-iboga analogs) or obtained from following sources and used as received: U50,488 hydrochloride (Tocris Bioscience), cyprodime hydrochloride (Tocris Bioscience), naloxone hydrochloride (Alfa Aesar), naltrindole hydrochloride (MedChemExpress), imipramine hydrochloride (Alfa Aesar), aticaprant free base (MedChemExpress), morphine sulfate (Gallipot, Inc., St. Paul, MN), penicillin G procaine (Butler Company, Columbus, OH), propofol (Patterson Veterinary Supply, Inc., Loveland, CO), fentanyl citrate (Fagron, Inc., St Paul, MN), ketamine hydrochloride (Ketaset), xylazine (Xylamed, Patterson Veterinary Supply, Inc., Greely, CO), cocaine hydrochloride and heroin hydrochloride (Drug Supply Program of the National Institute on Drug Abuse, Bethesda, MD). The compound solutions for pharmacology experiments (tail-flick, place conditioning preference, open field and forced swim test) were prepared on the same day as testing from pure solid material. Solids were dissolved in UPS grade 0.85% saline with addition of 2 molar equivalents of glacial acetic acid. Heat and sonication were used to assist in fully dissolving the solid to obtain a clear solution. Aticaprant was solubilized using 7% (v/v) Tween 80 in saline with addition of 2 molar equivalents of glacial acetic acid. Compound solutions were prepared at a concentration allowing for a volume of injection (190–350 µL) based on the body weight of the animal.

For self-administration and GDNF expression experiments test compounds (noribogaine, oxa-noribogaine and epi-oxa-noribogaine) were dissolved in vehicle (2 molar equivalents of acetic acid in water) at a concentration of (20 mg/mL) to allow for injection volume of (2 mL/kg). For example, a 40 mg/kg dose administered to a 300 g rat required injecting 12 mg of test substance in 0.6 mL of vehicle (0.5 mg/mL/kg). Heat and sonication were used to assist in fully dissolving the solid to obtain a clear solution.

For pharmacokinetic study, noribogaine hydrochloride solution for subcutaneous (s.c.) administration was prepared by dissolving the solid material in DMSO and the stock solution was diluted with normal saline to obtain a final formulation in 5% DMSO, 95% (v/v) normal saline (formulation strength 2 mg/mL). The final mixture was vortexed and sonicated to obtain a clear solution. Oxa-noribogaine was solubilized using 0.85% saline with addition of 2 molar equivalents of glacial acetic acid, formulations were vortexed and sonicated to obtain a clear solution (formulation strength 2 mg/mL for s.c. and 4 mg/mL for i.p. administration).

Tail-flick test

Tests were performed as previously reported²⁴. Mice were moved to the testing room 30 min before the experiment to allow for acclimation. The body weight of each mouse and base tail-flick value were recorded. Mice were administered a 1 mg/kg s.c. dose of compound solution. After injection, mice were returned to the home cage and allowed to rest for 30 min. Thirty minutes post injection, the tail-flick measurement was taken using thermal stimulation via IR on a Ugo Basile unit set to 52 PSU (to achieve a baseline between 2 and 3 s and 10 s was used as a maximum latency to prevent tissue damage). Mice were then administered 3 mg/kg s.c. dose, allowed to rest for 30 min, followed by another tail-flick measurement. This process was repeated for doses 10 and 30 mg/kg in increasing order. Tail-flick latencies for the different doses were expressed as percentage of maximum potential effect (% MPE) by subtracting the experimental value by the base tail flick value then dividing by the difference between the maximum possible latency (10 s) and the base tail-flick value and finally multiplying by 100.

ED₅₀ value for U50,488 was determined using four separate groups of naive mice (5 animals for each dose, baseline latency was determined for each animal prior to drug administration), as cumulative dosing of U50,488 in the same animal resulted in a rapid development of tolerance.

ED₅₀ values were calculated using the dose-response-stimulation nonlinear curve fitting model (log[agonist] vs. response (four parameters)) with applied constraints Top = 100 and Bottom = 0.

