Chester Clinic

 

Ajplung.physiology.org

Effects of fluoxetine, phentermine, and venlafaxineon pulmonary arterial pressure and electrophysiology HELEN L. REEVE,1,2 DANIEL P. NELSON,3 STEPHEN L. ARCHER,4AND E. KENNETH WEIR2,3Departments of 1Physiology and 2Medicine, University of Minnesota, Minneapolis 55455;3Department of Cardiology, Veterans Affairs Medical Center, Minneapolis, Minnesota 55417;and 4Department of Medicine, University of Alberta, Edmonton, Alberta, Canada T6G 2R7 Reeve, Helen L., Daniel P. Nelson, Stephen L. Archer,
PPH by an odds ratio of 23 (1). In 1996, it was reported and E. Kenneth Weir. Effects of fluoxetine, phentermine,
that the total number of prescriptions for the anorexic and venlafaxine on pulmonary arterial pressure and electro- combination of fenfluramine and another anorexic physiology. Am. J. Physiol. 276 (Lung Cell. Mol. Physiol. 20): agent, phentermine (Fen-Phen), was .18 million in the L213–L219, 1999.—The anorexic agents dexfenfluramine United States (8). Unfortunately, despite the wide- and fenfluramine plus phentermine have been associated spread use of anorexic agents, the mechanism by which with outbreaks of pulmonary hypertension. The fenflur- they may cause PPH remains unclear. The fenflur- amines release serotonin and reduce serotonin reuptake in amines cause serotonin release from neurons (21) and neurons. They also inhibit potassium current (IK), causing reduce reuptake, whereas phentermine inhibits seroto- membrane potential depolarization in pulmonary arterialsmooth muscle cells. The recent withdrawal of the fenflur- nin metabolism (27). It has been proposed that high amines has led to the use of fluoxetine and phentermine as an levels of serotonin might initiate pulmonary hyperten- alternative anorexic combination. Because fluoxetine and venlafaxine reduce serotonin reuptake, we compared the Weir et al. (31) reported that aminorex, fenfluramine, effects of these agents with those of phentermine and dexfen- and dexfenfluramine cause dose-dependent inhibition fluramine on pulmonary arterial pressure, IK, and membrane of the outward potassium current (IK) in isolated potential. Fluoxetine, venlafaxine, and phentermine caused pulmonary arterial (PA) smooth muscle cells (SMCs) minimal increases in pulmonary arterial pressure at concen- and that dexfenfluramine depolarizes the cell mem- trations , 100 µM but did cause a dose-dependent inhibition brane potential. More recently, it has been shown that of IK. The order of potency for inhibition of IK at 150 mV was dexfenfluramine inhibits the voltage-dependent potas- fluoxetine . dexfenfluramine 5 venlafaxine . phentermine.
Despite the inhibitory effect on I V) channel Kv2.1 (25), which may contribute to K at more positive membrane resting membrane potential (RMP) in PASMCs (6). The potentials, fluoxetine, venlafaxine, and phentermine, in con-trast to dexfenfluramine, had minimal effects on the cell same doses that inhibit IK also cause pulmonary vaso- resting membrane potential (all at a concentration of 100 constriction in isolated rat lungs, which is further µM). However, application of 100 µM fluoxetine to cells that enhanced after the inhibition of nitric oxide (NO) had been depolarized to 230 mV by current injection elicited synthase (31). Because inhibition of IK, membrane a further depolarization of .18 mV. These results suggest depolarization, and the resulting increase in intracellu- that fluoxetine, venlafaxine, and phentermine do not inhibit lar Ca21 concentration are thought to underlie hypoxic IK at the resting membrane potential. Consequently, they pulmonary vasoconstriction (30), Weir et al. (31) sug- may present less risk of inducing pulmonary hypertension gested that these drugs might initiate anorexic-induced than the fenfluramines, at least by mechanisms involving pulmonary hypertension in susceptible patients by a similar mechanism.
