Abstract
Fluoxetine (Prozac) inhibited the membrane currents elicited by serotonin (5-hydroxytryptamine; 5HT) in Xenopus oocytes expressing either cloned 5HT2C receptors or 5HT receptors encoded by rat cortex mRNA. Responses of 5HT2C receptors, elicited by nM concentrations of 5HT, were rapidly and reversibly blocked by micromolar concentrations of fluoxetine. For responses elicited by 1 μM 5HT, the IC50 of fluoxetine inhibition was ≈20 μM. In accord with the electrophysiological results, fluoxetine inhibited the binding of [3H]5HT to 5HT2C receptors expressed in HeLa cells (Ki ≈ 65–97 nM), and the binding to 5HT receptors in rat cortex membranes was also inhibited but less efficiently (Ki ≈ 56 μM). Our results show that fluoxetine is a competitive and reversible antagonist of 5HT2C receptors and suggest that some therapeutic effects of fluoxetine may involve blockage of 5HT receptors, in addition to its known blockage of 5HT transporters. Similar work may help to design more selective compounds for use in the treatment of brain disorders.
Keywords: membrane currents, receptor binding, Xenopus oocytes, HeLa cells
Fluoxetine (Prozac) is widely used in the treatment of a variety of brain disorders, such as mental depression, panic disorder, obesity, and alcoholism. It is generally believed that fluoxetine exerts its therapeutic effects by enhancing serotonergic transmission, exclusively through inhibition of serotonin (5-hydroxytryptamine; 5HT) transporters with minimal or no effects on other neurotransmitter receptors (1, 2). However, it has been shown that fluoxetine inhibits 5HT binding in the choroid plexus (3), that it appears to be an agonist of 5HT2C receptors in cultured astrocytes (4), and that it inhibits currents mediated by 5HT3 receptors in rat nodose ganglion neurons (5), as well as the binding of 5HT to 5HT3 and 5HT4 receptors (6). In addition, chronic treatment of fluoxetine may cause a down-regulation of 5HT1 receptors (2) and also alter the expression of other receptors, although the latter effects are somewhat controversial (7).
5HT2C receptors [formerly termed 5HT1C, (8)] are widely expressed in the brain and spinal cord, are particularly enriched in the choroid plexus (9), and appear to mediate many important effects of 5HT. For example, transgenic mice that are devoid of 5HT2C receptors are overweight and are prone to seizure-induced death, suggesting a role for this type of receptor in the control of appetite and neuronal network excitability (10). Given fluoxetine’s multiple therapeutic effects on a variety of mental and eating disorders, it was important to study in more detail the action of fluoxetine on 5HT2C receptors.
MATERIALS AND METHODS
RNA in Vitro Transcription.
NotI-linearized pSR1c (11) was transcribed using T7 RNA polymerase (Promega) in the presence of a cAMP-binding protein analog m7G(5′)ppp(5′)G (Pharmacia). Rat cortex RNA was extracted using the acid guanidinium thiocyanate-phenol-chloroform method (12), and poly(A)+ RNA was obtained by oligo(dT)-cellulose chromatography.
Translation in Xenopus Oocytes and Electrophysiological Recording.
Oocytes were injected with mRNA, and recordings were made 4–10 days later, essentially as described (13–15). Briefly, oocytes were injected with 1 ng of cloned rat 5HT2C mRNA, or with 50 ng of rat cortex mRNA, and kept in Barth’s medium containing 0.01 mg/ml gentamicin. Two days later, the oocytes were treated with collagenase to remove the follicular and other enveloping cells (13, 15). During the subsequent days, membrane currents were recorded, usually with the membrane potential clamped at −60 mV, digitized, and stored for analyses. Drugs were applied via continuous bath superfusion of Ringer’s solution at 5–8 ml/min (bath volume ≈ 100 μl).
Transfection and Membrane Preparation.
The EcoRI fragment of pSR1c, containing the entire rat 5HT2C receptor coding region, was subcloned into a eukaryotic expression vector pcDNA3 (Invitrogen), and transfected into HeLa cells by electroporation (Bio-Rad Gene Pulser, 500 μF, 300 V). Cell membranes were prepared according to Albert et al. (16) with slight modifications. Briefly, 2 days after transfection, the cells were harvested in a hypotonic buffer (50 mM Tris·HCl, pH 7.4/1 mM MgCl2) and precipitated at 25,000 × g for 30 min at 4°C. Membrane pellets were washed once, resuspended in a receptor binding assay buffer (50 mM Tris·HCl, pH 7.4/4 mM CaCl2/0.1% ascorbic acid), and stored at −70°C until use. Membrane protein was measured as described by Bradford (17).