Tail-flick test (KO animals)

For analgesic testing in knockout animals MOR-1 Exon-1 KO⁸² and KOR-1 Exon-3 KO⁸³ mice on a C57 background were bred in the Pintar laboratory at Rutgers University. All mice used were opioid naïve. Analgesia was tested in wild-type and KO animals by the radiant heat tail-flick technique using an IITC Model 33 Tail Flick Analgesia Meter as previously described^84,85. The intensity was set to achieve a baseline between 2 and 3 s. Tail flick antinociception was assessed as an increase in baseline latency, with a maximal 10 s latency to minimize damage to the tail. Data were analyzed as percent maximal effect, % MPE, which was calculated according to the formula: % MPE [(observed latency − baseline latency)/(maximal latency − baseline latency)] × 100. Compounds were administered subcutaneously (s.c.) and analgesia was assessed at the peak effect (15 min). Mice were tested for analgesia with cumulative subcutaneous doses of the drug until the mouse can withstand the maximal latency. Once the mouse reached the maximal latency, the mouse was no longer given higher doses.

ED₅₀ values were calculated using the dose-response-stimulation nonlinear curve fitting model (log[agonist] vs. response (four parameters)) with applied constraints Top = 100 and Bottom = 0. Statistical significance (WT vs KO-model) was assessed using the extra-sum-of-squares F test (alpha = 0.05).

Tail-flick test (antagonism pre-treatment)

Mice were transferred to the testing room 30 min before the experiment to allow for acclimation. The weight of each mouse and base tail-flick value were recorded. Mice were administered the selected antagonist s.c. Antagonists were prepared and used at the following doses: cyprodime 10 mg/kg, naloxone 10 mg/kg, naloxone 1 mg/kg, naltrindole 0.5 mg/kg, and aticaprant 0.1 mg/kg. After injection, mice were returned to the home cage and allowed to rest for a duration of time specific to the antagonist being used. The wait time after administration of antagonist are as follows: cyprodime 30 min, naloxone 30 min, naltrindole 45 min and aticaprant 20 min. Mice were then administered oxa-noribogaine s.c. prepared at the dose of 10 mg/kg. After injection, mice were returned to the home cage to rest for a duration of time specific to the experimental compound being used. The wait time after administration of oxa-noribogaine was 30 min. Afterwards, the tail-flick measurement was taken as described above.

Statistical analysis was performed using the unpaired t-test, two-tailed with Welch’s correction.

Open field test

Mice were transferred to the testing room 30 min before the experiment to allow for acclimation. Body weight of each mouse was recorded. For experiments with antagonist, pre-treatment, the dosing and after application wait times were the same as for the tail-flick test. Mice were administered an ED_50, ED₈₀ or >ED₉₅ doses of test compound (volume of injection 290–350 µL based on body weight). After injection, mice were returned to the home cage and allowed to rest for 30 min. Each mouse was then gently placed in the center of a clear Plexiglas arena (27.31 × 27.31 × 20.32 cm, Med Associates ENV-510) lit with dim light ( ~ 5 lux) and allowed to ambulate freely for 60 minutes. The locomotion of the mouse was tracked by infrared beams embedded along the X, Y, Z axes of the area and automatically recorded. Data was collected on Activity Monitor by Med Associates.

Statistical analysis was performed using the unpaired t-test, two-tailed with Welch’s correction to analyze total locomotion and a two-way ANOVA with Šidák’s multiple comparison test for post hoc comparison to analyze locomotion across time bins.

Forced swim test

One week after arriving in the facility, mice were handled for approximately 5 min by a male experimenter. All experiments were carried out by the same male experimenter who performed the handling. On the day of FST testing, mice were weighed and transferred to the testing room 30 min prior to experimentation. Mice were administered 0.85% saline (s.c.), imipramine 15 mg/kg (i.p.), oxa-noribogaine ED₈₀ (5.4 mg/kg) or >ED₉₅ 10 mg/kg dose s.c. 15 min after receiving saline or imipramine injection, or 30 min after receiving oxa-noribogaine injection, each subject mouse was gently placed into cylinder (plexiglass 40 cm tall, diameter 22 cm) filled with water (height 18 cm) kept at 23.8–25.0 °C. Mice were allowed to swim for 6 min after which they were removed, dried, and returned to their home cage. Video footage was analyzed with Noldus FST scoring software to obtain immobility scores. The mouse’s view of the FST cylinder was obstructed prior to initiating the experiment. Only the last 4 min of the 6 min test were analyzed for time spent immobile. All FSTs were performed between the hours of [12:00] and [18:00].

Statistical analysis was performed using the unpaired t-test, two-tailed with Welch’s correction.

