The effect of Banisteriopsis caapi (B. caapi) on the motor deficits in the MPTP-treated common marmoset model of Parkinson’s disease

Parkinson's disease (PD) is a progressive neurodegenerative disorder characterised principally by loss of dopaminergic neurons in the substantia nigra and resulting motor impairments. Symptomatic treatment centres on dopamine replacement, most commonly with L-3,4-dihydroxyphenylalanine (L-DOPA), but chronic L-DOPA use is limited by waning efficacy and the emergence of involuntary movements (dyskinesia). Monoamine oxidase B (MAO-B) inhibitors such as selegiline are used to prolong L-DOPA effects by reducing dopamine metabolism. Banisteriopsis caapi (B. caapi), a plant used traditionally in Amazonian preparations, contains β-carbolines (notably harmine, harmaline, and tetrahydroharmine) with monoamine oxidase inhibitory properties and some clinical reports suggest antiparkinsonian effects when taken as a tea, but controlled preclinical characterisation is limited. Fisher and colleagues set out to determine whether an aqueous extract of B. caapi and one of its principal β-carbolines, harmine, alter motor deficits in a primate model of PD. Using MPTP-treated common marmosets that had been primed with prior L-DOPA to exhibit dyskinesia, the investigators evaluated B. caapi extract, harmine, and the selective MAO-B inhibitor selegiline as monotherapies and in combination with submaximal doses of L-DOPA. The study aimed to establish whether B. caapi or harmine produce antiparkinsonian effects, whether these effects are consistent with MAO inhibition, and whether they modify the motor response or dyskinesia induced by L-DOPA.

The study used a within-subjects design in a single cohort of previously MPTP-treated common marmosets (Callithrix jacchus; n = 8, aged 3–10 years) that had been primed with L-DOPA to express dyskinesia. Animals had received MPTP 1–4 years before the experiments and had undergone washout of at least 4 weeks since any previous drug testing. Treatments were administered in a modified Latin square so that each animal received each treatment on a single occasion; experiments occurred twice weekly with at least 72 hr between treatments. For the selegiline crossover experiment, a 1-week washout was used due to selegiline's long half-life. Observers were blinded to treatment during behavioural scoring. An aqueous extract of the dried B. caapi stem was prepared by two sequential 1-hr water refluxes, filtration, freeze-drying, reconstitution in methanol and HPLC analysis. High-performance liquid chromatography with diode-array detection characterised the extract; reported concentrations in the freeze-dried extract were 35.2 mg/g harmine, 17.6 mg/g tetrahydroharmine and no detectable harmaline. Drug treatments comprised oral B. caapi extract (28.4–113.6 mg/kg, po), subcutaneous harmine (0.1 and 0.3 mg/kg, sc), and subcutaneous selegiline (10 mg/kg, sc), each given alone or combined with submaximal oral L-DOPA methyl ester (4 or 7 mg/kg, po) administered together with the peripheral DOPA-decarboxylase inhibitor benserazide (10 mg/kg po). Domperidone (2 mg/kg po) was given 1 hr prior to treatments to prevent vomiting where indicated. Initial dosing included a low B. caapi dose equivalent to harmine 0.3 mg/kg given in the home cage; no effect was noted, informing the choice of higher oral doses on test days. Behavioural testing took place in individual automated test units after 60 min acclimatisation. Baseline and post-dose assessments (every 30 min for up to 5–6 hr) included automated locomotor activity and blinded observer ratings of motor disability and dyskinesia. Motor disability was scored on an established scale combining alertness, checking, posture, balance, reaction to stimuli, vocalisation and motility to produce a summed score (0 = no disability, 18 = severe disability). Dyskinesia was rated on a 0–4 scale (0 = absent to 4 = severe, continuous). Data were summarised as median time courses and area under the curve (AUC) over 5 hr (trapezoid method). Motor disability and dyskinesia scores were square-root transformed prior to analysis. Statistical testing used repeated measures one-way analysis of variance with Newman–Keuls post hoc tests; significance was set at p < .05.

