Secondary Logo

Share this article on:

Palonosetron Exhibits Unique Molecular Interactions with the 5-HT3 Receptor

Section Editor(s): Durieux, Marcel E.; Gin, TonyRojas, Camilo PhD*; Stathis, Marigo MS*; Thomas, Ajit G. MSE*; Massuda, Edward B. MS*; Alt, Jesse BS*; Zhang, Jie PhD*; Rubenstein, Ed MD; Sebastiani, Silvia PhD; Cantoreggi, Sergio PhD§; Snyder, Solomon H. MD; Slusher, Barbara PhD*

doi: 10.1213/ane.0b013e318172fa74
Anesthetic Pharmacology: Research Report

BACKGROUND: Palonosetron is a 5-HT3-receptor antagonist (5-HT3-RA) that has been shown to be superior to other 5-HT3-RAs in phase III clinical trials for the prevention of acute, delayed, and overall chemotherapy-induced nausea and vomiting. The improved clinical efficacy of palonosetron may be due, in part, to its more potent binding and longer half-life. However, these attributes alone are not sufficient to explain the results with palonosetron. We sought to elucidate additional differences among 5-HT3-RAs that could help explain the observations in the clinic.

METHODS: Receptor site saturation binding experiments were performed with [3H] palonosetron, [3H] granisetron, and [3H] ondansetron to obtain the corresponding Scatchard analyses and Hill coefficients. Diagnostic equilibrium binding experiments and kinetic dissociation experiments were conducted to examine competitive versus potential allosteric interactions between ondansetron, granisetron and palonosetron and the 5-HT3 receptor. Finally, the long-term effect of the three antagonists on receptor function as measured by Ca2+ influx in HEK 293 cells expressing the 5-HT3-receptor was compared.

RESULTS: Analyses of binding isotherms using both Scatchard and Hill plots suggested positive cooperativity for palonosetron and simple bimolecular binding for both granisetron and ondansetron. Equilibrium diagnostic tests discriminated differential effects of palonosetron on [3H] ligand binding indicating that palonosetron was an allosteric antagonist whereas granisetron and ondansetron were competitive antagonists. Using dissociation rate strategies, palonosetron was shown to be an allosteric modifier that accelerated the rate of dissociation from the receptor of both granisetron and ondansetron. Differences in the binding mode of palonosetron to the 5-HT3 receptor were shown to have an impact on receptor function. In these experiments, cells were incubated with each antagonist, followed by infinite dilutions and dissociation for 2.5 h; cells previously incubated with either granisetron or ondansetron showed calcium-ion influx similar to control cells that had not been exposed to a 5-HT3 receptor antagonist. In contrast, substantial inhibition of calcium-ion influx was observed in cells that had been incubated with palonosetron.

CONCLUSIONS: Palonosetron exhibited allosteric binding and positive cooperativity when binding to the 5-HT3 receptor. Palonosetron also triggered functional effects that persisted beyond its binding to the 5-HT3 receptor at the cell surface. Differences in binding and effects on receptor function may be relevant to the unique beneficial actions of palonosetron. To our knowledge, this is the first report showing palonosetron's interaction with the 5-HT3 receptor at the molecular level, clearly differentiating it from other 5-HT3-RAs.

IMPLICATIONS: Palonosetron exhibits unique molecular interactions with the 5-HT3 receptor. Palonosetron's combined properties of allosteric binding, positive cooperativity, and long-term effect on receptor function make it a distinctive antagonist within the 5-HT3 receptor class.

From the *Research and Translational Development, MGI PHARMA INC, Baltimore, MD; †Medical & Scientific Affairs, MGI PHARMA INC, Bloomington, Minnesota; ‡Medical Marketing, HELSINN HEALTHCARE SA, Lugano, Switzerland; §Research and Development, HELSINN HEALTHCARE SA, Lugano, Switzerland; and ∥Department of Neuroscience, Johns Hopkins Medical School, Baltimore, Maryland.

Accepted for publication February 11, 2008.

Dr. Snyder is a consultant to MGI and a shareholder.

Address correspondence and reprint requests to Camilo Rojas, PhD, MGI PHARMA INC, 6611 Tributary St., Baltimore, MD 21224-6515. Address e-mail to

Serotonin 5-HT3 receptor antagonists (5-HT3-RAs) are the foundation of antiemetic prophylaxis for chemotherapy-induced emesis (CINV); they are also commonly used alone or in combination with other drugs to prevent or treat postoperative nausea and vomiting. As a class of drugs, 5-HT3-RAs have been considered competitive antagonists for serotonin, the naturally occurring ligand. Most of the scientific literature for many years has considered these drugs therapeutically equivalent and interchangeable.