Place conditioning preference

Place conditioning was conducted in a three-chamber apparatus with two equal sized chambers (16.8 × 12.7 × 12.7 cm) that differed in color (black, white) and flooring (mesh, bar), separated by a middle chamber (7.2 × 12.7 × 12.7 cm) with computer-controlled doors and equipped with infrared diodes (Med Associates). Mice were acclimated to the room for three days before the pretest phase.

For determination of pre-conditioning baseline preferences, mice were placed in the middle chamber; the doors were raised, and the animal had free access to the chamber for 20 min. Dependent measures included the amount of time spent in each compartment, activity in the compartments, and entries into the compartments. No statistically significant difference was observed in the amount of time spent between the compartments during the pretest phase. Using an unbiased approach, the drug-paired chamber was randomly assigned for each mouse. Mice spending more than 65% time in either compartment or 33% in the middle compartment during the pre-conditioning session were excluded from the study. Conditioning began the day following the pretest phase and consisted of three two-day pairing cycles. On the first, third and fifth days of conditioning, mice received either cocaine (10 mg/kg, i.p.), morphine (20 mg/kg, i.p.), oxa-noribogaine (10 or 40 mg/kg, i.p.) or epi-oxa-noribogaine (10 and 40 mg/kg, i.p.) and were confined to the drug paired compartment for 30 min. On the second, fourth, and sixth days, mice were administered saline (1 mL/kg, i.p.) and confined to the vehicle-paired compartment for 30 min.

For testing on the seventh day, mice were placed in the middle compartment, doors were raised, and the mice were allowed access to the entire chamber for 15 min. Dependent measures were identical to the pretest phase.

A CPP score was calculated as the difference in time spent in the drug-paired minus the vehicle paired compartment for both the pretest and testing phases. Data were analyzed using paired t-tests or two-way ANOVA with Šidák’s multiple comparison test for post hoc comparison.

General procedures for intravenous drug self-administration and food responding studies in rats

Rats were transferred to operant conditioning chambers (ENV-008CT; Med Associates, St. Albans, VT) enclosed in sound-attenuating cubicles (ENV-018; Med Associates). The front panel of the operant chambers contained two response levers (4 cm above the floor and 3 cm from the side walls), a cue light (3 cm above the lever) and a food chute centered on the front wall (2 cm above the floor) that was connected to a food pellet dispenser (ENV-023; Med Associates) located behind the front wall and a tone generator to mask extraneous noise. A syringe pump (PHM-100; Med Associates) holding a 20 mL syringe delivered infusions. A counter-balanced arm containing the single channel liquid swivel was located 8–8.5 cm above the chamber and attached to the outside of the front panel. An IBM compatible computer was used for session programming and data collection (Med Associates Inc., East Fairfield, VT).

For lever training, rats were transferred to operant chambers for daily experimental sessions. Food access was restricted in the home cage, such that rats were maintained at 90% of their normal body weight until responding was engendered at which point food was provided ad libitum. Responding was engendered by delivery of food pellets (45 mg pellets, TestDiet, Richmond, IN) under a fixed ratio (FR 1) schedule of reinforcement. The lever lights were illuminated when the schedule was in effect. Completion of the response requirement extinguished lights, delivered food, and was followed by a 20 s timeout (TO) period during which all lights were extinguished, and responses had no scheduled consequences. After the TO, the lights were illuminated, and the FR schedule was again in effect. Sessions lasted 20 min or until 50 food pellets were delivered.

After operant responding was acquired and maintained by food, subjects were surgically implanted with an intravenous jugular catheter. Venous catheters were inserted into the right jugular vein following administration of ketamine (90 mg/kg; i.p.) and xylazine (5 mg/kg; i.p.) for anesthesia as described previously^86,87,88. Catheters were anchored to muscle near the point of entry into the vein. The distal end of the catheter was guided subcutaneously to exit above the scapulae through a Teflon shoulder harness. The harness provided a point of attachment for a spring leash connected to a single-channel fluid swivel at the opposing end. The catheter was threaded through the leash and attached to the swivel. The other end of the swivel was connected to a syringe (for saline and drug delivery) mounted on a syringe pump. Rats were administered penicillin G procaine (75,000 units in 0.25 mL, i.m.) and allowed a minimum of 5 days to recover before self-administration studies were initiated. Catheter patency was maintained by hourly infusions of 0.2 ml of 0.9% saline (w/v) with 1.7 U/ml of heparin equivalent to 0.34 U heparin/infusion while in the home cage only. Infusions of propofol (6 mg/kg; i.v.) were manually administered as needed to assess catheter patency.

