Activities of extract and constituents of Banisteriopsis caapi relevant to parkinsonism

Schwarz and colleagues situate their study in the context of both traditional and more recent therapeutic interest in Banisteriopsis caapi, a principal botanical component of the ayahuasca brew. Earlier work has shown that B. caapi contains beta-carboline alkaloids such as harmine and harmaline that inhibit monoamine oxidases (MAOs), enzymes that would otherwise degrade monoamine neurotransmitters and compounds such as N,N-dimethyltryptamine. Historical, and more recent, anecdotal and small clinical reports have suggested symptomatic benefit from B. caapi extracts in patients with Parkinson's disease (PD), a disorder characterised by striatal dopamine deficiency for which current treatments such as L-DOPA can produce long-term complications including dyskinesia and motor fluctuations. This study therefore set out to examine in vitro activities of an aqueous stem extract of B. caapi and two of its major constituents, harmine and harmaline, that are relevant to Parkinsonism. Specifically, the investigators tested MAO-A and MAO-B inhibitory activity in rat liver homogenates and assessed the ability of the extract and the alkaloids to stimulate release of [3H]dopamine from rat striatal slices, with the aim of determining whether these pharmacological properties could provide a biochemical basis for the reported clinical effects.

Plant material consisted of dried B. caapi stem collected for clinical use; 50 g of powdered stem was extracted by two successive 1 h refluxes in water, the combined filtrates were freeze-dried and yielded 10.7 g of extract. Thin-layer chromatography (TLC) with densitometric scanning after Dragendorff staining was used to quantify harmine and harmaline in the extract. From a 200 mg sample applied to TLC, harmine was measured as 3.8 mg (approximately 1.9% wt/wt in the lyophilised extract) and harmaline as 0.22% wt/wt; the authors also report a comparable value of 19.2 mg harmine per gram of lyophilised extract from other samples. For MAO assays, male Wistar rats (about 190 g) provided livers that were homogenised in phosphate-buffered saline and stored at À70 °C until use. On assay day homogenates were diluted to a final concentration of 0.5% wt/vol. MAO-A activity was measured using [14C]5-hydroxytryptamine ([14C]5-HT) as substrate and MAO-B activity using [14C]phenylethylamine ([14C]PEA). Test substances (B. caapi extract, harmine, harmaline where available) and reference inhibitors (clorgyline for MAO-A and selegiline for MAO-B) were preincubated with homogenate for 30 min at 37 °C; incubations with substrate followed for 30 min, reactions were stopped with citric acid and deaminated metabolites extracted into organic solvent and quantified by liquid scintillation. Reported mean extraction efficiencies were 54% for 5-HIAA and 62% for phenylacetic acid (PAA). Enzyme activity was converted from d.p.m. to moles product per mg tissue per 30 min, expressed as percent of control, and IC50 values (concentration producing 50% inhibition) calculated from concentration–response data. Assays were performed in triplicate on four to six separate homogenates. To measure effects on dopamine release, male Wistar rats (280–320 g) provided brains for paired striata that were sliced (300 µm). Slices were preincubated and loaded with [3H]dopamine ([3H]DA, 50 nM) and then superfused; superfusate fractions were collected every 5 min and tritium counted. After establishing basal efflux over six fractions, test compounds were introduced into the superfusion medium for 60 min and fractional release expressed relative to total radioactivity remaining in slices. Experiments were performed in duplicate on four occasions and analysed by two-way ANOVA for time and treatment, with one-way ANOVA and Dunnett's post-test where appropriate.

In the MAO assays, reference inhibitors behaved as expected: clorgyline produced concentration-dependent inhibition of MAO-A with an IC50 of 0.66 nM and minimal inhibition of MAO-B (maximal 25% at 10À6 M). Selegiline produced concentration-dependent MAO-B inhibition with an IC50 of 5.0 nM and some inhibition of MAO-A only at high concentrations. B. caapi aqueous extract inhibited MAO-A in a concentration-related manner with an IC50 of 1.24 × 10À6 g/ml and produced little effect on MAO-B (maximal inhibition ~15% between 10À5 and 10À6 g/ml). Harmine inhibited MAO-A with an IC50 of 4.54 × 10À9 M and had little effect on MAO-B (maximal ~20% inhibition at 10À6 M). The authors note that the harmine IC50 is broadly consistent with prior reports. In the striatal slice experiments, superfusion with B. caapi extract at 2.5 mg/ml produced an immediate and substantial increase in [3H]DA release compared with control; a lower concentration (0.25 mg/ml) produced a trend that did not reach significance. Harmine increased [3H]DA release only at a high concentration (reported as 200 mM in the text) and not at lower concentrations. Harmaline increased DA release at both 58.3 mM and 5.83 mM; however, the total amount of [3H]DA released with harmaline was considerably less than that evoked by harmine at the higher concentration. The authors calculated that, based on their TLC quantification, a 2.5 mg/ml solution of the extract would contain about 20.1 mM harmine and 5.83 mM harmaline. They emphasise that these equivalent concentrations of the two alkaloids, when tested singly, did not reproduce the magnitude of the DA release produced by the whole extract. Additional reported quantitative details include that the TLC-derived harmine content (approximately 19.2 mg/g lyophilised extract) was similar to values reported previously (mean 23.8 mg/g).

The investigators interpret their MAO findings as showing that both the B. caapi extract and harmine preferentially inhibit MAO-A rather than MAO-B, a profile similar to clorgyline rather than selegiline. Given that MAO inhibition would be expected to reduce dopamine metabolism and thus raise central dopamine levels, this action could contribute to symptomatic benefit in Parkinsonism if the active compounds are absorbed and cross the blood–brain barrier. However, the MAO-A inhibitory activity of the whole extract appears disproportionately large compared with the amount of harmine present: the concentration of harmine corresponding to the extract IC50 would be at least two orders of magnitude lower than the IC50 of harmine alone. The authors therefore suggest either synergistic interactions among extract constituents or the presence of an as-yet unidentified, more potent MAO-A inhibitor in the extract. Regarding dopamine release, the extract, harmine and harmaline all increased [3H]DA release from striatal slices, with the extract producing a larger effect than would be expected from the measured content of harmine and harmaline. Harmaline showed activity at lower concentrations than harmine in some assays, though harmine produced greater total release at high concentration. The discrepancy between the effects of the crude extract and those of the isolated alkaloids again led the authors to propose synergism or additional active constituents. They also raise a possible mechanistic role for beta-carboline-sensitive receptors, noting that related compounds such as harman can stimulate dopamine release in tissues rich in these receptors. The authors acknowledge important uncertainties: the present work is in vitro and therefore cannot on its own demonstrate that the alkaloids reach effective concentrations in the CNS after oral dosing; the identity of any additional active constituents remains to be determined; and longer-term studies are required to assess whether chronic administration would produce dyskinesias similar to those seen with prolonged L-DOPA therapy.

Schwarz and colleagues conclude that an aqueous stem extract of B. caapi exhibits two pharmacological activities relevant to Parkinson's disease: preferential inhibition of MAO-A and stimulation of dopamine release from striatal tissue. Either or both mechanisms could, in principle, raise central dopamine levels and help alleviate Parkinsonian symptoms if the active constituents are bioavailable and cross the blood–brain barrier. The authors stress that the extract's effects exceed those predicted from the measured amounts of harmine and harmaline alone, implying synergism or additional active compounds, and they call for further work to identify these constituents, to confirm CNS penetration, and to evaluate long-term effects including the potential for dyskinesia.