Palonosetron is the newest 5-HT3-RA and has been shown to be superior to first generation 5-HT3-RAs in phase III clinical trials for the prevention of acute (0–24 h) and delayed (24–120 h) emesis after moderately emetogenic chemotherapy.1,2 A single IV dose of palonosetron (0.25 mg) was more effective than a single IV dose of dolasetron (100 mg) in preventing acute and delayed CINV; complete response rates, i.e., patients with no emetic episodes and no rescue medication, during the acute period were 63.0% for palonosetron and 52.9% for dolasetron. During the delayed period, complete response rates were 54.0% for palonosetron and 38.7% for dolasetron.1 In a separate study, a single IV dose of palonosetron (0.25 mg) was significantly superior to a single IV dose of ondansetron (32 mg) in the prevention of acute and delayed CINV; complete response rates during the acute period were 81.0% for palonosetron and 68.6% for ondansetron. During the delayed period, complete response rates were 74.1% for palonosetron and 55.1% for ondansetron.2 In another phase III randomized trial, with dexamethasone pretreatment, palonosetron (0.25 mg, IV) was significantly more effective (P < 0.05) in preventing emesis after highly emetogenic chemotherapy (53.3% patients emesis-free) over ondansetron (32 mg, IV, 33.3% emesis-free) throughout the 5-day postchemotherapy period.3 Palonosetron is the only 5-HT3-RA approved by the Food and Drug Administration for the prevention of both acute and delayed CINV.4 In addition, palonosetron significantly reduces the subjective sensation of nausea to a greater extent than other tested 5-HT3-RAs.2,5,6

Palonosetron exhibits significantly different characteristics from other drugs in its class that may help explain the clinical results. First, palonosetron has unique structural characteristics; while older drugs are based on a 3-substituted indole structure resembling serotonin, palonosetron is based on a fused tricyclic ring system attached to a quinuclidine moiety (see chemical structures below).

Second, palonosetron is much more potent at 5-HT3 receptors (pKi = 10.45) than the other drugs with an affinity constant at least an order of magnitude lower.7 Additionally, palonosetron is substantially longer acting and it has a plasma half-life exceeding 40 h,8 whereas the half-life for the other drugs ranges from 5 to 12 h.9 The improved clinical efficacy of palonosetron may be due, in part, to its more potent binding and longer half-life. However, these attributes alone are not sufficient to explain the results with palonosetron. If improved clinical efficacy was the result of enhanced potency, other drugs could be administered at higher doses provided the receptor is not saturated. Similarly, if improved efficacy was the result of longer half-life alone, other drugs with a shorter half-life could be administered more often. However, ondansetron does not mimic palonosetron's protective action in delayed emesis even when ondansetron is administered beyond 24 h after chemotherapy.10 Finally, the longer duration of action of palonosetron does not account for its greater efficacy in protecting patients from emesis within 24 h after moderately emetogenic chemotherapy compared with older drugs, such as ondansetron or dolasetron.

Since palonosetron differs in chemical structure from the other drugs, we wondered whether it might display different types of 5-HT3 receptor interactions. Even though the binding constants of 5-HT3 RAs have been reported in the literature,7,11 no equilibrium or kinetic diagnostic tests to extract potential allosteric interactions and long-term effects on receptor function have been performed on any of these antagonists. Accordingly, we have used tritium ([3H]) labeled preparations of palonosetron, granisetron and ondansetron and explored receptor binding of these drugs. We chose the latter drugs because, together with palonosetron, they constitute the majority of the 5-HT3-RAs used in United States clinical practice. Additionally, we have studied the 5-HT receptor-mediated enhancement of calcium influx in cells selectively expressing 5-HT3 receptors. We report that palonosetron exhibits allosteric interactions and positive cooperativity with the 5-HT3 receptor and that these characteristics are not displayed by the other two drugs. Moreover, receptor binding of palonosetron and its inhibition of calcium influx are not readily reversible implying the existence of a long-term alteration in the receptor or receptor internalization selectively for palonosetron. These receptor interactions may be relevant to the unique beneficial actions of palonosetron.

Back to Top | Article Outline


Membrane Preparation

Membranes used for these receptor binding experiments were prepared according to previously published protocols.12 N1E-115 cells, derived from mouse neuroblastoma tumors, were grown in Dulbecco's modified Eagle's medium (DMEM) containing 10% heat-inactivated fetal bovine serum, 5 mM l-glutamine, and 1% antibiotic/antimycotic (100-fold liquid, Invitrogen). Cells were harvested in their growth phase (i.e., at about 90% confluency). Cells were rinsed with Hank's buffered saline solution and treated with trypsin-EDTA for 5–10 min at 37°C to detach cells from their monolayer. Fresh growth media was added to stop the trypsin-catalyzed reaction and the cells harvested after brief dissociation and gentle centrifugation (228g for 10 min at 4°C). To prepare membranes, cells were homogenized in 50 mM Tris-HCl, 5 mM Na2EDTA buffer (pH 7.4 at 4°C) using a Polytron P-10 tissue homogenizer (setting 5, 2 × 10 s). Homogenates were centrifuged at 48,000g for 15 min and then washed by resuspension and centrifugation in the homogenizing buffer. A second and final wash, by resuspension and centrifugation, was done in 50 mM Tris-HCl, 0.5 mM Na2EDTA (pH 7.4 at 4°C). Membranes were stored at −80°C until needed.

