Address correspondence and reprint requests to Yukiko Nakamura, Department of Neuroscience and Cell Biology, Osaka University Graduate School of Medicine, 2-2, Yamada-oka, Suita, Osaka 565-0871, Japan. E-mail: email@example.com
The 5-hydroxytryptamine (serotonin, 5-HT) type 3 (5-HT3) receptor belongs to the superfamily of Cys-loop ligand-gated ion channels, and can be either homopentameric (5-HT3A) or heteropentameric (5-HT3AB) receptor. Several modulators are known, which either inhibit or potentiate this channel, but few have any appreciable selectivity between the two subtypes or can modulate one receptor differently to the other. In this study, we show that the anticancer drug, topotecan, bidirectionally modulates the 5-HT3 receptor using a two-electrode voltage clamp technique. Topotecan inhibited 5-HT-gated current through homomeric 5-HT3A receptors. Interestingly, however, additional expression of the 5-HT3B subunit changed the response to topotecan dramatically from an inhibitory to a potentiatory one. This effect was dependent on the level of 5-HT3B subunit expression. Moreover, the effect was reduced in the receptors containing the 5-HT3B(Y129S) polymorphic variant. These finding could explain individual differences in the sensitivity to topotecan-induced nausea and vomiting.
The 5-HT3 receptor belongs to the superfamily of Cys-loop ligand-gated ion channels, which also includes the nicotinic acetylcholine (nACh) receptor, gamma aminobutyric acid-type-A (GABAA) receptor, glycine receptor, and zinc-activated channels (reviewed in (Lester et al. 2004)). The 5-HT3 receptor complex comprises five subunits, which surround a cation-permeable (Na+, Ca2+, K+) channel pore (Green et al. 1995; Davies et al. 1999; Lummis et al. 2005). The human genome contains five genes encoding different 5-HT3 subunits (5-HT3A, B, C, D, and E) (Niesler et al. 2003). It is mainly the 5-HT3A and 5-HT3B subunits that are involved in the formation of functional receptors (Niesler et al. 2007). 5-HT3A subunits can form a functional homomeric receptor, whereas the 5-HT3B subunit alone cannot (Davies et al. 1999), and instead achieves functionality through the formation of heteromeric complexes with 5-HT3A subunits in a proposed subunit stoichiometry of 2A:3B (Davies et al. 1999; Barrera et al. 2005). The biophysical properties of the 5-HT3A and 5-HT3AB receptors exhibit differences. The 5-HT3AB receptor exhibits large single-channel conductance, low permeability to calcium ions, and a linear current-voltage relationship (Davies et al. 1999), and these properties of the heteromer more closely resemble those of the receptor characterized in the majority of mammalian systems (Yang et al. 1992; Hussy et al. 1994).
However, the two receptor subtypes are generally very similar pharmacologically (Brady et al. 2001), and there are no currently known drugs that can distinguish between 5-HT3A homomers and 5-HT3AB heteromers, although some drugs (e.g. picrotoxin (Das and Dillon 2003), tubocurarine (Davies et al. 1999) and irinotecan (Nakamura et al. 2011)) are less potent in blocking agonist-induced currents through the 5-HT3AB receptor than through the 5-HT3A receptor.
5-HT3 receptors are located in the gastrointestinal (GI) tract and in the chemoreceptor trigger zone (CTZ) located in the area postrema and nucleus tractus solitaries (NTS) of the vomiting center (Kilpatrick et al. 1989; Pratt and Bowery 1989; Champaneria et al. 1992). Following exposure to cytotoxic drugs, 5-HT is released from enterochromaffin cells in the mucosa of the small intestine adjacent to vagal afferent neurons expressing 5-HT3 receptors. The released 5-HT activates these neurons via 5-HT3 receptors, leading ultimately to a severe emetic response mediated via the medial solitary nucleus (reviewed in (Tyers and Freeman 1992)). Thus, 5-HT3 antagonists are commonly used as the gold standard in treating chemotherapy-induced nausea and vomiting (Billio et al. 2010). Although such an indirect stimulation of 5-HT3 receptor by cytotoxic drugs is believed to cause the nausea and vomiting, recently, we have shown some chemotherapy drugs directly modify the 5-HT-mediated current of 5-HT3 receptor (Nakamura et al. 2011). Here, we focus on an anticancer drug topotecan, a semi-synthetic derivative of the plant alkaloid, camptothecin (Kunimoto et al. 1987). Herein, we demonstrate that topotecan possesses the unique feature of having diametrically opposite actions on 5-HT-mediated 5-HT3 receptor current depending on the presence or absence of the 5-HT3B subunit. We also examined the effects of topotecan on Y129S polymorphic variant of 5-HT3B subunit.