Rat cerebral cortex membranes were prepared by homogenizing dissected cortex with a loose fitting Polytron in 0.32 M sucrose, precipitating at 800 × g for 10 min, and then reprecipitating the supernatant at 25,000 × g for 30 min. The membrane pellet was washed twice with 1 mM EGTA and resuspended in the receptor binding buffer.
Receptor Binding Assay.
Cell membranes (25–50 μg protein) were incubated with [3H]5HT (≈75 Ci/mmol; 1 Ci = 37 GBq; Amersham) in the binding assay buffer (200 μl) at 4°C for 30 min. Nonspecific binding was determined by adding 100 μM 5HT together with [3H]5HT. Binding assays were terminated by centrifugation, and the membranes were solubilized with a tissue solubilizer (TS-2, Research Product International) before quantification by liquid scintillation spectrometry (≈30% efficiency). In competition assays, different concentrations of fluoxetine were used in the reaction. Ki was calculated from the equation: Ki = IC50/(1 + radioligand concentration/Kd of the radioligand) (18), where IC50 is the concentration of the competing ligands required for 50% inhibition of the radioligand binding. Data were analyzed using the program prism (GraphPad Software, San Diego).
RESULTS
Effects of Fluoxetine on 5HT2C Receptor-Mediated Responses.
Defolliculated oocytes from Xenopus frogs do not have native 5HT receptors, although, very rarely, 5HT elicits small oscillatory currents in some oocytes (R.M., unpublished results). In the experiments reported here, all the control, noninjected oocytes from more than 20 donor frogs did not respond to 5HT. In contrast, oocytes injected with 5HT2C mRNA gave large membrane currents in response to 5HT. Fluoxetine applied alone at concentrations up to 100 μM did not induce a membrane current response or, occasionally, it elicited a very small inward current (Fig. 1). When fluoxetine was applied together with 5HT, the responses to 10 nM 5HT were completely abolished and those to 10 μM 5HT, a concentration that elicits near maximal responses in Xenopus oocytes, were greatly attenuated (Figs. 1 and 2A). The blocking effect of fluoxetine was rapid in onset and was also rapidly reversible. For example, in Fig. 1, the response to 10 μM 5HT had already recovered to near its control level 17 min after removal of fluoxetine (Fig. 1, lower trace). However, in some oocytes, the responses to 5HT did not recover fully even 1 h after exposure to high concentrations of fluoxetine. A dose-response analysis of fluoxetine inhibition of the membrane current responses elicited by 1 μM 5HT gave an IC50 of ≈20 μM (Fig. 3).
It is known that injection of rat cerebral cortex mRNA into oocytes leads to the expression of functional 5HT receptors (19), whose molecular types have not yet been clearly established, although the 5HT2C receptor seems to be the predominant 5HT receptor subtype in the rat central nervous system (20). Therefore, we decided to see if the 5HT receptors expressed by rat cortex mRNA were also blocked by fluoxetine. Here again, fluoxetine (100 μM) blocked completely the responses to 10 nM 5HT and greatly reduced the responses to 10 μM 5HT (Fig. 2B).
Fluoxetine was an effective blocker even at relatively low concentrations. For example, in the oocyte used for Fig. 4, 1 μM fluoxetine exerted a rapid and fully reversible inhibition of the response to 2 nM 5HT. On average, the amplitude of the response to 2 nM 5HT was reduced ≈50% by a coapplication with 1 μM fluoxetine, and a second coapplication of fluoxetine exerted a similar inhibitory effect. Moreover, when 1 μM or 10 μM fluoxetine was applied briefly during prolonged application of low concentrations of 5HT, the responses were blocked rapidly and almost completely (Fig. 5 A and C), and the 5HT current recovered rapidly after removing the fluoxetine (Fig. 5A).
Effects of Fluoxetine on Other Receptors Also Linked to the Phosphoinositide Pathway.
The membrane current responses to 5HT in Xenopus oocytes, injected with either rat cortex mRNA or cloned 5HT2C RNA, result from activation of an endogenous receptor-channel coupling pathway. The binding of 5HT to 5HT2C receptors activates the phosphoinositide pathway via a G protein, thus leading to the formation of inositol triphosphate and the release of Ca2+ from intracellular stores (21). This Ca2+ in turn opens Ca2+-gated Cl− channels (14, 19, 22, 23). Therefore, it was possible that fluoxetine inhibited the membrane currents elicited by 5HT by acting at any of the multiple steps on the receptor-channel coupling pathway. To examine this possibility we used two approaches. In one, we studied the effects of fluoxetine on the responses mediated by native or expressed angiotensin II receptors or by native serum factor receptors, all of which activate the same phosphatidylinositol pathway used by the 5HT2C receptors (24, 25). In the other, we studied the effect of fluoxetine on the binding of 5HT to the 5HT2C receptors.