Back to Top | Article Outline

Radioligand Binding Assays

Binding assays were based on published procedures.7,13 For saturation binding reactions, N1E-115 cell membranes were incubated with varying concentrations of each [3H] antagonist in Tris-Krebs buffer (154 mM NaCl, 5.4 mM KCl, 1.2 mM KH2PO4, 2.5 mM CaCl2, 1.0 mM MgCl2, 11 mM d-Glucose, 10 mM Tris, pH 7.4 at 25°C) for 30 min. Nonspecifically bound radioligand was measured in the presence of 100-fold K d unlabeled antagonist; nonspecific binding was then subtracted from total binding to obtain specific binding. For diagnostic assays, [3H] antagonist at a given concentration was incubated for 30 min with N1E-115 cell membranes. Unlabeled antagonist was then added at increasing concentrations to determine the concentration needed to displace half of the [3H] antagonist (IC50). The same procedure was performed using increasing concentrations of [3H] antagonist; the range of concentrations included values below and above the corresponding K d values. The IC50 of unlabeled antagonist was then plotted as a function of [3H] antagonist concentration. For kinetic dissociation studies, the association phase was conducted by incubating N1E-115 cell membranes with [3H] antagonist at twice the dissociation constant (K d) for 30 min. The dissociation phase was then initiated by addition of excess unlabeled antagonists. The antagonist being evaluated for allosteric modification was added during the dissociation phase.13 The amount of [3H] antagonist still bound to the receptor was measured at various times during the first hour after addition of unlabeled antagonists. Incubations were terminated by vacuum filtration through Whatman GF/B filters pretreated with 0.3% polyethylenimine. Filters were washed three times with ice-cold 0.1 M NaCl buffer, dried, and the radioactivity retained on the filters measured using a scintillation counter (TopCount). Prism and IDBS ExcelFit programs were used for curve-fitting and statistical analyses. Analysis of the binding data using Hill plots was based on the guidelines given by Cornish-Bowden and Koshland.14

Back to Top | Article Outline

Plasmid Preparation and Cell Transfections

Recombination reactions (LR Clonase) were set up to transfer 5-HT3A and 5-HT3B coding sequences (Ultimate ORF Clone Library, Invitrogen) into the Gateway® compatible cytomegalovirus expression vectors with neomycin or blasticidin selectable markers. The resulting plasmids were named pcDNA3.2-5-HT3A, pcDNA3.2-5-HT3B, pcDNA6.2-5-HT3A, and pcDNA6.2- 5-HT3B. The correct sequences for all four plasmids were confirmed by DNA sequencing through the entire coding sequences. In vitro translations were performed with pcDNA3.2-5-HT3A and pcDNA3.2-5-HT3B as templates using rabbit reticulocytes and 35S-methionine. The translated proteins were resolved by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). The dried gel was exposed to radiograph film (−80°C overnight). Single protein bands with a molecular weight of 55 and 49 kD were observed in the product of 5-HT3A and 5-HT3B template respectively. Plasmids were purified using anion exchange columns (Qiagen Endo Free Plasmid Maxi Kit).

HEK-293 cells were transfected with pcDNA3.2-5-HT3A using Roche Fugene 6 transfection reagent and expanded in the presence of 500 μM genetecin. After 2 wk of genetecin exposure, cells were plated out in 96-well plates with a density averaging 1 cell per well. Wells that grew a single colony were isolated for expansion. Expanded clones were tested for the presence of 5-HT3A receptors using radioligand binding and calcium influx assays. A HEK-293 cell line with stable expression of 5-HT3A was then cotransfected with pcDNA6.2-5-HT3B and grown in the presence of 3 μg/mL blasticidin. A HEK-293 cell line with stable expression of both 5-HT3A and 5-HT3B proteins was selected by the same procedure as above.

