Oxa-Iboga alkaloids lack cardiac risk and disrupt opioid use in animal models

Abstract

Ibogaine and its main metabolite noribogaine provide important molecular prototypes for markedly different treatment of substance use disorders and co-morbid mental health illnesses. However, these compounds present a cardiac safety risk and a highly complex molecular mechanism. We introduce a class of iboga alkaloids - termed oxa-iboga - defined as benzofuran-containing iboga analogs and created via structural editing of the iboga skeleton. The oxa-iboga compounds lack the proarrhythmic adverse effects of ibogaine and noribogaine in primary human cardiomyocytes and show superior efficacy in animal models of opioid use disorder in male rats. They act as potent kappa opioid receptor agonists in vitro and in vivo, but exhibit atypical behavioral features compared to standard kappa opioid agonists. Oxa-noribogaine induces long-lasting suppression of morphine, heroin, and fentanyl intake after a single dose or a short treatment regimen, reversal of persistent opioid-induced hyperalgesia, and suppression of opioid drug seeking in rodent relapse models. As such, oxa-iboga compounds represent mechanistically distinct iboga analogs with therapeutic potential.

Fig. 1. Oxa-iboga is a distinct class of iboga alkaloids discovered by structural editing of ibogaine, enabled by efficient de novo chemical synthesis.

a Ibogaine is a substance with broad therapeutic effects and complex pharmacology, that is distinct from classical psychedelic tryptamines. Relative potencies at known molecular targets are shown. b Oxa-iboga analogs are defined by the replacement of indole with benzofuran, resulting in accentuation of the KOR activity on the iboga pharmacological background. c De novo synthesis of iboga molecular framework rests on the catalytic union of the two main structural components of oxa-iboga skeleton, the isoquinuclidine and benzofuran ring systems. d Docking pose of noribogaine (carbon frame in green) in sticks representation inside KOR structure (active receptor state). Hydrogen bonding near the C10 phenol and tertiary amine are highlighted by yellow dashed lines. KOR (kappa opioid receptor), α3β4 (α3β4 nicotinic acetylcholine receptors), SERT (serotonin transporter), 5-HT (serotonin), NMDAR (N-methyl-D-aspartate receptor).

Fig. 2. KOR dependent upstream and downstream molecular signaling pathway effects of oxa-iboga compounds.

a Illustration of the KOR-induced molecular signaling pathway assays in vitro and in vivo. The location and shape of medial prefrontal cortex (mPFC), nucleus accumbens (NAc) and ventral tegmental area (VTA) are highlighted in a rodent brain - representative illustration of a left hemisphere (sagittal slice, mouse, Allen Institute). b Oxa-iboga compounds are agonists of rat KOR in vitro, as demonstrated by a G protein activation BRET assay. c Oxa-noribogaine displays signaling efficacy across Gα isoforms, efficacy compared to U50,488 in the TRUPATH BRET assay. Pooled data (n = 3 biological replicates) for each subunit were analyzed using a nonlinear regression and the calculated span ± SEM parameters are presented (dose response curves Supplementary Fig. S6). d In the nanobody Nb33 sensor recruitment assay, which approximates true intrinsic signaling efficacy, oxa-noribogaine analogs are partial agonists. e Additionally, markedly reduced signaling efficacy for oxa-noriboga analogs was detected in the β-arrestin2 recruitment assay (KOR results on panels b, d, e are presented as mean ± SEM, n = 3 biological replicates). f Administration of oxa-noribogaine (40 mg/kg; i.p.) significantly increased GDNF protein levels in the mPFC and VTA after 5 days (OXA5). Pre-treatment of rats with KOR selective antagonist aticaprant (ATI, 1 mg/kg s.c.) before oxa-noribogaine administration prevents the increase of GDNF expression in mPFC and VTA. mPFC: CONTROL - OXA5 (P < 0.0001), VEH + VEH - OXA5 (P = 0.0078), VEH + ATI - OXA5 (P = 0.0123), OXA5 - ATI + OXA5 (P = 0.0022) and VTA: CONTROL - OXA5 (P = 0.0010), VEH + VEH - OXA5 (P = 0.0004), VEH + ATI - OXA5 (P < 0.0001), OXA5 - ATI + OXA5 (P = 0.0001). Experiments were conducted using separate groups of animals (Control n = 7, all other groups n = 8 subjects). Data are presented as mean ± SEM. Statistical test used: One-way ANOVA with Tukey’s multiple comparisons test, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. Source data are provided in the Source Data file.