As illustrated in Fig. 5, the responses to 0.3 nM 5HT were almost completely blocked by a 1 μM concentration of fluoxetine (Figs. 5 A and C). In contrast, the response to 1 nM angiotensin III was not inhibited, and may even be increased slightly, by a 105-fold higher concentration of fluoxetine (100 μM) (Fig. 5B). Similarly, the oscillatory current responses elicited by serum were not blocked by 100 μM fluoxetine (Fig. 6). All this suggested strongly that fluoxetine did not affect appreciably the phosphatidylinositol receptor-channel coupling pathway. Therefore, it appeared very likely that the inhibiting action of fluoxetine on the responses to serotonin was exerted at the receptor level.
Effects of Fluoxetine on [3H]5HT Binding to 5HT2C Receptors.
To determine whether fluoxetine affects directly the binding of 5HT to its receptors, we examined the effects of fluoxetine on the binding of [3H]5HT to membranes from HeLa cells transiently expressing 5HT2C receptors and to membranes from rat cerebral cortex. To reduce complications from potential binding to 5HT transporters, all these experiments were carried out at 4°C and in a Na+-free binding assay buffer. Under these conditions, the binding of fluoxetine to 5HT transporters should be less than 0.5% of that occurring under more physiological conditions (37°C and with Na+ present) (26). Fluoxetine inhibited the binding of [3H]5HT to 5HT2C receptors with a relatively high efficiency. Thus, when 3.2 nM [3H]5HT was used to label the 5HT2C receptors, fluoxetine inhibited the binding at concentrations as low as 10 nM, with an IC50 of 0.11 ± 0.01 μM (Ki ≈ 65 nM, n = 3). At close to saturating concentrations for binding of [3H]5HT (18 nM), the IC50 was 0.51 ± 0.06 μM (Ki ≈ 97 nM, n = 3). At both concentrations of [3H]5HT, the specific binding was completely inhibited by 100 μM fluoxetine (Fig. 7). In membranes from HeLa cells expressing 5HT2C receptors, the Kd of [3H]5HT binding was 4.3 ± 0.8 nM (n = 4), with Bmax ranging from 1.4 to 6.8 pmol/mg protein. There was no specific [3H]5HT binding in nontransfected HeLa cell membranes.
In agreement with the electrophysiological results, fluoxetine also blocked the binding of [3H]5HT to rat cortical membranes, although with a much lower efficiency as compared with its effect on the 5HT2C receptors. As shown in Fig. 7, a higher concentration of fluoxetine (>1 μM) was needed to detect inhibition of [3H]5HT binding to rat cortical membranes and the IC50 was 103 ± 18 μM (Ki ≈ 56 μM, n = 5).
DISCUSSION
This study shows clearly that fluoxetine has a potent blocking effect on 5HT2C receptors expressed in both Xenopus oocytes and HeLa cells, as well as on the native 5HT receptors present in cerebral cortical cell membranes. The evidence is many fold. (i) In Xenopus oocytes expressing cloned 5HT2C receptors, fluoxetine inhibited rapidly the currents elicited by 5HT. Appreciable inhibition was observed with fluoxetine concentrations lower than 1 μM, and the inhibition was rapidly reversible. For responses induced by 1 μM 5HT, the IC50 of fluoxetine inhibition was ≈20 μM. (ii) Fluoxetine did not block the oocyte responses to cloned (angiotensin II) or native serum receptors that are mediated by the same receptor-channel coupling pathway used by 5HT. (iii) Fluoxetine did not have an agonist action on 5HT2C receptors in oocytes because it failed to elicit appreciable currents. (iv) In HeLa cell membrane preparations, the binding of [3H]5HT to expressed rat 5HT2C receptors was inhibited completely by fluoxetine, with a Ki of ≈65–97 nM. Therefore, in oocytes, as well as in membranes of HeLa cells, fluoxetine acts as a reversible competitive antagonist of recombinant 5HT2C receptors.