Back to Top | Article Outline

Inhibition of Calcium Influx

Calcium influx measurements were based on published procedures.15,16 HEK-293 cells transfected with either the human 5-HT3A homomeric or the 5-HT3AB heteromeric serotonin receptors were plated onto glass- bottomed dishes treated with poly-d-lysine and allowed to grow for at least 4 days in RPMI-1640 media supplemented with 10% heat-inactivated fetal bovine serum and 2 mM l-glutamine. Cells were treated with the three antagonists (at five times their respective K d values; palonosetron 1 nM, granisetron 5 nM, ondansetron 30 nM) or with saline for 24 h. Subsequent to the treatment, antagonists were removed and antagonist- free media was added to the cells. An hour later, cells were rinsed with F-12/DMEM and incubated for an hour at room temperature with the acetoxymethyl (AM) ester form of the fluorescent Ca2+ indicator (Fluo-4 AM ester, 2 μM). Pluronic acid (0.04%) was added as nonionic surfactant to sequester the AM ester into micelles for cell uptake. Both the am ester and pluronic acid were dissolved in HEPES buffered saline (130 mM NaCl, 2 mM KCl, 1 mM MgCl2 and 2 mM CaCl2, 20 mM HEPES, pH 7.4). Cells were then allowed to recover in fresh buffer for 30 min in order to allow for de-esterfication of the dye into the polycarboxylate form. The polycarboxylate form of the dye is the form that binds Ca2+ when it enters the cell to form a fluorescent complex. After the recovery period, cells were challenged with serotonin (10 μM) and changes in fluorescence intensity at 470 nm were captured every second for 3 min using Intracellular Imaging's InCyt program. Student's t-test was used for statistical analyses of the results.

Back to Top | Article Outline

Incubation of 5-HT3A/HEK or 5-HT3AB/HEK Cells with [3H] Antagonists

Inhibition of calcium influx experiments were performed on 5-HT3A or 5-HT3AB transfected HEK cells using [3H] antagonists. At the end of the serotonin challenge, media was removed, cells scraped into 50 μL of fresh buffer, and the radioactivity present in the scraped material measured using a scintillation counter. Student's t-test was used for statistical analyses of the results.

Back to Top | Article Outline

Binding of [3H] Palonosetron After Incubation of Cells with Unlabeled Antagonists

Experiments were performed using 5-HT3A-transfected HEK cells. Cells were treated with unlabeled antagonists or saline for 24 h. Subsequent to the treatment, the antagonists were removed and fresh growth media added to the cells. An hour later, cells were rinsed with F-12/DMEM and incubated for an hour at room temperature in buffer. Cells were incubated in fresh buffer for an additional 30 min. The buffer was removed and cells incubated with [3H] palonosetron (1 nM) for 40 min at room temperature. The label was removed and the cells washed with ice-cold buffer. After the wash, cells were scraped into 50 μL of fresh ice-cold buffer and the radioactivity associated with the cells was measured using a scintillation counter. Student's t-test was used for statistical analyses of the results.

Back to Top | Article Outline


Positive Cooperativity and Allosteric Receptor Actions of Palonosetron

Binding Isotherms

We examined the binding of increasing concentrations of [3H] granisetron, [3H] ondansetron, and [3H] palonosetron to cell membranes from N1E-115 cells at a constant protein concentration (Figs. 1a, d, and g). All [3H] ligands displayed saturable binding with K d values of 6 nM for ondansetron, 1 nM for granisetron, and 0.2 nM for palonosetron, consistent with previously reported values.7 [3H] palonosetron binding isotherm at low concentrations exhibited a sigmoidal shape in contrast to [3H] granisetron and [3H] ondansetron (insets in Figs. 1a, d, and g). Scatchard plots can be used to determine if ligands bind to single or apparent multiple sites reflecting positive or negative cooperativity.13 Scatchard plots were linear for granisetron and ondansetron but displayed a complex inverted U form for palonosetron (Figs. 1b, e, and h). The results are consistent with simple bimolecular binding for granisetron and ondansetron and positive cooperativity for palonosetron.13 Since binding of palonosetron displays positive cooperativity, the calculated K d of palonosetron is a composite value representing different affinities at each level of occupancy. Hill coefficients can also discriminate multiple sites and positive or negative cooperativity.14 Hill coefficients for ondansetron and granisetron were approximately 1 (ondansetron: 0.99; granisetron: 1.08) indicating the absence of cooperativity (Figs. 1c and f). By contrast, the Hill coefficient for palonosetron was 1.5 (Fig. 1i), consistent with positive cooperativity.

Figure 1

Figure 1

Back to Top | Article Outline

Equilibrium Diagnostic Tests

When an allosteric antagonist affects the affinity of the receptor for a [3H] antagonist, competition experiments between the two antagonists can discriminate allosteric antagonism from simple competitive interactions.13,17 Accordingly, we examined the competition of palonosetron, granisetron, and ondansetron for binding of the radiolabeled ligands (Fig. 2). In such competition studies, at concentrations of the [3H] ligand lower than the K d, one would not expect variations in the IC50 of the unlabeled drug, as its IC50 simply reflects its affinity for the receptor. At concentrations above the K d, the unlabeled drug dilutes the [3H] ligand so there should be a linear increase in IC50 with increasing concentrations with the [3H] ligand. For granisetron inhibiting [3H] granisetron binding and for ondansetron inhibiting [3H] ondansetron binding, there was a linear relationship between the IC50 and the [3H] drug concentration (Figs. 2a and d). For palonosetron inhibiting [3H] palonosetron binding, we observed a similar linear relationship with a modest concavity in the plot (Fig. 2g).

