This page respects the original work's rights terms and independently introduces its bibliography, abstract, main results, and limitations. It is not a full translation.
About this resource
This is not a full-text translation of the original paper, but a detailed introduction based on the publicly available author manuscript and bibliographic information. The study compound, tabernanthalog (TBG), is an analogue designed from the structure and pharmacology of ibogaine; it is not ibogaine itself. Findings about TBG cannot be treated directly as evidence of ibogaine’s efficacy or safety.
Rather than preserving the full structure of a complex natural product, the study used function-oriented synthesis to design candidates around elements related to neurite and dendrite growth. The preclinical question was whether neuroplasticity and behavioral activity could be retained while reducing hallucination-like effects, hERG inhibition, cardiac abnormalities, and difficulties in synthesis and formulation associated with ibogaine. The terms “non-hallucinogenic” and “safer” must both be interpreted within the cell, zebrafish, and rodent tests used here.
Study design
The study spanned compound design, cellular pharmacology, early safety, neuroplasticity, and animal behavior. Dendritic growth and spine density were measured in cortical neurons derived from rat embryos; hERG-channel inhibition was assessed in HEK293 cells; and functional activity was examined at human and mouse serotonin and other receptors. Zebrafish larvae were used to assess heart rate and arrhythmia-like indicators, behavioral profiles in response to stimuli, developmental toxicity, and neural activity.
In mice, the head-twitch response (HTR) served as a behavioral proxy for hallucination-like activity. The researchers also used conditioned place preference, two-photon imaging of cortical spine dynamics, the forced-swim test, and intermittent access to 20% ethanol in a two-bottle choice. Rat experiments used heroin self-administration, extinction, and cue-induced reinstatement, with sucrose self-administration as a comparison for nondrug reward. The Methods state that treatment was randomized, analysts were blinded to treatment condition, and power analyses were conducted for animal experiments.
Content
Compound, receptor, and safety indicators
Structure–activity studies selected TBG, a simplified portion of ibogaine that could be synthesized in one step, as a candidate. TBG functioned as an agonist at the 5-HT2A receptor and as an antagonist at 5-HT2B. Screening across 81 targets showed relatively high selectivity for the 5-HT2 receptor family. These are observations of target pharmacology, not demonstrations of clinical efficacy.
The positive control 5-MeO-DMT produced a strong HTR in mice, while no response was detected with TBG. TBG inhibited hERG approximately 100 times less strongly than ibogaine; the ibogaine IC50 reported in the paper was 1 μM. In zebrafish, the reduced heart rate, arrhythmia-like findings, developmental abnormalities, and deaths seen with ibogaine were not clearly reproduced with TBG, and there were fewer nonsurviving animals in the 100 μM TBG group than in the ibogaine group. At 66 μM, the alive-to-dead ratio after five days was not statistically distinguishable from vehicle control.
Plasticity and behavior
In cultured rat cortical neurons, TBG increased dendritic complexity and spine density in mature cultures; dendritic growth was blocked by the 5-HT2A antagonist ketanserin. Two-photon imaging of living mouse cortex showed increased spine formation 24 hours after administration, with no change in the rate of spine loss. This is an observation of structural change; it does not establish that the change caused a particular behavioral improvement.
Results
In the mouse forced-swim test after seven days of unpredictable mild stress, 50 mg/kg TBG reduced the elevated immobility time, while 10 mg/kg did not produce the same result. In a comparison without the stress procedure, both TBG and ketamine reduced immobility 24 hours after administration, but the effect of TBG appeared less persistent than ketamine’s. Ketanserin pretreatment blocked the TBG-associated behavioral change, and spontaneous locomotion did not clearly decrease at 24 hours. This is an “antidepressant-like” behavioral indicator, not a treatment outcome for human depression.
In mice given intermittent ethanol access for seven weeks, pre-administration intake was 11.44 ± 0.76 g/kg/24 hours and 3.89 ± 0.33 g/kg during the first four hours. After TBG was administered three hours before a drinking session, ethanol intake and preference during the first four hours decreased, with the reduction observed for at least 48 hours. Water intake did not decrease, and in a separate 5% sucrose two-bottle choice, neither sucrose preference nor total fluid volume decreased.
In the rat heroin task, 40 mg/kg TBG was given during self-administration, on the first extinction day, or immediately before cue-induced reinstatement. Acute administration reduced heroin seeking, but strongly reduced sucrose self-administration at the same time, so the acute effect may have included nonspecific disruption of operant responding. In contrast, rats that received TBG once 12–14 days before the reinstatement test responded less to heroin cues without reduced responding to sucrose cues. The authors discussed the possibility of a lasting reduction in drug seeking, while leaving optimal dose, timeline, and mechanism for future work.
Limitations
All findings came from cells, fish, or rodents; administration, safety, subjective experience, and efficacy by diagnosis were not studied in humans. HTR is only a proxy for hallucination-like activity, and the Methods describe a small assessment of two male and two female mice per treatment. An absent HTR does not guarantee that perceptual or cognitive changes will not occur in people. hERG testing, zebrafish heart rate, and the 81-target panel are important early screens, but they do not cover metabolites, repeated exposure, organ toxicity, or drug interactions.
Immobility in the forced-swim test is a limited construct and not equivalent to human depressive symptoms. Ethanol two-bottle choice and heroin-cue tests also differ from human substance use disorders embedded in social environments. In particular, sucrose responding was suppressed during the acute phase of the heroin task, weakening a simple explanation in terms of a selective drug effect. A causal relationship between neuroplasticity and persistent behavior was not established through direct manipulation.
The corresponding author disclosed serving as president and chief scientific officer of Delix Therapeutics, which licensed TBG-related technology, and the company funded the large-scale receptor screening. This does not invalidate the findings, but it heightens the importance of independent replication and subsequent safety assessment.
Safety
The study showed relative, early safety findings in selected test systems: TBG produced weaker hERG inhibition and fewer zebrafish abnormalities than ibogaine, and did not produce an HTR in mice. “Nontoxic” and “non-hallucinogenic” must not be generalized to humans. In high-dose place conditioning, 1 mg/kg did not change preference, while mild place aversion was observed at 10 and 50 mg/kg, also showing that behavioral effects vary with dose.
TBG is a distinct research compound, and the study does not show that it can be self-administered as a substitute for ibogaine. Human pharmacokinetics, ECG, blood pressure, psychiatric symptoms, liver and kidney function, repeated dosing, and interactions with concomitant medications cannot be determined from this research. “Safer” in preclinical tests is not a clinical safety guarantee or treatment recommendation.
Source and rights
The original source can be checked through the author manuscript in PubMed Central, the PubMed record, and the DOI and publisher page. It was published in Nature in 2021, volume 589, pages 474–479 (first published online in December 2020).
PMC hosts an author manuscript. The displayed Springer Nature terms concern viewing, printing, copying, downloading, and text and data mining for scholarly research; they are not a Creative Commons license clearly permitting public distribution of adaptations. This page is limited to a detailed introduction using brief paraphrase and critical organization. It does not reproduce the full text, figures, or consecutive passages as a substitute for the original. Readers using exact figure values or supplemental conditions should consult the source and publisher terms directly.