In addition, our study shows that fluoxetine also blocks the responses of 5HT receptors expressed in oocytes from rat cortex mRNA, and inhibits also the binding of [3H]5HT to 5HT receptors present in rat cortical membranes. So far, about 10 different subtypes of 5HT receptors have been found in rat brain (27), including the 5HT2C receptor. Our results show that the Ki of fluoxetine for rat cortex 5HT receptors is very low compared with that for 5HT2C receptors (56 μM versus 65 nM). This suggests that fluoxetine has a stronger influence on the 5HT2C receptor than on the other subtypes of 5HT receptors present in rat cortex. Furthermore, this disparity in inhibitory potency between recombinant 5HT2C receptors and rat cortical 5HT receptors suggests that the main 5HT receptor in the rat cortex may not be of the 5HT2C type.
So far, the therapeutic effects of fluoxetine have been attributed primarily to its inhibition of 5HT transporters. Interestingly, it has been shown that the therapeutic plasma concentration of fluoxetine is in the micromolar range (28), and our studies show that, at this concentration range, fluoxetine can potently inhibit the membrane current responses mediated by 5HT2C receptors. Moreover, the affinity of fluoxetine for 5HT2C receptors (Ki = 65 nM) is close to its affinity for 5HT transporters (Ki = 33 nM) (29), which is also well below the therapeutic plasma concentration of fluoxetine. Thus, some therapeutic effects of fluoxetine may be a consequence of blocking both 5HT transporters and 5HT2C receptors. It should be noted that the blockage of 5HT transporters and that of 5HT2C receptors would have opposing actions on serotonergic synaptic transmission. Moreover, in addition to its effects at serotonergic synapses fluoxetine may exert important actions via volume transmission (30) at sites far away from the synaptic regions. In these areas, probably containing extra-junctional receptors (31) of many types, the extracellular concentration of 5HT is very likely much lower, and more sustained, than the concentrations reached within the synaptic gap. Thus, the volume transmission effect of fluoxetine on the extra-junctional receptors may resemble its effect on oocytes exposed to low concentrations of 5HT, where 1 μM fluoxetine blocked almost completely the response to 5HT. Because of the highly nonlinear dose/response relationship of 5HT2C receptors the blockage of even a small number of receptors in a cell would lead to very profound changes, not only in its responses to 5HT but also in those to other neurotransmitters which act on receptors linked to the same phosphatidylinositol receptor-channel coupling mechanism (32).
Thus, the mechanisms of the medicinal actions of fluoxetine appear to be more complicated than hitherto anticipated. The results presented here may help not only to advance our understanding of the therapeutic mechanisms of fluoxetine and related drugs, but also in the development of new families of drugs that could lead to improved treatments for depression and other dysfunction of the brain.
Acknowledgments
We thank Dr. D. J. Julius for the pSR1c, Eli Lilly Company and Dr. R. E. Purdy for the fluoxetine HCl, and Drs. L. Pulakat and G. M. Deng for preparing the bovine angiotensin II receptor mRNA. This work was supported by a grant from the National Institute of Neurological Disorders and Stroke (NS 23284).
ABBREVIATION
- 5HT
5-hydroxytryptamine (serotonin)
References
- 1.Cusack B, Nelson A, Richelson E. Psychopharmacology (Berlin) 1994;114:559–565. doi: 10.1007/BF02244985. [DOI] [PubMed] [Google Scholar]
- 2.Wong D T, Bymaster F P, Engleman E A. Life Sci. 1995;57:411–441. doi: 10.1016/0024-3205(95)00209-o. [DOI] [PubMed] [Google Scholar]
- 3.Wong D T, Threlkeld P G, Robertson D W. Neuropsychopharmacology. 1991;5:43–47. [PubMed] [Google Scholar]
- 4.Chen Y, Peng L, Zhang X, Stolzenburg J U, Hertz L. Brain Res Bull. 1995;38:153–159. doi: 10.1016/0361-9230(95)00082-p. [DOI] [PubMed] [Google Scholar]