Figure 2

Figure 2

If an unlabeled drug competes with a different [3H] drug at the same site, using concentrations of the [3H] ligand exceeding the K d, one should also observe linear relationships.13,17 Such linear relationships were apparent for ondansetron inhibiting [3H] granisetron binding and granisetron inhibiting [3H] ondansetron binding (Figs. 2b and e). In contrast, we detected strikingly nonlinear relationships in the competition by palonosetron for the binding of [3H] granisetron and [3H] ondansetron (Figs. 2c and f). We also observed a modest concavity in the competition curve of granisetron and ondansetron competing for [3H] palonosetron binding implying that the drugs interact at different sites.

The competition between [3H] ondansetron and palonosetron illustrates the prediction that the IC50 of the unlabeled drug simply reflects its affinity for the receptor when concentrations of the [3H] ligand are lower than the K d. The change in palonosetron's IC50 when the [3H] ondansetron concentration is below its K d varies little resulting in a lag phase (Fig. 2f). The lag is quite apparent in this case because the affinity of palonosetron for the 5-HT3 receptor is more than 100-fold higher than ondansetron's so that it takes a substantial amount of ondansetron to compete with palonosetron and elicit an increase in its IC50.

When evaluating nonradioactive palonosetron as a potential allosteric antagonist against [3H] granisetron or [3H] ondansetron, curved relationships with a plateau were observed (Figs. 2c and f). On the other hand, when using [3H] palonosetron as radioligand to be displaced, it led to relationships with a slight curvature upwards (Figs. 2g–i). Both sets of results might reflect palonosetron's positive cooperativity. In the first case, binding of palonosetron to the allosteric site may trigger a higher binding affinity for the orthosteric site (serotonin binding site) preventing binding of [3H] granisetron or [3H] ondansetron; increases in [3H] antagonist concentrations did not change the IC50 resulting in the observed plateaus (Figs. 2c and f). Similarly, [3H] palonosetron binding to the allosteric site triggered an increased affinity for the orthosteric site resulting in an increase in the concentration of unlabeled antagonist needed to displace [3H] palonosetron and the corresponding upward concavity relationships (Figs. 2 g–i).

Back to Top | Article Outline

Kinetic Diagnostic Tests

The equilibrium diagnostic experiments implied that palonosetron influences 5-HT3 receptors at a different site than the other ligands.13,17 Such allosteric interactions can also be studied in dissociation experiments.13 Dissociation of a radiolabeled ligand from its binding site can be initiated with excess concentrations of the unlabeled ligand with allosteric drugs altering the dissociation rate. Accordingly, we labeled 5-HT3 receptors in N1E-115 cell membranes with [3H] palonosetron, [3H] granisetron, and [3H] ondansetron and initiated dissociation with excess concentrations of various drugs (Fig. 3). The dissociation of [3H] granisetron elicited by 0.2 μM granisetron was monophasic with a half-life of about 5 min; dissociation initiated with a five-fold higher concentration (1 μM) provided the same half-life (Table 1, Fig. 3a). The dissociation rate with 0.2 μM granisetron plus 1 μM ondansetron was the same as with 0.2 μM granisetron (Table 1, Fig. 3b). By contrast, dissociation of [3H] granisetron was substantially accelerated in the presence of 1 μM palonosetron plus 0.2 μM granisetron consistent with an allosteric action of palonosetron (Table 1, Fig. 3c). Similar findings were obtained with [3H] ondansetron whose dissociation rate was the same with excess granisetron as with excess ondansetron, whereas dissociation was markedly accelerated by excess palonosetron (Table 1, Figs. 3d–f). Excess granisetron or ondansetron failed to alter the dissociation rate of [3H] palonosetron obtained in the presence of excess palonosetron (Figs. 3g–i).

Figure 3

Figure 3

Table 1

Table 1

Back to Top | Article Outline

Persistent Association of [3H] Palonosetron to 5-HT3 Receptor-Expressing Cells Suggests Receptor Internalization