Material and methods
All animal experiments were conducted in accordance with the institutional guidelines set by the Osaka University School of Medicine Animal Care and Use Committee. Every effort was made to minimize animal suffering and to reduce the number of animals used.
Synthesis of cRNA
The cDNA clones of human 5-HT3A and 5-HT3B subunits were obtained from OriGene Technologies, Inc. (Rockville, MD, USA). These clones were fully sequenced and checked using Gene WebIII software to confirm that the clones were NM_000869.2 (for the 5-HT3A subunit) and NM_006028.3 (for the 5-HT3B subunit). 5-HT3B(Y129S) was generated by site-direct mutagenesis, following the manufacturer's instructions (PrimeSTAR Mutagenesis Basal kit, Takara, Japan). These were then subcloned into a pBlueScriptII KS vector containing Xenopus β-globin (Krieg and Melton 1984) to increase its efficiency of translation in oocytes. cRNAs were synthesized in vitro using a mMESSAGE mMACHINE RNA transcription kit with T3 RNA polymerase, following the manufacturer's instructions (Ambion, Austin, TX, USA).
Frog surgery, oocyte isolation, and microinjection
Mature female Xenopus laevis frogs were purchased from Hamamatsu Seibutsu Kyozai (Hamamatsu, Japan). Preparation of oocytes and microinjection were carried out largely as described previously (Nakamura et al. 2011). In brief, frogs were anesthetized in ice-cold water and their ovarian lobes excised. These were then sequentially digested with 2 mg/mL collagenase type IA (Sigma Aldrich, St. Louis, MO, USA) in ND-96 calcium-free solution (96 mM NaCl, 2 mM KCl, 1 mM MgCl2, and 5 mM HEPES; pH 7.5) for 2 h on a rocking platform at (23°C). Mature stage V or VI oocytes were selected for microinjection. Oocytes were microinjected with human 5-HT3A cRNA (0.05 ng) or a combination of human 5-HT3A cRNA (0.05 ng) and a variable amount of human 5-HT3B cRNA (0.005, 0.025, 0.05, 0.075, 0.25, or 0.5 ng) depending on the experiment. All cRNA preparations for microinjection were dissolved in 100 mM KCl in a total injection volume of 25 nL. Following microinjection, oocytes were incubated in ND96 at 20°C until recording.
Two-electrode voltage clamping
Over a 16-h period after injection, oocytes were then impaled with voltage-sensing- and current-passing electrodes filled with 3 M KCl. Following implantation, oocytes were voltage-clamped to −60 mV, and experimental recording was initiated using a two-electrode voltage clamp technique employing a MEZ 7200 microelectrode amplifier and CEZ 1200 dual-electrode voltage clamp amplifier (Nihon Kohden, Tokyo, Japan). Data were acquired at 100 Hz with no low-pass filtering. Current response was digitized by a Power Lab 4/26 converter (AD instruments, Nagoya, Japan) and analyzed using Lab Chart v7 & Scope (AD instruments, Nagoya, Japan).
Compounds were applied by adding them to the superfusate. Stock solutions of 5 mM 5-HT (Sigma Aldrich) and 10 mM topotecan hydrochloride (Enzo Life Science, Plymouth Meeting, PA, USA) were dissolved in distilled water. To compare peak currents, all drugs were applied for 60 s.
To analyze the pharmacological mechanism by which topotecan influences 5-HT3A and 5-HT3AB receptor responses, peak 5-HT-induced current amplitudes were quantified. To illustrate response relationships for topotecan, current amplitudes in the presence of this compound were normalized to control 5-HT currents in its absence. All graphs were generated using GraphPad Prism v5 software (MDF Co. Ltd., Tokyo, Japan). All data are presented as mean ± SEM.