- 5.Fan P. Neuroscience. 1994;62:515–522. doi: 10.1016/0306-4522(94)90384-0. [DOI] [PubMed] [Google Scholar]
- 6.Lucchelli A, Santagostino B M, Barbieri A, Candura S M, Tonini M. Br J Pharmacol. 1995;114:1017–1025. doi: 10.1111/j.1476-5381.1995.tb13307.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Beasley C M, Masica D N, Potvin J H. Psychopharmacology (Berlin) 1992;107:1–10. doi: 10.1007/BF02244958. [DOI] [PubMed] [Google Scholar]
- 8.Martin G R, Humphrey P P. Neuropharmacology. 1994;33:261–273. doi: 10.1016/0028-3908(94)90058-2. [DOI] [PubMed] [Google Scholar]
- 9.Mengod G, Nguyen H, Le H, Weber C, Lubbert H, Palacios J M. Neuroscience. 1990;35:577–591. doi: 10.1016/0306-4522(90)90330-7. [DOI] [PubMed] [Google Scholar]
- 10.Tecott L H, Sun L M, Akana S F, Strack A M, Lowenstein D H, Dallman M F, Julius D. Nature (London) 1995;374:542–546. doi: 10.1038/374542a0. [DOI] [PubMed] [Google Scholar]
- 11.Julius D, MacDermott A B, Axel R, Jessell T M. Science. 1988;241:558–564. doi: 10.1126/science.3399891. [DOI] [PubMed] [Google Scholar]
- 12.Chomczynski P, Sacchi N. Anal Biochem. 1987;162:156–159. doi: 10.1006/abio.1987.9999. [DOI] [PubMed] [Google Scholar]
- 13.Kusano K, Miledi R, Stinnakre J. J Physiol (London) 1982;328:143–170. doi: 10.1113/jphysiol.1982.sp014257. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Miledi R. Proc R Soc London B. 1982;215:491–497. doi: 10.1098/rspb.1982.0056. [DOI] [PubMed] [Google Scholar]
- 15.Miledi R, Woodward R M. J Physiol (London) 1989;416:601–621. doi: 10.1113/jphysiol.1989.sp017780. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Albert P R, Zhou Q Y, Van T H, Bunzow J R, Civelli O. J Biol Chem. 1990;265:5825–5832. [PubMed] [Google Scholar]
- 17.Bradford M M. Anal Biochem. 1976;72:248–254. doi: 10.1016/0003-2697(76)90527-3. [DOI] [PubMed] [Google Scholar]
- 18.Cheng Y, Prusoff W H. Biochem Pharmacol. 1973;22:3099–3108. doi: 10.1016/0006-2952(73)90196-2. [DOI] [PubMed] [Google Scholar]
- 19.Gundersen C B, Miledi R, Parker I. Proc R Soc London B. 1983;219:103–109. doi: 10.1098/rspb.1983.0062. [DOI] [PubMed] [Google Scholar]
- 20.Molineaux S M, Jessell T M, Axel R, Julius D. Proc Natl Acad Sci USA. 1989;86:6793–6797. doi: 10.1073/pnas.86.17.6793. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Berridge M J. Nature (London) 1993;361:315–325. doi: 10.1038/361315a0. [DOI] [PubMed] [Google Scholar]
- 22.Miledi R, Parker I. J Physiol (London) 1984;357:173–183. doi: 10.1113/jphysiol.1984.sp015495. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Parker I, Miledi R. Proc R Soc London B. 1987;231:27–36. doi: 10.1098/rspb.1987.0033. [DOI] [PubMed] [Google Scholar]
- 24.Tigyi G, Dyer D, Matute C, Miledi R. Proc Natl Acad Sci USA. 1990;87:1521–1525. doi: 10.1073/pnas.87.4.1521. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Woodward R M, Miledi R. Proc R Soc London B. 1991;244:11–19. doi: 10.1098/rspb.1991.0044. [DOI] [PubMed] [Google Scholar]
- 26.Chang A S, Frnka J V, Chen D N, Lam D M. Proc Natl Acad Sci USA. 1989;86:9611–9615. doi: 10.1073/pnas.86.23.9611. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Boess F G, Martin I L. Neuropharmacology. 1994;33:275–317. doi: 10.1016/0028-3908(94)90059-0. [DOI] [PubMed] [Google Scholar]
- 28.Goodnick P J. Psychopharmacol Bull. 1991;27:503–512. [PubMed] [Google Scholar]
- 29.Hoffman B J, Mezey E, Brownstein M J. Science. 1991;254:579–580. doi: 10.1126/science.1948036. [DOI] [PubMed] [Google Scholar]
- 30.Fuxe K A, Agnati LF, editors. Volume Transmission in the Brain: Novel Mechanisms for Neural Transmission. New York: Raven; 1991. [Google Scholar]
- 31.Miledi R. J Physiol (London) 1960;151:24–30. [PMC free article] [PubMed] [Google Scholar]
- 32.Miledi R, Parker I, Sumikawa K. In: Fidia Research Foundation Neuroscience Award Lectures Series. Smith J, editor. New York: Raven; 1989. pp. 57–90. [Google Scholar]