We wondered whether the unique receptor interactions of [3H] palonosetron might reflect influences on receptor dynamics in intact cells. Accordingly, we used HEK293 cells stably over-expressing 5- HT3A receptors and evaluated the persistence of [3H] palonosetron, [3H] ondansetron and [3H] granisetron binding (Table 2). To ensure saturation of receptors, we incubated the cells for 24 h with concentrations of [3H] ligands five times their K d values. The media in one set of cells was removed and radioactivity associated with cells measured and normalized to 100%. In parallel sets of cells, media was replaced with fresh drug-free media, which was changed three times over a 2.5-h period, which is more than 15 times the longest dissociation half-life of any of the ligands7,11 (Table 1). After this procedure, the percentages of radioactivity associated with cells that were initially incubated with [3H] ondansetron and [3H] granisetron were 4% ± 2 and 15% ± 3, respectively. In contrast, 53% ± 11 (P < 0.01 and <0.05 when compared with ondansetron and granisetron, respectively) of the initially bound [3H] palonosetron was retained (Table 2). We wondered whether the retained [3H] palonosetron reflected nonspecifically bound ligand or was associated with the receptor. Accordingly, we conducted the same experiments using HEK293 cells not expressing 5-HT3 receptors. Under these conditions, [3H] palonosetron, like [3H] ondansetron and [3H] granisetron, was undetectable.

Table 2

Table 2

In another approach, we pretreated the 5-HT3A-HEK293 cells for 24 h with each unlabeled drug in the same concentrations previously used for [3H] ligands. After the 24-h incubation, we subjected the cells to the same dilution and washing to fully dissociate receptor-bound drugs. All cells were then incubated with [3H] palonosetron (1 nM) for 40 min. At the end of this time, cells were washed with ice-cold saline (2 mL) and radioactivity associated with the cells was measured. After ondansetron or granisetron treatment, there was minimal change in [3H] palonosetron binding compared cells that had not been pretreated with antagonist (Table 2). By contrast, palonosetron treatment caused a reduction to 45% ± 4 binding (P < 0.01 when compared to results with ondansetron and granisetron) (Table 2).

Back to Top | Article Outline

Palonosetron Causes Long-Lasting Inhibition of 5-HT3 Receptor Function

To determine whether the putative receptor internalization elicited by palonosetron upon ligand binding impacted 5-HT3 receptor physiology, we monitored serotonin-elicited calcium-ion influx in HEK293 cells stably transfected with 5-HT3A receptors (Fig. 4). We treated the cells in the same way as the long-term binding studies. Thus, cells were incubated with drug concentrations five times their K d for 24 h, after which they were maintained in fresh drug-free medium for 2.5 h with three additional media changes designed to ensure complete dissociation of drug from receptors. Cells were then exposed to serotonin (10 μM) and calcium-ion influx monitored. Prior exposure to palonosetron led to a major reduction in response to serotonin-induced calcium-ion influx (37% ± 5, P < 0.001 compared with 100% response of cells that had not been preincubated with antagonist). In contrast, there was a minor decrease in cells treated with granisetron (79% ± 8, P < 0.05) and no significant decrease in responses of cells pretreated with ondansetron (89% ± 7, n.s.) (Fig. 4). Similar results were obtained when the experiment was conducted using 5-HT3AB receptor-expressing HEK293 cells (data not shown).

Figure 4

Figure 4

Back to Top | Article Outline


The principal findings of this study are that palonosetron interacts with 5-HT3 receptors very differently than granisetron or ondansetron and that this differential interaction triggers a receptor alteration or internalization resulting in a long-lived inhibition of receptor function.

Three independent sets of experiments, binding isotherms, equilibrium diagnostic tests, and kinetic diagnostic tests, indicate that palonosetron acts as an allosteric antagonist with positive cooperativity in clear contrast to ondansetron and granisetron. Analyses of binding isotherms using both Scatchard and Hill plots indicate that palonosetron exhibits positive cooperativity, whereas granisetron and ondansetron exhibit simple bimolecular binding.

The differential effects of palonosetron on [3H] ligand binding indicate that palonosetron interacts with 5-HT3 receptors at different or additional sites than those binding granisetron or ondansetron. In addition, using dissociation rate strategies, palonosetron was shown to be an allosteric modifier that accelerated the rate of dissociation from the receptor of both granisetron and ondansetron. Neither granisetron nor ondansetron had an effect on dissociation rates of other antagonists. These kinetic investigations also support an interaction of palonosetron with sites distinct to those labeled by granisetron or ondansetron. These data are consistent with palonosetron binding to an allosteric site on the receptor, inducing a conformational change and decreasing the affinity of bound [3H] granisetron or [3H] ondansetron resulting in the observed increase in dissociation rate.

This unique mode of receptor interaction is associated with long-term effects on receptor ligand binding and functional responses to serotonin. Thus, incubation of cells with [3H] palonosetron for 24 h, followed by infinite dilution and incubation with antagonist-free media, led to 53% [3H] palonosetron associated with cells (Table 2). The substantial amount of receptor-associated [3H] palonosetron that persisted after prolonged dilution and washing suggested that the bound [3H] palonosetron could have been internalized. Furthermore, when cells were preincubated with unlabeled palonosetron for 24 h, followed by infinite dilution and incubation with antagonist-free media, binding of [3H] palonosetron was reduced to 45% compared with that of cells that were not preincubated with antagonist (Table 2). This major decrease in [3H] palonosetron binding despite procedures that should have fully dissociated palonosetron implies that the pretreatment caused movement of the receptor to a site inaccessible to [3H] ligands, consistent with receptor internalization.