Fig. 3. Oxa-noribogaine is an atypical KOR agonist in vivo with no aversion and no pro-depression-like effects at efficacious analgesic doses.

a Oxa-iboga analogs induce potent analgesia in the mouse tail-flick test (male mice), comparable in potency and efficacy to the standard kappa agonist, U50,488. b Analgesia of oxa-noribogaine and epi-oxa-noribogaine (c) is KOR dependent as demonstrated in KOR knock-out mice (KOR-KO), compared to mu receptor knock-out (MOR-KO) and wild type (WT) mice (KOR-KO vs WT, P < 0.0001). d Oxa-noribogaine induces potent analgesia in female mice, primarily driven by KOR as demonstrated in e female KOR-KO mice compared to MOR-KO and WT. f Traces visualizing ambulatory distance traveled by WT mice in open field test (OF) show different effects by the isomers at equianalgesic doses. g Quantification of OF test (ED₈₀ doses) for oxa-noribogaine (5.4 mg/kg, both male and female) and epi-oxa-noribogaine (5.2 mg/kg), data are normalized to initial locomotion of vehicle group. h Sedation of epi-oxa-noribogaine (5.2 mg/kg, P < 0.0001) is KOR-driven as demonstrated by pre-treatment (P < 0.0001) by the selective KOR antagonist aticaprant (ATI, 0.1 mg/kg). Total locomotion over 60 min period is normalized to vehicle. i No pro-depressive-like effects were detected using the forced swim test after oxa-noribogaine administration (30 min post administration). Imipramine was used as a positive effect control (P = 0.0031). j Male and female mice do not develop a conditioned place preference or aversion (CPP/CPA) after administration of oxa-noribogaine (P = 0.4531, P = 0.5994, respectively), in contrast to cocaine (10 mg/kg: P = 0.0029 and 0.0013) and morphine (20 mg/kg: P = 0.0010 and 0.0153). k Pharmacokinetic distribution of oxa-noribogaine in mice plasma and brain tissue (10 mg/kg). All drugs were administered via subcutaneous (s.c.) route except for intraperitoneal (i.p.) route for CPP/CPA test. % MPE (percentage of maximum potential effect). Data are presented as mean ± SEM. Statistical tests used: Unpaired (h, i) and paired (j) t-test, two-tailed, One-way (i) and Two-way (g, j) ANOVA with Šidák multiple comparisons test, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. Source data are provided in the Source Data file.

Fig. 4. Oxa-iboga analogs do not show pro-arrhythmia risk in adult human primary cardiomyocytes.

a Human heart is digested to isolate single cardiomyocytes, which are field stimulated to produce a regular pattern of contractility transients. These cells are phenotypically stable and provide a robust preclinical assay with high translational validity. b Representative traces capturing contraction-induced change in sarcomere length after administration of vehicle, noribogaine and oxa-noribogaine solutions. c Noribogaine demonstrates pro-arrhythmic potential by causing after-contractions and contraction failures in cardiomyocytes as quantified in the plot, providing validation of this assay for the iboga compounds. No pro-arrhythmic potential was detected for oxa- or epi-oxa-noribogaine (for donor information and replicates see Supplementary Fig. S8). Source data are provided in the Source Data file. Alternans % (percentage of repetitive alternating short and long contractility amplitude transients), STV % (Short-Term Variability in percentage).