anorexic; serotonin; potassium channels; pulmonary hyperten- Despite the recent withdrawal of fenfluramine and sion; membrane potential dexfenfluramine because of their association with carci-noid syndrome-like cardiac valve disease (8), there hasalready been a move to replace them with new agentssuch as the combination of fluoxetine (Prozac) and THE AMPHETAMINE-LIKE ANOREXIC AGENT aminorex was phentermine (Pro-Phen). Fluoxetine is a serotonin reup- associated with an epidemic of pulmonary hyperten- take inhibitor (28) and venlafaxine inhibits reuptake of sion in Austria, Germany, and Switzerland between serotonin and norephinephrine (17). To determine 1967 and 1972 (13, 22). More recently, a similar epi- whether these agents might have membrane effects demic of primary pulmonary hypertension (PPH) oc- similar to the fenfluramines, we investigated the effects curred after the use of two other chemically related of fluoxetine, venlafaxine, and phentermine on PA anorexic agents, fenfluramine and its D-isomer dexfen- pressures in isolated, perfused rat lungs and on IK and fluramine (7). An epidemiologic study carried out in membrane potential recorded from isolated rat PASMCs Europe between 1992 and 1994 showed that the use of and compared them with those with dexfenfluramine.
fenfluramine for .3 mo increased the risk of developing Isolated perfused rat lungs. Male Sprague-Dawley rats The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby 6 5 g; n 5 59) were anesthetized with pentobarbital marked ‘ advertisement' in accordance with 18 U.S.C. Section 1734 sodium (50 mg/kg body wt ip). The rats were intubated with solely to indicate this fact.
PE-200 tubing (ID 1.44 mm, OD 1.90 mm), a thorocatomy was ELECTROPHYSIOLOGY AND PULMONARY VASOCONSTRICTION performed, and the animal was heparinized (100 units). The potential, the cells were held in current clamp at either their pulmonary artery was cannulated with a double-lumen can- RMP (244 6 3 mV; n 5 21) or a potential of 230 mV. The nula so perfusion and pressure measurements could be baseline was recorded for at least 1 min to ensure stability.
obtained simultaneously. The left atrium was cannulated for Data were recorded and analyzed with pClamp 6.04 software efferent flow in a recirculating manner at a rate of 0.04 ml · (Axon Instruments, Foster City, CA).
min21 · g body wt21. Fifty milliliters of Krebs solution contain- Drugs used. Dexfenfluramine, phentermine, fluoxetine, ing 4% albumin and 5 µg/ml of meclofenamate were used for and ketanserin were obtained from RBI (Natick, MA). Venla- the perfusate. The lungs were ventilated with humidified faxine (Effexor) was a gift from Knoll Pharmaceutical (Mt.
gases containing 20% O Olive, NJ). The drugs were dissolved in normal saline, except 2-5% CO2-balance N2 (normoxia) or ketanserin that was dissolved in 1 part ethanol to 4 parts 2-5% CO2-balance N2 (hypoxia). The lung chamber and perfusate were maintained at 37°C. Respiration was set to normal saline. All other drugs and salts were obtained from physiological values (frequency 70 breaths/min; tidal volume Sigma (St. Louis, MO). Vehicle controls were done for all 1.5 ml), with a positive end-respiratory pressure of 2.5 Statistics. Data are expressed as means 6 SE. The effects 2O. To determine lung reactivity, the lungs were sub- jected to two consecutive cycles, each consisting of 10 min of of drugs on IK and PA pressure were compared with a normoxia, a bolus injection of angiotensin II (0.15 µg) into the repeated-measures ANOVA (Staview II, version 4.0, Abacus afferent line, and, after 8 min, a 6-min hypoxic challenge.