Following recovery, rats were transferred to their respective operant chambers for daily self-administration sessions. Before each session, the swivel and catheter were flushed with 500 µL of heparinized saline before connecting the catheter to the syringe mounted on the syringe pump outside of the sound attenuating chamber via a 20 ga luer hub and 28 ga male connector. At the start of each self-administration session, the stimulus light above the active lever was illuminated and both the active and inactive levers were extended. A response on the active lever (FR1) resulted in a 20 s time out (FR1:TO 20 s) during which time the subject received a 200 µL intravenous drug infusion (over the first six seconds), lever light extinguished, levers are retracted, tone is generated, and the house light was illuminated. At the end of the TO, the levers were extended, lever light illuminated, tone silenced, and the house light extinguished. The self-administration session continued for 2 h or until 70 infusions were delivered. Self-administration studies were initiated once stable responding was achieved (defined as total number of infusions ± <20% of the mean of the three previous sessions)^88,89,90.

Effect of oxa-noribogaine on morphine self-administration

Separate groups of rats were trained to self-administer morphine (10, 20 or 40 µg/infusion) under an FR1 schedule of reinforcement. After stable responding was established, rats were administered vehicle (2 mL/kg, i.p.) 30 min prior to the subsequent experimental session. Four days following vehicle administration, rats were administered oxa-noribogaine in an ascending dose manner (3, 10 or 30 mg/kg, i.p.), administered 30 min prior to the beginning of the session. Responding was assessed for seven days following each dose. For each morphine dose, baseline and vehicle active and inactive lever responding was compared using paired t-tests, two tailed. The acute effects of oxa-noribogaine on responding (active and inactive lever) maintained by each morphine dose was assessed independently using a two-way ANOVA with Tukey’s multiple comparison test. The effects of oxa-noribogaine on morphine self-administration on active and inactive lever responding for the seven days following each dose of oxa-noribogaine were assessed using a mixed-model analysis of repeated measures data (due to missing data points) with Šidák’s multiple comparison test for post hoc comparison.

Effect of oxa-noribogaine on cue-induced reinstatement of responding previously maintained by morphine or fentanyl

Two groups of rats were trained to self-administer morphine (20 µg/infusion) or fentanyl (0.625 µg/infusion) in 2 h sessions under FR1 schedule of reinforcement. After a minimum of two weeks of self-administration and three days of stable responding (defined as total number of infusions ± <20% of the mean of the three previous sessions)^88,89,90 immediately prior to the extinction session, rats underwent operant extinction training in 2 h session in the absence of the drugs and associated cues until responding was <20% of baseline responding maintained by intravenous self-administration of the drug. During the extinction phase, levers were available, but responding on the active lever did not result in cue light presentation or drug delivery.

The day following the last extinction session, each group was divided into two subgroups to receive either vehicle (2 mL/kg, i.p.) or oxa-noribogaine (10 mg/kg, i.p.) 30 min prior to the 60 min session to test for cue-induced reinstatement of morphine or fentanyl. During the assessment, cues previously associated with morphine or fentanyl self-administration were presented following each response on the previously active lever. Responses on the previously inactive lever were recorded but did not result in stimulus presentation.

Baseline responding (three days prior to extinction) on the active and inactive levers were compared separately for the morphine and fentanyl groups receiving either vehicle or oxa-noribogaine using a two-way ANOVA with time as the repeated measure with Šidák’s multiple comparison test for post hoc comparison. For the cue-induced reinstatement test session, responding on the active and inactive levers were analyzed for the morphine and fentanyl groups separately receiving either vehicle or oxa-noribogaine using an unpaired t-test.

Effect of acute administration of oxa-noribogaine, epi-oxa-noribogaine and noribogaine on morphine self-administration

Rats were trained to self-administer morphine (10 µg/infusion) as described in the general self-administration training. After stable responding was established, rats were administered vehicle (2 mL/kg, i.p.) 15 min prior to the subsequent experimental session. Three days following vehicle administration, rats were administered oxa-noribogaine (10 or 40 mg/kg, i.p.), epi-oxa-noribogaine (40 mg/kg, i.p.) or noribogaine (40 mg/kg, i.p.), administered 15 min prior to the beginning of the session.