Inhibition and enhancement of 5-HT-gated current by topotecan depends 5-HT3B subunit
Topotecan is a semisynthetic derivative of camptothecin, a plant alkaloid (Kunimoto et al. 1987). The structure of topotecan and 5-HT are shown in Fig. 1. We found that the current stimulated through 5-HT3A receptors by 2 μM 5-HT (an EC20 concentration) was inhibited by topotecan at a high concentration (> 10 μM) (Fig. 2a). Topotecan itself was without effect at the 5-HT3A receptor (data not shown). Because absolute current amplitude varied between oocytes, these responses were normalized to the control (2 μM 5-HT) response. The IC50 for topotecan inhibition of 5-HT-gated currents was 114.1 μM (pIC50 = 3.9 ± 0.1) (Fig. 2c).
Several compounds are reported to modulate 5-HT-driven 5-HT3A receptor-mediated currents (Davies 2011), and some have been found to lose potency when the 5-HT3B subunit is introduced into the receptor complex (Davies et al. 1999; Das and Dillon 2003; Nakamura et al. 2011). We therefore examined the effect of topotecan on heteromeric 5-HT3AB receptors by injecting 5-HT3A cRNA and 5-HT3B cRNA in a 1 : 1 ratio. In oocytes injected this way, we found that, unlike other modulators, the 5-HT current via the 5-HT3AB receptor was actually enhanced by topotecan, although topotecan itself had no effect (Fig. 2b). The potentiating effect was topotecan dose-dependent, but somewhat bell-shaped, such that very high concentrations of topotecan (316.2 μM) produced smaller responses than 100 μM topotecan (Fig. 2c). These data showed that topotecan is a subunit-dependent modulator, which allows it to inhibit 5-HT-stimulated current in homomeric 5-HT3A receptors, but to enhance it in heteromeric 5-HT3AB receptors.
Topotecan is a competitive inhibitor at the 5-HT3A receptor
We examined whether topotecan is a competitive or non-competitive inhibitor of 5-HT3A receptors. A high concentration (100 μM) of 5-HT, either alone or in combination with 100 μM topotecan, was applied to oocytes expressing homomeric 5-HT3A receptors (Fig. 3a). Currents recorded in the presence and absence of topotecan were indistinguishable, although responses to 2 μM 5-HT were inhibited by the presence of topotecan (Figs 2a and 3b). The desensitization kinetics of 100 μM 5-HT-evoked currents were also unaffected by topotecan (Fig. 3c). These data demonstrate that topotecan acts as competitive antagonist of the 5-HT3A receptor.
Characteristics of topotecan-mediated modulation of 5-HT3AB receptor responses
Oocytes injected with 5-HT3A and 5-HT3B subunit cRNA should express both 5-HT3A and 5-HT3AB receptors. To eliminate confounding interference from the 5-HT3A receptors, we investigated the properties of the 5-HT3AB receptor at topotecan concentrations of less than 10 μM, at which this compound potentiates the 5-HT3AB receptor, but does not inhibit the 5-HT3A receptor (Fig. 2c). To examine whether topotecan can access its binding site without prior channel opening by 5-HT, oocytes expressing these receptors were incubated with 10 μM topotecan alone for 3 min prior to the application of 5-HT (Fig. 4a). The maximal response to 2 μM 5-HT following pre-incubation with topotecan was 123.8 ± 6.0% when compared with responses in control cells not pre-incubated with topotecan (100% with 2 μM 5-HT; p < 0.02), suggesting that topotecan does not require 5-HT3AB channel opening to access its binding site. Importantly, 10 μM topotecan itself did not potentiate the 5-HT3AB receptor (Fig. 4a), suggesting that topotecan is not an agonist for the 5-HT3AB receptor. Moreover, at a high concentration of 5-HT (100 μM), topotecan did not enhance the 5-HT-mediated 5-HT3AB receptor current (Fig. 4b and c). In support of this result, we simultaneously applied 100 μM topotecan (which shows maximum potentiation but might also modulate 5-HT3A receptor current (Fig. 2c)) and a high concentration of 5-HT (100 μM) to the 5-HT3AB receptor. Topotecan was still unable to change the current generated by 100 μM 5-HT (Fig. 4d and e). The desensitization kinetics of 100 μM 5-HT-evoked currents were also unaffected by topotecan (Fig. 4f). These data showed that the potentiation of the 5-HT3AB receptor response by topotecan is completely surmountable by 5-HT.