Concepts). Membrane potential data were compared with Lungs were only accepted for study if they had a pressor Student's paired t-test. A value of P , 0.05 was considered response . 8 mmHg to hypoxia. After a return to baseline, the lungs were given an NO synthase inhibitor [N-nitro-L- arginine methyl ester (L-NAME); 50 µM] and perfused for afurther 20 min. At this point, increasing doses of the test Fluoxetine, phentermine, and venlafaxine effects on drugs (0.1, 1, and 10 µM) were administered at 5-min PA pressure. There was minimal effect on PA pressure intervals. The type 2 5-hydroxytryptamine (5-HT2)-receptor at concentrations of ,10 µM for all drugs tested.
antagonist ketansarin (1 µM) or vehicle was then adminis- Dexfenfluramine caused a small constriction at 10 µM, tered. We have found that this concentration of ketanserin whereas venlafaxine, fluoxetine, and phentermine had prevents the vasoconstriction caused by 100 µM serotonin(data not shown). After a further 10 min, drugs were given minimal effect on baseline pressures at this concentra- again at concentrations of 10 and 100 µM. To determine tion (Fig. 1A). At a dose of 100 µM, dexfenfluramine responsiveness of the lungs to a combination of phentermine and fluoxetine caused significant constriction. Pretreat- and dexfenfluramine (to mimic the use of Fen-Phen), the ment of the lungs with the 5-HT2 blocker ketanserin lungs were given two consecutive doses of phentermine (10 (1 µM) caused no significant reduction in the response and 100 µM) or vehicle after L-NAME, followed by two doses to dexfenfluramine (Fig. 1A). Lungs treated with phen- of dexfenfluramine (10 and 100 µM).
termine (10 and 100 µM) constricted significantly more Cell dispersal. Rat PASMCs were obtained fresh on each to subsequent doses of dexfenfluramine (10 and 100 day of experimentation. Male Sprague-Dawley rats (316 6 µM) than control lungs given vehicle before dexfenflur- 14 g; n 5 30) were anesthetized with 50 mg/kg of pentobarbi- amine (Fig. 1B).
tal sodium, and the heart and lungs were removed en bloc.
Whole cell I Fourth-, fifth-, and sixth-generation pulmonary arteries were K recorded from PASMCs with the conven- dissected free and placed in Ca21-free Hanks' solution com- tional and perforated-patch clamp. IK recorded from posed of (in mM) 145 NaCl, 4.2 KCl, 1.0 MgCl single PASMCs (average cell capacitance 8.6 6 0.2 pF; 10 HEPES, and 0.1 EGTA (pH 7.4) for 10 min at 4°C. The n 5 67) with the conventional whole cell configuration arteries were then transferred to Hanks' solution containing were typically fast activating and slowly inactivating, 1 mg/ml of papain, 0.75 mg/ml of albumin, and 0.85 mg/ml of with an average current amplitude of 2,326 6 183 pA at dithiothreitol without EGTA and kept at 4°C for 17 min. After 150 mV (n 5 27 cells). Currents recorded with the this time, the arteries were incubated at 36°C for 10 min. The perforated-patch clamp, which prevents dialysis of the arteries were washed in enzyme-free Hanks' solution and cell cytosol, displayed similar kinetics and were not maintained at 4°C. Several digestions were done each day to significantly different in amplitude (n 5 7 cells). Cur- ensure cell viability. Gentle trituration produced a cell suspen- rents were inhibited by 1 and 2 mM 4-aminopyridine sion that was divided into aliquots in a perfusion chamber on (4-AP), suggesting that they were primarily due to the stage of an inverted microscope (Diaphot 200, Nikon) forwhole cell patch-clamp studies (14). The cells were allowed to activation of 4-AP-sensitive KV channels (Fig. 2A).