Effect of acute oxa-noribogaine administration on fentanyl self-administration

Rats were trained to self-administer fentanyl (625 ng/ infusion) as described in the general self-administration training. After stable responding was established, rats were administered vehicle 15 min prior to the subsequent experimental session. Three days following vehicle administration, rats were administered oxa-noribogaine (40 mg/kg, i.p.) administered 30 min prior to the beginning of the session.

Food maintained responding

For food maintained responding studies, food access was restricted such that rats were maintained at 90% of their normal body weight until responding stabilized, at which point rats were provided 12–15 g of rodent chow in addition to the food pellets earned during the experimental session.

Rats were lever trained as described in the general methods. Responding was maintained as described with notable difference in timeout (TO) period 6 min and sessions lasting 2 h or until a maximum of 20 pellets was delivered. Once responding stabilized, subjects were assigned randomly to receive oxa-noribogaine (10 and 40 mg/kg, i.p.), epi-oxa-noribogaine (40 mg/kg, i.p.), noribogaine (10 and 40 mg/kg, i.p.) or vehicle administered intraperitoneal 15 min prior to the initiation of the session. Responding was considered stable when variation in the number of reinforcers for three consecutive sessions was less than 20%. For the group receiving oxa-noribogaine, data were analyzed using a two-way ANOVA with Šidák’s multiple comparisons test for post hoc comparison.

GDNF expression studies

Male Fisher F344 rats were assigned to groups and administered oxa-noribogaine (40 mg/kg; i.p.) or noribogaine (40 mg/kg; i.p.). One day or five days after drug administration, rats from these groups as well as non-treated controls were decapitated, brains removed and placed in a stainless-steel rat brain matrix. Coronal slices were taken and the ventral tegmental area (VTA), nucleus accumbens (NAc) and medial prefrontal cortex (mPFC) were dissected and immediately frozen on dry ice. Total protein was isolated from pulverized tissue (8–15 mg tissue per region was harvested from one subject) from these regions. GDNF contents were assayed using the BiosSensis GDNF, Rat, Rapid™ ELISA assay (Biosensis Pty Ltd, SA, Australia). Protease (Thermo Scientific, Rockford, IL), and phosphatase inhibitors (Cocktails 1 and 2, Sigma-Aldrich, St. Louis, MO) were added to the extraction buffers. Regardless of sample weight 230uL of RIPA buffer (Sigma; R0278) was added into each sample along with protease inhibitors. Samples were sonicated to lyse the tissue (horn sonicator or Diagenode Bioruptor) twice for 15–20 s and incubated on ice for 30 s between sonication. After resting on ice for 30 min, samples were centrifuged at 8,000 RPM (equivalent to 6010 x g) for 15 min at 4 °C and the supernatant (total protein lysate) was transferred to a new tube. Next 8.5 µL of 1 N HCl solution was added to each sample to obtain a pH of 3.5. After 15 min 1.6 µL of 1 N NaOH solution was added to raise the pH to 7.0–7.4. (note: it is important to check pH during acid/base addition steps and ideally practice on blank tissue samples to ensure consistent pH change and measurements; it is critical that pH does not decrease below 3.0 – the risk of permanently denaturing the target). Protein concentrations were measured using the bicinchoninic acid protein assay kit (Pierce, Rockford, IL, USA) on a spectrophotometer (iD5, Molecular Devices, Sunnyvale, CA). Aliquots of 100 μL of isolated protein from each region were transferred to a 96-well ELISA plates. The abundance of GDNF was normalized to the amount of total protein (pg/mg protein). In a separate cohort, we assessed the ability of the kappa antagonist aticaprant to attenuate the oxa-noribogaine induced increase in GDNF in the VTA and mPFC five days after administration. Rats were administered either vehicle (7% (v/v) Tween 80, 2 molar equivalents glacial acetic acid in physiologic saline) or aticaprant (1 mg/kg; s.c.) followed twenty minutes later by administration of either vehicle (2 molar equivalents glacial acetic acid in physiologic saline) or oxa-noribogaine (40 mg/kg; i.p.). Along with non-treated controls, rats were decapitated, and tissue processed as described above. Data were analyzed by protein and region using one way ANOVA with Tukey’s multiple comparison test for post hoc comparison.