Current enhancement by topotecan is dependent on the 5-HT3B expression
The 5-HT3B subunit is a key factor in topotecan-mediated enhancement of 5-HT currents (Fig. 2). When experimenting under conditions designed to eliminate interference from 5-HT3A receptors (as mentioned above), the degree of potentiation of the 5-HT current by topotecan depends on the level of 5-HT3B subunit expression (Fig. 5). These data suggested that current modification through 5-HT3 receptors depends on the amount of 5-HT3B subunit incorporated in the receptor.
Reduction of topotecan-mediated enhancement of 5-HT3AB receptor by 5-HT3B(Y129S) mutation
A human variant 5-HT3B subunit, 5-HT3B (Y129S), is a worldwide high-frequency polymorphism (Krzywkowski et al. 2008). To examine the effect the topotecan of 5-HT3AB(Y129S) receptor, 5-HT3A cRNA and 5-HT3B or 5-HT3B(Y129S) cRNA was injected at 1 : 1, 1 : 1.5, and 1 : 10 ratio (Fig. 6a–c, Table 1). Topotecan-mediated potentiation curve of 5-HT3AB receptor was significantly different from that of 5-HT3AB(Y129S) receptor in all conditions. More than 10 μM topotecan showed mixed comportment caused by inhibition of 5-HT3A receptor and potentiation of 5-HT3AB receptor (Figs 2 and 5). Therefore, we focused on the effect at 10 μM topotecan concentration which can eliminate interference from homomeric 5-HT3A receptor. The topotecan-mediated enhancement was significantly reduced in 5-HT3AB(Y129S) receptor compared with 5-HT3AB receptor (Fig. 6d). These results showed that the tyrosine 129 of 5-HT3B subunit is involved in the response to topotecan-mediated enhancement of 5-HT3AB receptor.
Table 1. Summary of topotecan-mediated potentiation of control 2 μM 5-HT responses for 5-HT3AB and 5-HT3AB(Y129S) receptor. Concentration-response relationships were fitted with a logistic equation and are shown in Fig. 6, yielding the tabulated values
5-HT3A : B = 1 : 1
5.1 ± 0.1
202.7 ± 9.5
1.3 ± 0.4
5-HT3A : B(Y129S) = 1 : 1
5.0 ± 0.1
132.0 ± 4.6
6.0 ± 16.6
5-HT3A : B = 1 : 1.5
5.3 ± 0.1
185.4 ± 5.5
1.9 ± 0.6
5-HT3A : B(Y129S) = 1 : 1.5
5.3 ± 0.1
136.3 ± 4.4
2.3 ± 1.5
5-HT3A : B = 1 : 10
5.2 ± 0.03
185.4 ± 3.3
1.9 ± 0.6
5-HT3A : B(Y129S) = 1 : 10
4.9 ± 0.1
197.8 ± 11.9
2.3 ± 1.5
Topotecan exerts divergent and opposite effects on the 5-HT3 receptor depending on its subunit composition. This compound inhibited the 5-HT-gated current through the 5-HT3A receptor, but enhanced the current through receptors containing the 5-HT3B subunit (Fig. 2). 5-HT3 receptor currents are modulated by several inhibitors and enhancers (reviewed in (Davies 2011)). It has been reported that there is minimal difference between their pharmacological profiles at homomeric 5-HT3A and heteromeric 5-HT3AB receptors (Brady et al. 2001), although some drugs, including picrotoxin (Das and Dillon 2003), tubocurarine (Davies et al. 1999), and irinotecan (Nakamura et al. 2011) have less potency at inhibiting 5-HT3AB currents than at blocking 5-HT3A responses (Solt et al. 2005; Stevens et al. 2005). In contrast, our findings suggest that topotecan acts as a dual modulator of the 5-HT3 receptor and can either inhibit or potentiate it activity depending on its subunit composition. Until now, a 5-HT3AB receptor-selective enhancer has evaded identification. The discovery of these unique properties of topotecan suggests that this drug might be a good tool for distinguishing between 5-HT3A and 5-HT3AB receptors, which has been pharmacologically difficult.