adhere to the bottom of the organ bath for several minutes Fluoxetine, phentermine, and venlafaxine inhibition before perfusion with a solution composed of (in mM) 145 of IK. Fluoxetine and venlafaxine caused dose-depen- NaCl, 5.4 KCl, 1.0 MgCl dent and reversible inhibition of I 2, 1.5 CaCl2, 10 HEPES, and 10 K recorded from single glucose (pH 7.4 with NaOH). For conventional whole cell PASMCs. Dose-response curves (1–100 µM) were con- recordings, electrodes were filled with a solution of (in mM) structed for both drugs and compared with those 140 KCl, 1.0 MgCl2, 5 HEPES, 1 EGTA, and 1 ATP (dipotas- obtained with dexfenfluramine (Fig. 2B). Because phen- sium salt) (pH 7.2 with KOH). For perforated-patch record- termine had minimal effect on IK even at 100 µM, a ings (26), ATP was omitted from the pipette solution and dose-response curve was not constructed. Venlafaxine amphotericin B was included at a final concentration of 120 and dexfenfluramine inhibited a similar percentage of µg/ml. The electrodes had a resistance of 2–3 MV after beingfire polished. All drugs were applied via the extracellular IK at all concentrations tested. Fluoxetine inhibited a perfusate at a rate of 1–2 ml/min at room temperature significantly greater percentage of the total IK and, at a (21–23°C). For voltage-clamp experiments, the cells were membrane potential of 150 mV, almost completely held at a potential of 270 mV and stepped to more depolar- eliminated the current at 30 µM (79.8 6 3.0% inhibi- ized potentials in 120-mV steps. For recordings of membrane tion; n 5 6 cells; Fig. 2B), with no additional inhibition ELECTROPHYSIOLOGY AND PULMONARY VASOCONSTRICTION Fig. 1. A: changes (D) in pulmonary arterial pressure(Ppa) measured after increasing doses of venlafaxine, dexfenfluramine, and fluoxetine before and after admin-istration of ketanserin (arrow) in isolated perfused ratlungs. Values are means 6 SE; n, no. of lungs. B: Ppameasured during indicated interventions in isolatedperfused rat lungs. Control, vehicle treated; ANG II,angiotensin II (0.15 µg); L-NAME, N-nitro-L-argininemethyl ester; Phen, phentermine; Dex, dexfenflur-amine. Values between interventions indicate baseline measurements pre- and postintervention. Values aremeans 6 SE; n, no of lungs. Note that ANG II andhypoxia have similar responses in both sets of lungs.
* P , 0.05 compared with control value.
at 100 µM. The EC50 of fluoxetine for the inhibition of IK 3B). Venlafaxine (100 µM) had little or no effect on cells at 150 mV was calculated as 4.3 µM. Inhibition of IK by held either at their RMP (n 5 6; Fig. 3A) or at 230 mV dexfenfluramine (30 µM) with the perforated patch- (n 5 3; Fig. 3B). One hundred micromolar phentermine clamp technique was not significantly different from had no effect on RMP (n 5 4; Fig. 3A) but caused a small that found with the conventional whole cell technique depolarization if cells were predepolarized to 230 mV (data not shown).
(n 5 3; Fig. 3B). Dexfenfluramine (100 µM) depolarized Fluoxetine, phentermine, and venlafaxine modula- PASMCs from their RMP by ,14 mV (n 5 4; Fig. 3A).
tion of membrane potential. The average RMP recorded Cells that did not respond to the test drugs were from fresh PASMCs was 244 6 3 mV (n 5 21). Cells exposed to 1 mM 4-AP to ensure reactivity (16.2 6 could be consistently depolarized from RMP by applica- 4 mV; n 5 17).
tion of 4-AP (1 mM; 18.6 6 4-mV depolarization; n 5 3)but not by tetraethylammonium (5 mM; data not shown), suggesting that a KV channel controls RMP inthese cells as previously described (4, 5, 33). Despite Obesity, which is estimated to contribute to 300,000 the significant inhibition of I deaths annually, is a significant medical problem in the K at 150 mV by 100 µM fluoxetine, when this concentration was applied to cells United States (19). Aminorex, fenfluramine, and dexfen- held at their RMP, it had minimal effect on the mem- fluramine were developed to treat obesity but have brane potential (3-mV depolarization; n 5 7; Fig. 3A).