Opiate induced hyperalgesia (OIH) study

Rats were randomly divided into two groups to receive either two morphine (MOR; 75 mg each) or two placebo pellets. Prior to pellet implantation, baseline measurements of paw withdrawal threshold (PWT) using Von Frey (VF) assay were taken. Sustained morphine administration was accomplished by subcutaneous implantation of two morphine pellets (75 mg each). Under light isoflurane anesthesia, an area approximately one inch above the pelvic bone was shaved, a small incision made, two morphine or two placebo pellets were implanted, and the wound was closed by suture^91,92. Steady state plasma levels of morphine were reported to be achieved 3 days post-pellet implantation and maintained through 12 days post-implantation^91,92.

On Day 8 post-implantation, morphine and placebo pellets were removed under light isoflurane anesthesia and the wound was washed with sterile saline to remove residual morphine. Two hours following pellet removal rats were administered either vehicle (2 molar equivalents acetic acid in saline) or oxa-noribogaine solution (30 mg/kg; I.P.) to obtain four separate study groups.

Paw withdrawal thresholds were recorded for each group in selected time intervals: 8 h, 24 h and 168 h post-implant removal (6, 22 and 166 h post oxa-noribogaine treatment respectively). The presence and development of nociceptive tolerance was assessed by two-way ANOVA with Tukey’s multiple comparison test for post hoc comparison.

Electronic Von Frey Assessments: Mechanical sensitivity was measured using the electronic Von Frey 5 with embedded camera (BIO-EVF5; Bioseb US, Pinellas Park, FL). Rats were placed in a modular holder cage (BIO-PVF, Bioseb US) on a wire mesh elevated stand and allowed to acclimate for 5 min prior to assessment. The spring tip of the device was applied to the plantar surface of the hind paw. Brisk paw withdrawal was considered as a positive response, the force (grams) required to initiate paw withdrawal was recorded by the EVF software. The average of a minimum of Two replicates per hind paw was calculated for each rat. The average of all values was calculated as the paw withdrawal threshold (PWT).

Pharmacokinetic studies (PK, Sai Life Sciences Limited)

Total of twenty-four male mice (twenty-seven male rats) were used per study (3 animals per each time point). Mice were administered subcutaneously (10 mg/kg dose, s.c.), rats intraperitoneally (40 mg/kg dose, i.p.). The dosing volume for subcutaneous administration was 5 mL/kg and 10 mL/kg for intraperitoneal administration.

Blood samples ( ~ 60 µL from mice, ~120 from rats) were collected under light isoflurane anesthesia (Surgivet®) from retro orbital plexus from a set of three animals at specified time points into labeled micro-tubes, containing K₂EDTA solution (20% K₂EDTA solution) as an anticoagulant. Immediately after blood collection, plasma was harvested by centrifugation at 4000 rpm (2000 x g), 10 min at 40 °C and samples were stored at −70 ± 10 °C until bioanalysis. Following blood collection, animals were immediately sacrificed, the abdominal vena-cava was cut open and whole body was perfused from heart using (10 mL for mice or 20 mL for rats) of normal saline. Brain samples were collected from a set of three animals at specified time points. After isolation, brain samples were rinsed three times in ice cold normal saline (for 5–10 s/rinse using ( ~ 5–10 mL for mice or ~10–20 mL) of normal saline in disposable petri dish for each rinse) and dried on blotting paper. Brain samples were homogenized using ice-cold phosphate buffer saline (pH − 7.4). Total homogenate volume was three times the tissue weight. All homogenates were stored below −70 ± 10 °C until bioanalysis. The extraction procedure for plasma and brain samples and the spiked plasma and brain calibration standards were identical. A 25 µL of study sample or spiked plasma calibration standard was added to individual pre-labeled micro-centrifuge tubes followed by 100 µL of internal standard prepared in acetonitrile (Glipizide, 500 ng/mL) was added except for blank, where 100 µL of acetonitrile was added. Samples were vortexed for 5 min. Samples were centrifuged for 10 min at a speed of 4000 rpm ( ~ 2000 x g) at 4 °C. Following centrifugation, 100 µL of clear supernatant was transferred in 96 well plates and the concentrations of analyte were determined by fit for purpose LC-MS/MS method. Instrument (Waters ACQUITYTM, UPLC, Canada) equipped with (API-4000 MDS Sciex, Applied Biosystems, Canada) detector.