Fig. 5. Oxa-noribogaine induces acute and long-lasting suppression of morphine, heroin, and fentanyl self-administration in rats.

a A schematic depiction of the experimental treatment paradigm of opioid use (male rats used in this study). b Oxa-noribogaine (40 mg/kg) is more efficacious than noribogaine (40 mg/kg) in suppressing morphine self-administration (P = 0.0001, seven sessions). One injection of oxa-noribogaine (40 mg/kg) results in statistically significant suppression of morphine intake for 7 days. The effect on morphine self-administration was dose dependent. c Food operant intake (natural reward) is reduced following administration of 40 mg/kg but not 10 mg/kg (P < 0.0001). The moderate dose (10 mg/kg) has a marginal effect on food intake (all study groups n = 7). d One injection of oxa-noribogaine (40 mg/kg, n = 9) results in statistically significant suppression of fentanyl intake for 4 days (comparison with morphine data from Fig. 5b). e Visualization of acute (Day 1) and post-acute (Day 5) dose-dependent effects of oxa-noribogaine (3 and 10 mg/kg) on morphine dose-effect function in morphine self-administration (10 µg/inf, n = 8; 20 µg/inf, n = 10; 40 µg/inf, n = 8). On Day 1, oxa-noribogaine dose dependently decreases self-administration of morphine 20 µg/inf (oxa-noribogaine doses, 3 mg/kg: P = 0.0419 and 10 mg/kg: P < 0.0001) and 40 µg/inf (10 mg/kg: P = 0.0327), and intake continued to be significantly decreased on Day 5 for 20 µg/inf (Oxa-noribogaine doses, 3 mg/kg: P = 0.0243 and 10 mg/kg: P = 0.0044). f A temporal profile across all sessions of the dose dependent effects of oxa-noribogaine in the rat cohort self-administering 20 µg/inf (n = 10) of morphine (active lever – shades of blue, inactive – grey). Control parallel cohort received injections of vehicle (n = 5) solution. g Repeated escalating dosing of oxa-noribogaine (3, 10 and 30 mg/kg; n = 11) enable acute suppression of heroin (4.5 µg/infusion) intake (10 mg/kg, session 8: P = 0.0012 and 30 mg/kg, session 15: P < 0.0001) and the long term suppression effect propagates up to 7 days post last intervention (session 16: P = 0.0009, session 18: P = 0.0137 and session 21: P = 0.0234). All drugs were administered via intraperitoneal (i.p.) route. Data are presented as mean ± SEM. Statistical tests used: Unpaired (b, g) and Paired (c, e) t-test, two-tailed, One-way (c) and Two-way (b, c, d, e, g), repeated measures ANOVA or Mixed Model with Šidák multiple comparisons test *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. Source data are provided in the Source Data file.

Fig. 6. Acute suppression of reinstatement to opioid responding and long-lasting alleviation of opioid-induced hyperalgesia in rats by a single dose of oxa-noribogaine.

a A single dose of oxa-noribogaine (10 mg/kg, i.p.) reduces cue-induced fentanyl (P = 0.0141, n = 8) and morphine (P = 0.042, n = 5) seeking. b Robust OIH response resulting from a chronic exposure to morphine, measured as paw withdrawal threshold (PWT; Von Frey assay), persisting for up to 7 days (Placebo-Veh vs. Morphine-Veh: 8 h P = 0.004, 24 h P = 0.0258 and 168 h P = 0.0157) is suppressed by a single dose of oxa-noribogaine (30 mg/kg, n = 8), Placebo-Oxa vs. Morphine-Veh (8 h P = 0.0127, 24 h P = 0.0305 and 168 h P = 0.0003) and Morphine-Veh vs. Morphine-OXA (8 h P = 0.0164). Data are presented as mean ± SEM. Statistical tests used: Unpaired t-test, two-tailed (a) and Two-way repeated measures ANOVA with Tukey’s multiple comparisons test (b), *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. Source data are provided in the Source Data file.