Palonosetron treatment also provided a 63% decline in serotonin-elicited calcium influx. None of these effects was evident with ondansetron and was minimal with granisetron (Table 2 and Fig. 4). Current efforts in the laboratory are focused upon visualizing palonosetron-triggered receptor internalization using 5-HT3 receptors labeled with enhanced cyan fluorescent protein.18

The persistent, seemingly irreversible actions of palonosetron might reflect some long-term alteration in the receptor causing augmentation of ligand affinity to a pseudoirreversible state. We are not aware of instances of altered receptor affinity resembling our findings with palonosetron. Desensitization of nicotinic cholinergic receptors in the electric organ of an elasmobranch is associated with a unique high affinity state for agonists, but the kinetics of this effect differ markedly from what we have observed for palonosetron.19,20

Alternatively, palonosetron exposure may have caused receptor internalization. Receptor internalization is well characterized for many receptors including steroid receptors, polypeptide receptors, G-protein coupled receptors,21 and ionotropic receptors comparable to 5-HT3 receptors such as glutamate-AMPA receptors.22 Receptor internalization is typically elicited by agonists although some receptor antagonists have been shown to cause internalization.23 It is possible that palonosetron's antagonist effects on receptors reflect inverse agonism. An inverse agonist is an agent which decreases basal receptor function (e.g., calcium ion influx) in the absence of agonist exposure. Studies indicate that many drugs, specially G-protein coupled receptors antagonists, previously characterized as conventional antagonists are, in fact, inverse agonists.24 It was not possible to evaluate inverse agonism in our system because the 5-HT3 receptor expressed in HEK293 cells did not exhibit basal calcium-ion influx.

It is clear from the binding, calcium-ion influx and potential receptor internalization results that palonosetron interacts with 5-HT3 receptors differently than ondansetron and granisetron. These pharmacological findings may have clinical relevance. In a side-by-side comparison, palonosetron was statistically and clinically more effective in protecting patients from acute CINV than ondansetron.1 Moreover, palonosetron decreased the subjective sensation of nausea more than other 5-HT3 receptor antagonists.5 This greater efficacy under acute conditions is not readily explained simply by greater molar potency at receptors, as more potent drugs are administered at lower doses, so that receptor occupancy typically is similar under clinical conditions for potent and weak drugs. The longer half-life of palonosetron would not impact acute clinical effects. Alternatively, long-term inhibition of calcium-ion influx could have a differential effect on intracellular signaling that could translate into superior inhibition of emesis. Furthermore, the greater therapeutic efficacy of palonosetron may reflect actions at an allosteric site, which elicits receptor internalization. Such internalization providing fewer available receptor sites may diminish agonist responses more than simple reversible competition. Receptor internalization may also account for the ability of palonosetron to prevent delayed nausea and vomiting, which is not evident with all other 5-HT3 antagonists examined.4,25 The long plasma half-life of palonosetron is less likely to explain these clinical effects, because long-term infusion of ondansetron does not relieve delayed nausea and vomiting.10

In summary, our results provide strong evidence that palonosetron exhibits allosteric binding and positive cooperativity when binding to the 5-HT3 receptor. The prolonged inhibition of serotonin-induced calcium-ion influx also indicates that palonosetron triggers functional effects that persist beyond its immediate binding to 5-HT3 receptors. To our knowledge, this is the first report showing palonosetron's unique interaction with the 5-HT3 receptor at the molecular level, clearly differentiating it from older 5-HT3 receptor antagonists.