Topotecan on 5-HT3AB receptor has the following properties: (i) topotecan alone cannot induce current; (ii) it can potentiate current induced by a low concentration of 5-HT; (iii) it does not increase the maximal response. A similar phenomenon has been reported as an effect of EtOH at the 5-HT3 receptor and chlordiazepoxide, a well-known benzodiazepine, at the GABAA receptor, and the molecular mechanisms underlying this change in potency are thought to be diverse (Lovinger and White 1991; Maksay et al. 2000). According to these reports, it is possible that topotecan could increase the affinity of 5-HT for its recognition site. Another possibility is that topotecan increases the probability of opening of the 5-HT3 receptor-associated ion channel. Such a change could account for the larger increase in current amplitude at lower agonist concentrations, where receptor occupancy and probability of channel opening are low under basal conditions.
Topotecan enhanced 5-HT-mediated current via the 5-HT3AB receptor, although effects diminished slightly at a very high (316 μM) topotecan concentration (Fig. 2c). The most likely explanation is that injection of 5-HT3A and 5-HT3B subunit cRNA into oocytes results in the coexistence of homomeric 5-HT3A and heteromeric 5-HT3AB receptors in single oocytes (Walstab et al. 2008). Therefore, we also examined the effects of topotecan in cells expressing a higher proportion of 5-HT3AB receptors (i.e. in which the ratio of injected 5-HT3A:5-HT3B subunit cRNA was 1 : 10), which would be expected to diminish effects of 5-HT3A receptor interference in the results. However, the maximum response to 2 μM 5-HT incubated with 100 and 316 μM topotecan was 170.3 ± 3.9 μM and 108.9 ± 1.7 μM, respectively. Therefore, the diminished effect did not seem to be caused only from homomeric 5-HT3A receptor. These biphasic modulations can arise from the existence of distinct high- and low-affinity sites mediating the potentiation and inhibition of the modulator, respectively (Thompson et al. 2002; Hapfelmeier et al. 2003; Trattnig et al. 2012). The 5-HT3AB receptor could bring about the potentiation observed at low concentrations of topotecan, whereas occupation of additional sites in the pentamer containing 5-HT3A subunits at the high concentrations might cause the inhibitory response.
In the presence of the 5-HT3B subunit, the 5-HT3A subunit is likely to form a heteromeric 5-HT3AB receptor rather than a homomeric 5-HT3A receptor (Walstab et al. 2008). If the 5-HT3A subunit is sufficiently in excess, the amount of 5-HT3B subunit can restrict the formation of 5-HT3AB receptors. In this study, current enhancement by topotecan almost reached a plateau when the 5-HT3B subunit is expressed 1.5 times more than 5-HT3A subunit (Fig. 5), implying that the 5-HT3AB receptor is composed of this ratio. This finding is consistent with previous study by Edwardson and colleagues, which used atomic force microscopy to show that the subunit stoichiometry in heteromeric 5-HT3AB receptors was 2A:3B (Barrera et al. 2005).