Non-Compartmental-Analysis tool of Phoenix WinNonlin® (Version 8.0 for oxa-noribogaine, Version 7.0 for noribogaine) was used to assess the pharmacokinetic parameters. Peak plasma concentration (C_max) and time for the peak plasma concentration (T_max) were the observed values. The areas under the concentration time curve (AUC_last and AUC_inf) were calculated by linear trapezoidal rule. The terminal elimination rate constant (k_el) was determined by regression analysis of the linear terminal portion of the log plasma concentration-time curve. The terminal half-life (T_1/2z) was estimated by 0.693/k_el. Clearance (CL) was estimated as Dose/AUC_inf and steady state volume of distribution (V_ss) as CL × MRT (Mean Residence Time). Tissue-to-plasma ratio (Kps) were calculated using Microsoft Excel.

Plasma protein and brain tissue binding (Sai Life Sciences Limited)

Noribogaine, oxa- and epi-oxa-noribogaine plasma protein and brain tissue binding were determined using rapid equilibrium dialysis (RED) followed by LC-MS/MS quantification in MRM mode. Brain tissue homogenate samples were prepared by diluting one volume of whole brain tissue with three volumes of dialysis buffer (phosphate buffered saline pH 7.4 with 0.1 M sodium phosphate and 0.15 M sodium chloride) to yield 4 times diluted homogenate. A 1 mM stock solution of test compounds was prepared in DMSO and diluted 200-folds in rat plasma (male Sprague Dawley, n = 5), human plasma (drug free volunteers; n = 6) or rat brain homogenate to prepare a concentration of 5 µM. The final DMSO concentration was 0.5%. Rapid equilibrium dialysis was performed with RED device containing dialysis membrane with a molecular weight cut-off of 8000 Daltons (Thermo Scientific). A 200 µL aliquot of positive control 5 µM and test compound at 5 µM (triplicates) were separately added to the plasma/brain homogenate chamber and 350 µL of phosphate buffer saline (pH 7.4, Thermo Scientific) was added to the buffer chamber of the inserts. After sealing the RED device with an adhesive film, dialysis was performed in incubator at 37 °C with shaking at 100 RPM for 4 h.

A 50 µL aliquot of positive control and test compounds were added to four 0.5 mL microcentrifuge tubes. Two aliquots were frozen immediately (0 min sample). The other two aliquots were incubated at 37 °C for 4 h along with the RED device.

Following dialysis, an aliquot of 50 µL was removed from each well (both plasma or brain homogenate and buffer side) and diluted with equal volume of opposite matrix (dialyzed with the other matrix) to nullify the matrix effect. Similarly, 50 µL of buffer was added to recovery and stability samples. An aliquot of 100 µL was submitted for LC-MS/MS analysis. Instrument (Waters ACQUITYTM, UPLC, Canada) equipped with (API-4000 MDS Sciex, Applied Biosystems, Canada) detector.

A 25 µL aliquot of positive control and test compounds were crashed with 100 µL of acetonitrile containing internal standard and vortexed for 5 min. The samples were centrifuged at 4000 rpm ( ~ 2000 x g) at 4 °C for 10 min and 100 µL of supernatant was submitted for LC-MS/MS analysis in MRM mode. The samples were run without calibration curve and the peak area ratios (analyte versus internal standard) obtained were used to determine the fraction of compound bound to plasma and brain proteins.

Estimated free drug concentrations, target engagement, and correlation with in vitro pharmacology

Oxa-noribogaine exhibits rapid and high brain penetration (brain/plasma = 5.5, mice) after subcutaneous administration (s.c.). Approximately 85% of the compound is cleared in the first 2 h and only traces remain after 8 h. We estimated the free plasma and brain drug concentrations based on plasma protein and brain tissue binding (C_max(brain) = 166 nM, 10 mg/kg, s.c.). Assuming linearity of PK below 10 mg/kg, C_max(brain) at an ED₅₀ analgesic dose (3 mg/kg) is expected to be ~ 50 nM, which matches well the in vitro KOR activation potency (EC₅₀ = 41 nM in BRET assay, EC₅₀ = 49 nM, in [³⁵S]GTPgS assay). Thus, the pharmacodynamic readout (analgesia) correlates with the estimated free drug PK measures. Further, the C_max and area under the curve (AUC, drug exposure) values for estimated free drugs are much smaller for oxa-noribogaine in comparison to noribogaine, which is relevant for interpreting the oxa-noribogaine’s superior efficacy in addiction models.