Back to Top | Article Outline


1. Eisenberg P, Figueroa-Vadillo J, Zamora R, Charu V, Hajdenberg J, Cartmell A, Macciocchi A, Grunberg S. Improved prevention of moderately emetogenic chemotherapy-induced nausea and vomiting with palonosetron, a pharmacologically novel 5-HT3 receptor antagonist: results of a phase III, single-dose trial versus dolasetron. Cancer 2003;98:2473–82
2. Gralla R, Lichinitser M, Van Der Vegt S, Sleeboom H, Mezger J, Peschel C, Tonini G, Labianca R, Macciocchi A, Aapro M. Palonosetron improves prevention of chemotherapy-induced nausea and vomiting following moderately emetogenic chemotherapy: results of a double-blind randomized phase III trial comparing single doses of palonosetron with ondansetron Ann Oncol 2003;14:1570–7
3. Aapro MS, Grunberg SM, Manikhas GM, Olivares G, Suarez T, Tjulandin SA, Bertoli LF, Yunus F, Morrica B, Lordick F, Macciocchi A. A phase III, double-blind, randomized trial of palonosetron compared with ondansetron in preventing chemotherapy-induced nausea and vomiting following highly emetogenic chemotherapy. Ann Oncol 2006;17:1441–9
4. Siddiqui MA, Scott LJ. Palonosetron. Drugs 2004;64:1125–32; discussion 33–4
5. Decker GM, DeMeyer ES, Kisko DL. Measuring the maintenance of daily life activities using the functional living index-emesis (FLIE) in patients receiving moderately emetogenic chemotherapy. J Support Oncol 2006;4:35–41, 52
6. Grunberg SM, Deuson RR, Mavros P, Geling O, Hansen M, Cruciani G, Daniele B, De Pouvourville G, Rubenstein EB, Daugaard G. Incidence of chemotherapy-induced nausea and emesis after modern antiemetics. Cancer 2004;100:2261–8
7. Wong EH, Clark R, Leung E, Loury D, Bonhaus DW, Jakeman L, Parnes H, Whiting RL, Eglen RM. The interaction of RS 25259–197, a potent and selective antagonist, with 5-HT3 receptors, in vitro. Br J Pharmacol 1995;114:851–9
8. Stoltz R, Cyong JC, Shah A, Parisi S. Pharmacokinetic and safety evaluation of palonosetron, a 5-hydroxytryptamine-3 receptor antagonist, in U.S. and Japanese healthy subjects. J Clin Pharmacol 2004;44:520–31
9. Constenla M. 5-HT3 receptor antagonists for prevention of late acute-onset emesis. Ann Pharmacother 2004;38:1683–91
10. Geling O, Eichler HG. Should 5-hydroxytryptamine-3 receptor antagonists be administered beyond 24 hours after chemotherapy to prevent delayed emesis? Systematic re-evaluation of clinical evidence and drug cost implications. J Clin Oncol 2005;23:1289–94
11. Nelson DR, Thomas DR. [3H]-BRL 43694 (Granisetron), a specific ligand for 5-HT3 binding sites in rat brain cortical membranes. Biochem Pharmacol 1989;38:1693–5
12. Wong EH, Bonhaus DW, Wu I, Stefanich E, Eglen RM. Labelling of 5- hydroxytryptamine3 receptors with a novel 5-HT3 receptor ligand, [3H]RS-42358–197. J Neurochem 1993;60:921–30
13. Limbird L. Cell surface receptors, a short course on theory and methods. 3rd ed. New York: Springer; 2005
14. Cornish-Bowden A, Koshland DE Jr. Diagnostic uses of the Hill (Logit and Nernst) plots. J Mol Biol 1975;95:201–12
15. Dubin AE, Huvar R, D'Andrea MR, Pyati J, Zhu JY, Joy KC, Wilson SJ, Galindo JE, Glass CA, Luo L, Jackson MR, Lovenberg TW, Erlander MG. The pharmacological and functional characteristics of the serotonin 5-HT(3A) receptor are specifically modified by a 5-HT(3B) receptor subunit. J Biol Chem 1999; 274:30799–810
16. Kao JP. Practical aspects of measuring [Ca2+] with fluorescent indicators. Methods Cell Biol 1994;40:155–81
17. Kenakin TP. Quantitation in receptor pharmacology. Receptors Channels 2001;7:371–85
18. Ilegems E, Pick HM, Deluz C, Kellenberger S, Vogel H. Noninvasive imaging of 5-HT3 receptor trafficking in live cells: from biosynthesis to endocytosis. J Biol Chem 2004;279:53346–52
19. Conti-Tronconi BM, Raftery MA. The nicotinic cholinergic receptor: correlation of molecular structure with functional properties. Annu Rev Biochem 1982;51:491–530
20. Weber M, David-Pfeuty T, Changeux JP. Regulation of binding properties of the nicotinic receptor protein by cholinergic ligands in membrane fragments from Torpedo marmorata. Proc Natl Acad Sci USA 1975;72:3443–7
21. Waldhoer M, Bartlett SE, Whistler JL. Opioid receptors. Annu Rev Biochem 2004;73:953–90
22. Liang F, Huganir RL. Coupling of agonist-induced AMPA receptor internalization with receptor recycling. J Neurochem 2001;77:1626–31
23. Lin JW, Ju W, Foster K, Lee SH, Ahmadian G, Wyszynski M, Wang YT, Sheng M. Distinct molecular mechanisms and divergent endocytotic pathways of AMPA receptor internalization. Nat Neurosci 2000;3:1282–90
24. Kenakin T. Efficacy as a vector: the relative prevalence and paucity of inverse agonism. Mol Pharmacol 2004;65:2–11
25. Rubenstein EB. Palonosetron: a unique 5-HT(3) receptor antagonist indicated for the prevention of acute and delayed chemotherapy-induced nausea and vomiting. Clin Adv Hematol Oncol 2004;2:284–8
© 2008 International Anesthesia Research Society