Most important finding of this article is that topotecan has the unique property of having opposite effects (enhancement or inhibition) depending on the presence or absence of the 5-HT3B subunit. Although the difference in ligand gating concentration could affect the magnitude of allosteric modulation, it is difficult to be explained the dynamic topotecan modulation by a slight difference in the responsiveness between 5-HT3A and 5-HT3AB receptor to 2 μM 5-HT (Davies et al. 1999). Such as a subunit-dependent dynamic modulation has been reported for GABAA receptors, where addition of the ε subunit of the GABAA receptor causes the functional response to tracazolate to change dramatically from a potentiatory to an inhibitory phenotype (Thompson et al. 2002). The presence of the 5-HT3B subunit may generate a new or altered binding site for topotecan that might change its mechanism of current modulation in the ways described above. Until now, there have been no modulators identified that can bind to the 5-HT3B subunit using its specific amino acid sequence, rather than the 5-HT3A subunit (Lochner and Lummis 2010; Thompson et al. 2011). However, one amino acid substitution from tyrosine to serine at residue 129 of 5-HT3B subunit showed greatly reduction of the toptoecan-mediated enhancement of 5-HT3AB receptor (Fig. 6). It has been reported that the surface expression of 5-HT3AB(Y129S) receptor was not inhibited, and the sensitivity to 5-HT of 5-HT3AB receptor is indistinguishable that of 5-HT3AB(Y129S) receptor (Krzywkowski et al. 2008). Therefore, Tyr129 of 5-HT3B subunit is one possible modification site for topotecan.
Topotecan is a chemotherapeutic agent usually given to treat ovarian cancer, small-cell lung cancer (SCLC), and solid cancer (O'Brien et al. 2007; Peng et al. 2008) and has been in clinical use for over 10 years. There are various regimens for the dosing (Peng et al. 2008). Although the plasma concentration is usually lower than 0.1 μM (Kobayashi et al. 2002), sometimes the peak topotecan concentration becomes to 2.5 μM (Andreopoulou et al. 2011). In addition, topotecan binds to tissues, especially in the gut and kidney (Shah and Balthasar 2011). Like antipsychotic drugs which accumulate at a restrict region for high concentration (Tischbirek et al. 2012), local concentrations of topotecan achieved in patients might be sufficient to modulate the 5-HT3AB receptor.
Another important finding of this article is that Tyr 129 of 5-HT3B subunit is modification site for topotecan. The 5-HT3AB(Y129S) receptor showed significant reduction of topotecan-mediated enhancement of 5-HT3AB receptor (Fig. 6). Actually, 5-HT3B(Y129S) is a worldwide high-frequency polymorphism (Krzywkowski et al. 2008). In addition, patients with the Ser allele had lower risk of developing nausea than patients with Tyr/Tyr genotype induced by paroxetine (Sugai et al. 2006). Because the activation of 5-HT3 receptor induces nausea and vomiting, the diminished current by 5-HT3B(Y129S) subunit might connect to the lower risk of nausea with the Ser allele. Despite of a low (10-30%) emetic potential of topotecan (Herrstedt and Dombernowsky 2007), some patients still suffer from severe nausea and vomiting after receiving this drug (Forbes et al. 2001). Until now there are no data which focused on the relationship between the severity of nausea and vomiting caused by topotecan and the polymorphism of 5-HT3B(Y129S). However, according to this fact, the directly modification of 5-HT3 receptor and the effect by 5-HT3B(Y129S) subunit might be a part of the reason for the individual differences in the severity of nausea and vomiting after chemotherapy.
In summary, we have shown that an established anticancer therapeutic agent, topotecan, has a previously unknown pharmacological feature: the ability to act as an inhibitor of 5-HT3A receptors, but as an enhancer of the 5-HT3AB receptors. The potentiating effect of topotecan on 5-HT3AB receptor was reduced in the receptors containing Y129S polymorphic variant of 5-HT3B subunit. Although stimulations of 5-HT3 receptors on vagal nerve by 5-HT released from enterochromaffin cells are considered to be main causes of chemotherapy-induced nausea and vomiting, the present findings suggest that topotecan could directly modify 5-HT3 receptors. Moreover, different sensitivity to topotecan of 5-HT3B subunit and polymorphic 5-HT3B(Y129S) subunit could also explain individual differences in sensitivity to topotecan-induced nausea and vomiting.
This study was supported by a Grant-in-Aid for Young Scientists (B) (23791895) from the Ministry of Education, Culture, Sports, Science and Technology of Japan. Authors have no conflict of interest. Yukiko Nakamura and Shoichi Shimada contributed to conceptual design of the experiments. Yukiko Nakamura contributed to acquisition and interpretation of data and article drafting. Yukiko Nakamura, Yusuke Ishida, Takahiro Yamada, Makoto Kondo, and Shoichi Shimada contributed to article revising and all authors approved the version for publication.