Ibogaine analogs docking studies on KOR

The active state KOR X-ray crystal structures corresponding to PDB accession code 6B73⁹³ was extracted from the RCSB server. Three conserved crystallographic waters close of H^6.52 and Y^3.33, elucidated in the high-resolution active state MOR X-ray crystal structure (PDB 5C1M)⁹⁴ were transplanted to the active state KOR model and their steric positions inside the active state KOR model were optimized by minimization protocols. All the objects except the receptor protein subunit, the crystallized ligand, and the three transplanted waters were deleted, and this was followed by the addition of hydrogens and optimization of the side-chain residues. Ligands were sketched, assigned formal charges, and energy-optimized prior to molecular docking. The ligand docking box for potential grid docking was defined to contain the extracellular half of the protein, and all-atom docking was performed using the energy minimized structures for all ligands with an effort value of 10. The best-scored docking solutions were further optimized by iterative rounds of minimization and Monte Carlo sampling of the ligand conformation, including the surrounding side-chain residues (within 5 Å of the ligand) and the three water molecules in the KOR orthosteric site. All the above-mentioned molecular modeling operations were performed in ICM-Pro v3.9 molecular modeling and drug discovery suite (Molsoft LLC)⁹⁵.

The optimized ligand conformations inside of active state KOR presented on Fig. 1d and Supplementary Fig. 4e–g are provided as supplementary datasets 3 (noribogaine) and 4 (oxa-noribogaine).

X-ray structure determination

X-ray diffraction data were collected on a Bruker Apex II diffractometer. The structures were solved by using direct methods and standard difference map techniques and were refined by full-matrix least-squares procedures on F2 with SHELXTL (Version 2014/7)^96,97,98. Crystallographic data have been deposited with the Cambridge Crystallographic Data Centre (CCDC #2215015).

The X-ray structure and checkcif report are provided as supplementary datasets 1 and 2.

Synthesis

Reagents and solvents (including anhydrous solvents) were obtained from commercial sources and used without further purification unless otherwise stated. All compounds, except noribogaine, were prepared in racemic form. All reactions were performed in oven- or flame-dried glassware under an argon atmosphere unless otherwise stated and monitored by TLC using solvent mixtures appropriate to each reaction. All column chromatography was performed on silica gel (40–63 μm). Preparative TLC was performed on 20 × 20 cm plates with a 1 mm silica layer. For compounds containing basic nitrogen, triethylamine (Et₃N) was often used in the mobile phase to provide better resolution. In these cases, TLC plates were pre-soaked in the Et₃N containing solvent and then briefly dried before use, to yield an accurate R_f value. Nuclear magnetic resonance spectra were recorded on Bruker 400 or 500 MHz instruments as indicated, using Bruker TopSpin (3.5 pl6) software. Spectra were further analyzed using MestReNova 14. Chemical shifts are reported as δ values in ppm referenced to CDCl₃ (¹H NMR = 7.26 and ¹³C NMR = 77.16). Multiplicity is indicated as follows: s (singlet); d (doublet); t (triplet); q (quartet); p (pentet); dd (doublet of doublets); ddd (doublet of doublet of doublets); dt (doublet of triplets); td (triplet of doublets); m (multiplet); br (broad). For compounds containing a carbamate group, complex spectra with split peaks were observed (presence of conformers). As a result of these effects, multiple peaks may correspond to the same proton group or carbon atom. When possible, this is indicated by an “and” joining two peaks or spectral regions. Alternatively, when certain carbon peaks overlap and thus represent two carbons they were indicated by (2 C) designation. For easier identification of carbamate containing compounds on the included ¹H NMR spectra, groups of peaks arising from the hindered rotation were integrated together to obtain whole number integrals. All carbon peaks are rounded to one decimal place. Low-resolution mass spectra were recorded on either a JEOL LCmate (ionization mode: APCI+ ) or Advion expression-L CMS quadrupole mass spectrometer equipped with ESI and APCI sources. High-resolution mass spectra (HRMS) were acquired on a high-resolution Waters XEVO G2-XS QToF mass spectrometer equipped with a UPC2 SFC inlet, on-board fluidics and an ESI probe.

Synthetic conditions and compound characteristics are provided in the Supplementary information file.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.