4-Bromo-2,5-dimethoxyphenethylamine (2C-B) and structurally related phenylethylamines are potent 5-HT2A receptor antagonists in Xenopus laevis oocytes

Earlier research links the subjective and behavioural effects of many indolylalkylamine and phenylethylamine psychostimulants to activation of central 5-HT2 receptors, with particular emphasis on 5-HT2A as a key CNS target. Radioligand and functional assays have generally associated hallucinogenic potency with 5-HT2 receptor affinity and efficacy, and phenylisopropylamines (PIAs) that are full or partial 5-HT2A/2C agonists have often been taken as models for the mechanism of action of related drugs. However, prior work in Xenopus laevis oocytes suggested that a set of 2,5-dimethoxy-4-substituted phenylethylamines (PEAs) show low or negligible intrinsic efficacy at 5-HT2A receptors, raising the possibility that some PEAs might instead behave as antagonists at that subtype. Villalobos and colleagues designed the present study to test that hypothesis directly. They expressed rat 5-HT2A and 5-HT2C receptors in X. laevis oocytes and examined how a series of PEAs (2C-I, 2C-B, 2C-D and 2C-H) affected 5-HT-evoked currents, with the aim of determining whether these compounds antagonise 5-HT2A responses, whether the action is subtype-selective, and how potency varies with the C(4) substituent.

Stage V–VI X. laevis oocytes were prepared, defolliculated and microinjected intracytoplasmatically with 10–20 ng of cRNA encoding either the rat 5-HT2A or 5-HT2C receptor. Following 36–48 h incubation at 15°C in Barth's solution, receptor expression was assayed using two-electrode voltage clamp with the membrane held at −70 mV. Test concentrations close to each receptor's EC50 were used to evoke currents: 100 nM 5-HT for 5-HT2A and 10 nM 5-HT for 5-HT2C, and these test pulses were repeated at 15–20 min intervals to avoid desensitisation. Antagonism protocols evaluated coapplication of PEAs (concentration range 10−11 to 10−4 M) with the test 5-HT concentration for 10 s, and separate sets of experiments assessed effects after preincubations of 20, 60, 120 or 180 s with selected PEA concentrations. Complete 5-HT concentration–response curves were obtained within single oocytes both before and after PEA application or after a 2-min preincubation plus coapplication, allowing assessment of rightward/downward shifts. The primary outcome measure was the amplitude of 5-HT-evoked currents; antagonist potency was summarised by IC50 values interpolated from antagonism curves. A minimum of four oocytes were used per protocol, with oocytes derived from at least two separate frogs. Data were expressed as mean ± s.e.m. and analysed to obtain EC50 and IC50 (and pIC50) values using GraphPad software. Nonparametric tests (Kruskal–Wallis, Mann–Whitney and Friedman & Quade) were also applied, with significance defined as P < 0.05. The PEAs (2C-I, 2C-B, 2C-D, 2C-H) were synthesised and prepared as hydrochloride salts (2C-B as the hydrobromide), dissolved in Barth's solution for application.

All four PEAs tested produced concentration- and time-dependent inhibition of 5-HT-evoked currents at oocytes expressing the 5-HT2A receptor, while failing to inhibit currents at the 5-HT2C receptor under the same conditions, indicating subtype selectivity in this preparation. 2C-I: This compound displayed partial agonist behaviour at 5-HT2A with intrinsic efficacy of about 15% relative to 5-HT; 100 nM 2C-I produced a small current. However, coapplication of 1 nM 2C-I with 100 nM 5-HT reduced the 5-HT-evoked current, and higher concentrations abolished the response. Antagonism was reversible and time-dependent: a 2-min preincubation increased antagonist potency by more than 100-fold compared with simple coapplication. Recovery from high-concentration blockade was slow, with complete recovery at the highest 2C-I concentrations after a 45-min washout. By contrast, at the 5-HT2C receptor 2C-I had greater intrinsic activity (about 35%) and did not significantly inhibit 5-HT-evoked currents. 2C-B: Substituting bromine for iodine produced an antagonist at 5-HT2A with reversible, time-dependent properties similar to 2C-I but approximately 30-fold less potent. A 2-min preapplication of 10 nM 2C-B reduced the maximal 5-HT response at 5-HT2A to 22%, while leaving 5-HT2C responses unaffected. The time course and reversibility of antagonism resembled those reported for 2C-I. 2C-D and 2C-H: Replacement of the C(4) substituent with methyl (2C-D) or hydrogen (2C-H) also yielded reversible, time-dependent antagonists at 5-HT2A. Two-minute applications of 10 or 30 nM 2C-D reduced maximal 5-HT responses at 5-HT2A to 61.3 ± 11.3% and 40.6 ± 13.4%, respectively. Neither compound altered 5-HT-evoked currents at 5-HT2C. Across the series the PEA concentration–response curves were parallel. Relative potency followed the rank order 2C-I > 2C-B > 2C-D > 2C-H for 5-HT2A antagonism. PEAs generally shifted the 5-HT concentration–response curves to the right and downward, and a preincubation of about 2 min was required to attain maximal inhibition in this oocyte model.

Villalobos and colleagues interpret their findings as evidence that 2C-B and structurally related 2,5-dimethoxy-4-substituted PEAs function as potent, reversible antagonists at the 5-HT2A receptor in the X. laevis oocyte expression system, with apparent selectivity over the 5-HT2C subtype. They stress that these data challenge the prevailing assumption that phenylalkylamine hallucinogens necessarily act as full or partial 5-HT2A agonists, because several recognised human hallucinogens in this chemical family behaved as antagonists in their model. Several mechanistic explanations are discussed. One proposal is that a hydrophobic pocket in the 5-HT2A receptor accommodates the C(4) substituent, such that bulkier hydrophobic groups (iodine, bromine) increase antagonist affinity; by this account 2C-I is most potent because iodine fits optimally. The authors suggest the 5-HT2C receptor may lack a similar hydrophobic pocket or possess a more hydrophilic site, consistent with the lack of antagonism observed at 5-HT2C and with the generally higher intrinsic activity of PEAs at that subtype. Alternative explanations acknowledged by the investigators include functional coupling selectivity (differential G-protein engagement), spare receptor reserve effects, and the possibility that PEAs act at other receptor systems implicated in psychostimulation and hallucinogenesis, including NMDA or metabotropic glutamate receptors and dopaminergic pathways. They note that some classical hallucinogens (for example LSD) have been reported to exhibit antagonist properties in other preparations, and that intracellular signalling measures (such as phosphoinositide hydrolysis) do not always correlate neatly with in vivo hallucinogenic activity. The authors also discuss the slow onset of PEA antagonism in their assay, suggesting it may reflect slow association kinetics or access to the binding site—factors that might be influenced but not wholly explained by bulky C(4) substituents, since even the hydrogen-substituted compound showed slow onset. Limitations they acknowledge include the restricted number of PEAs tested, the known constraints of the X. laevis oocyte model (transient/transfected protein expression and potential differences in G-protein coupling), and the absence of behavioural data; these caveats preclude definitive conclusions about the mechanisms underlying human psychostimulant or hallucinogenic effects. The investigators conclude that further pharmacological, behavioural and molecular studies are needed, and propose that these PEAs could serve as leads for development of competitive, selective 5-HT2A antagonists and tools to clarify 5-HT2 receptor physiology.