Prolonged inhibition of 5-HT₃ receptors by palonosetron results from surface receptor inhibition rather than inducing receptor internalization

Abstract

Background and purpose: The 5-HT₃ receptor antagonist palonosetron is an important treatment for emesis and nausea during cancer therapy. Its clinical efficacy may result from its unique binding and clearance characteristics and receptor down-regulation mechanisms. We investigated the mechanisms by which palonosetron exerts its long-term inhibition of 5-HT₃ receptors for a better understanding of its clinical efficacy.

Experimental approach: Cell surface receptors (recombinantly expressed 5HT₃A or 5HT₃AB in COS-7 cells) were monitored using [³H]granisetron binding and ELISA after exposure to palonosetron. Receptor endocytosis was investigated using immunofluorescence microscopy.

Key results: Chronic exposure to palonosetron reduced the number of available cell surface [³H]granisetron binding sites. This down-regulation was not sensitive to either low temperature or pharmacological inhibitors of endocytosis (dynasore or nystatin) suggesting that internalization did not play a role. This was corroborated by our observation that there was no change in cell surface 5-HT₃ receptor levels or increase in endocytic rate. Palonosetron exhibited slow dissociation from the receptor over many hours, with a significant proportion of binding sites being occupied for at least 4 days. Furthermore, our observations suggest that chronic receptor down-regulation involved interactions with an allosteric binding site.

Conclusions and implications: Palonosetron acts as a pseudo-irreversible antagonist causing prolonged inhibition of 5-HT₃ receptors due to its very slow dissociation. In addition, an irreversible binding mode persists for at least 4 days. Allosteric receptor interactions appear to play a role in this phenomenon.

Figure 1

The long-term effect of prior exposure to 5-HT₃ receptor antagonists on the availability of 5-HT₃ receptor binding sites. [³H]granisetron binding was performed in COS-7 cells expressing 5-HT₃A (A) or 5-HT₃AB (B) receptors. Cells were incubated with palonosetron (PAL, 1 nM) or ondansetron (OND, 30 nM) for 150 min at 37°C and incubated in drug-free buffer for 150 min at 37°C with three washes. Remaining surface binding sites were detected using [³H]granisetron. Data are an average of at least three separate experiments performed in triplicate. ***P < 0.001, significantly different from untreated cells UT; anova.

Figure 2

The inhibition of endocytosis does not prevent the palonosetron-mediated down-regulation of 5-HT₃ receptor binding sites. [³H]granisetron binding was performed on 5-HT₃A-expressing COS-7 cells. (A) Palonosetron (1 nM) pretreatment (150 min) at 4, 15 or 37°C and incubated in drug-free buffer for 150 min, at the same temperatures, prior to radioligand binding. (B) Palonosetron (1 nM) was incubated in the presence of dynasore (DYN, 80 μM), nystatin (NYS, 21 μM), or both and incubated at 37°C (150 min). Following incubation in drug-free buffer (150 min) radioligand binding was performed (37°C). Data are expressed as % of untreated cells and experiments were performed on three independent occasions in triplicate.

Figure 3

Cell surface levels of 5-HT₃ receptors are not altered by palonosetron treatment. Cell surface ELISA was performed on COS-7 cells expressing 5-HT₃A-myc (A) or 5-HT₃A-myc/B-HA receptors (B), probing with anti-myc or –HA antibodies, respectively. Cells were incubated with palonosetron (PAL, 1 nM) or ondansetron (OND, 30 nM) for 150 min at 37°C and then fixed to prevent further membrane trafficking. Cell surface receptors were then probed using ELISA. Data are an average of three experiments performed in octuplicate and normalized to untreated (UT).

Figure 4

5-HT₃ receptor internalization is not induced by palonosetron (PAL). (A) COS-7 cells expressing 5-HT₃A-myc (3A) or 5-HT₃A-myc/B-HA (3A/B) were probed with anti-myc or –HA antibodies (4°C, 60 min). Cell surface receptors were identified using Alexa 568 conjugated secondary antibody (surface). Receptor internalization (37°C, 30 min) was identified following the removal of surface antibody by a low pH (acid wash) and probed with the secondary antibody (surface). Internalized receptors were detected following permeabilization (intracellular). (B) Cells were pre-bound with primary antibodies, as above. They were then incubated in the absence or presence of palonosetron (1 nM, 30 min, 37°C). Cells were then permeabilized and probed with secondary antibody. Images represent at least three separate experiments. Scale bar = 50 μm. UT, untreated.

Figure 5

Recovery of 5-HT₃ receptor binding sites following palonosetron exposure. (A) Cells were incubated with palonosetron (1 nM) for 150 min at 4°C, and then washed in acidic buffer (at indicated pH), prior to [³H]granisetron binding (performed at pH 7.4). (B) Following palonosetron exposure (1 nM, 150 min, 37°C) in either live or fixed cells, recovery of 5-HT binding sites was monitored over time (0–96 h) using [³H]granisetron binding and related to the level of binding sites in untreated (UT) cells. (C) Cells were allowed to recover for 8 h after removal of palonosetron in the absence (0.1% ethanol vehicle, VEH) or presence of cycloheximide (CHX, 35 μM), anisomycin (ASN, 10 μM), monensin (MSN, 5 μM) or Exo-1 (EXO, 50 μM). Radioligand binding signal is expressed as a % of control cells that were treated with inhibitor or vehicle but not exposed to palonosetron. Data are average of three independent experiments performed in triplicate.

Figure 6

Concentration-dependency of 5-HT₃ receptor binding and down-regulation by palonosetron and ondansetron. A range of concentrations of palonosetron (A) and ondansetron (B) were tested for their ability to compete for [³H]granisetron binding (competition binding) or cause chronic 5-HT₃ receptor down-regulation. Competition binding was performed by co-incubating 5-HT_3A receptor expressing cells with palonosetron (PAL) or ondansetron (OND) with 0.6 nM [³H]granisetron for 120 min at 19°C. Down-regulation was measured by incubating cells with palonosetron or ondansetron for 150 min at 19°C, washing excess ligand and then incubating in drug-free media for a further 150 min at 19°C before binding to [³H]granisetron. (C) Cells were incubated with or without 10 nM ondansetron for 30 min at 19°C, followed by incubation with 1 nM palonosetron plus 10 nM ondansetron, or 10 nM ondansetron alone for 150 min at 19°C. [³H]granisetron binding was performed after washout of excess ligand and incubation in drug-free media for 150 min at 19°C. Binding was expressed as a percentage of untreated (UT) cells. Data represent an average of three experiments performed in triplicate. *P < 0.5, ***P < 0.001, significantly different as shown; anova.