HPLC analysis of the prepared B. caapi extract identified harmine (35.2 mg/g extract) and tetrahydroharmine (17.6 mg/g extract) but not harmaline. All animals were responsive to L-DOPA, and a submaximal dose of L-DOPA (7 mg/kg, po) produced significant reversal of motor disability, increased locomotor activity and induced moderate to severe dyskinesia, confirming the primed MPTP model. B. caapi extract given alone produced a dose-dependent, U-shaped effect on motor disability with a maximal, moderate reversal at 56.8 mg/kg; AUC analyses showed significant improvement versus vehicle at both 56.8 and 113.6 mg/kg. Across doses (28.4–113.6 mg/kg), B. caapi did not increase locomotor activity and did not induce dyskinesia. When co-administered with a submaximal L-DOPA dose (7 mg/kg, po), B. caapi did not significantly alter the L-DOPA-induced reversal of motor disability, and although locomotor activity tended to decline with higher B. caapi dose, this change was not statistically significant. The lowest B. caapi dose (28.4 mg/kg) produced a small, non-significant reduction in total L-DOPA-induced dyskinesia score. Harmine alone at the higher dose (0.3 mg/kg sc) produced a modest but statistically significant reversal of motor disability lasting about 2 hr, whereas harmine did not alter locomotor activity at either dose tested (0.1 and 0.3 mg/kg). Dyskinesia scores after harmine alone were not significantly different from vehicle, although three animals displayed transient abnormal movements and two animals experienced worsened existing action tremor. In combination with submaximal L-DOPA (4 mg/kg, po), harmine did not significantly change the L-DOPA-induced reversal of motor disability or increase in locomotor activity; peak-dose dyskinesia tended to be higher with harmine but this was not statistically significant. Selegiline (10 mg/kg sc) given alone produced a reversal of motor disability and an increase in locomotor activity similar in magnitude to the submaximal L-DOPA (4 mg/kg, po), though with a reduced peak effect; motor disability was significantly improved versus vehicle. When administered prior to L-DOPA (4 mg/kg), selegiline did not alter the overall L-DOPA-induced motor disability score or locomotor activity but extended the duration of improved motor disability (scores < 8) to about 210 min versus 150 min with L-DOPA alone. Selegiline alone produced fleeting low-grade dyskinesia in some animals but significantly reduced the peak dyskinesia induced by L-DOPA.

Fisher and colleagues interpret their findings as evidence that B. caapi extract and its constituent harmine exert a modest antiparkinsonian effect in the MPTP-treated marmoset when given as monotherapy. The extract improved motor disability without increasing locomotor activity or eliciting dyskinesia, a profile that differs from L-DOPA and may be of interest for early-stage symptomatic treatment. However, B. caapi did not potentiate the antiparkinsonian effect of submaximal L-DOPA nor substantially reduce L-DOPA-induced dyskinesia when co-administered. The authors note that harmine reproduced some of the extract's effects, consistent with a contribution from β-carbolines. HPLC data showed harmine and tetrahydroharmine present in the tea-derived extract but no detectable harmaline; the absence of harmaline may reflect extraction-method differences. Because harmine is a relatively selective MAO-A inhibitor and has been reported to increase dopamine efflux, the investigators propose that direct facilitation of dopamine release rather than MAO-B inhibition is a plausible mechanism for the observed motor benefit. Comparison with selegiline—a selective MAO-B inhibitor—supports this view: selegiline produced a more marked reversal of disability and an increase in locomotion than B. caapi or harmine, suggesting that the latter do not exert strong MAO-B inhibition at the doses used. Selegiline extended the duration of L-DOPA's effect and reduced peak dyskinesia, findings the authors note were somewhat unexpected in the primate model and not entirely concordant with prior animal literature, though they align with clinical reports of MAO-B inhibitors extending L-DOPA benefit. The investigators acknowledge limitations and uncertainties: the extent of central MAO inhibition achieved by the doses of harmine and extract is not directly measured, the selectivity between MAO isoforms may be critical, and other pharmacological actions cannot be excluded. They also flag variability in extraction and analytic methods as a factor influencing constituent profiles. Finally, the authors point out clinical relevance: the lowest effective extract dose approximated a human-consumable tea volume and aligns with previous clinical observations of single-dose benefit, supporting the idea that B. caapi or harmine might offer mild symptomatic benefit in early PD but are unlikely to provide additive advantages when used alongside L-DOPA in later disease stages.