been associated with epidemics of PPH (1, 7) and, more In contrast, if cells were held at a more depolarized recently, carcinoid-like cardiac valve disease (8). These potential of 230 mV, 100 µM fluoxetine caused signifi- drugs are inhibitors of potassium-channel activity in cant membrane depolarization (,18 mV; n 5 3; Fig.
resistance PASMCs (31). In a susceptible population, ELECTROPHYSIOLOGY AND PULMONARY VASOCONSTRICTION Fig. 2. A, left: actual representative traces of wholecell potassium currents recorded from a single pulmo-nary arterial smooth muscle cell (PASMC). Currents were evoked from a holding potential of 270 mV to atest potential of 150 mV during control period and1-min application of 2 mM 4-aminopyridine (4-AP).
A, right: current (I)-voltage (V) relationship of effectof 4-AP shown on left. B: dose-dependent inhibitionof potassium current (IK) recorded at 150 mV byincreasing doses of fluoxetine, Dex, and venlafaxineand inhibition by 100 µM phentermine. [drug], Drugconcentration. Values are means 6 SE; n 5 4–7 between doses.
this channel inhibition might result in membrane antagonist ketanserin (Fig. 1A). In 1996, there were depolarization, increased levels of intracellular cal- reported to be 18 million prescriptions for the anorexic cium, and pulmonary vasoconstriction, hence contribut- combination Fen-Phen (8). For this reason, we investi- ing to pulmonary hypertension. Since the withdrawal gated the effects of the combination of dexfenfluramine of fenfluramine and dexfenfluramine, a new generation and phentermine. In isolated lungs, in the presence of of antiobesity drug regimens has already emerged, phentermine, dexfenfluramine caused significantly including fluoxetine (Prozac) in combination with phen- greater pulmonary vasoconstriction than in lungs termine (Pro-Phen) (2). Fluoxetine and other drugs like treated with vehicle only (Fig. 1B). It is possible that it venlafaxine act, at least in part, through modulation of might similarly enhance the slight vasoconstriction the serotonergic system, leading to increased serotonin caused by lower concentrations of the serotonin reup- levels in the brain (28). In light of our previous data take inhibitors.
(31), we tested the effects of fluoxetine, phentermine, The patch-clamp studies show that all the drugs and venlafaxine on PA pressure in isolated rat lungs tested cause a dose-dependent inhibition of whole cell and on IK and membrane potential in single PASMCs.
IK in resistance PASMCs. Interestingly, fluoxetine All three drugs caused a slight increase in PA pressure causes the most potent inhibition, with nearly 60% of at a dose of 10 µM, but none constricted the lungs to the the total current at 150 mV blocked by 10 µM com- same extent as dexfenfluramine at the same concentra- pared with only 10% by dexfenfluramine. This would tion (Fig. 1A). At the high dose of 100 µM, venlafaxine appear to contradict the results in the whole lung, and phentermine caused a slight, additional increase in which indicate that fluoxetine has little effect on pulmo- pressure, whereas fluoxetine constricted the lungs as nary pressure at 10 µM. However, this may be ex- effectively as dexfenfluramine. The vasoconstriction at plained by considering the membrane potential data.
high concentrations of fluoxetine and dexfenfluramine At RMP, fluoxetine causes virtually no inhibition of IK, appeared to be via a mechanism independent of seroto- even at 100 µM, and, consequently, does not initiate nin because it could not be prevented by the 5-HT2 membrane depolarization. However, if the cell is held at

ELECTROPHYSIOLOGY AND PULMONARY VASOCONSTRICTION through several subtypes of the KV channel (6, 25, 29),and it is possible that anorexic subjects inhibit differentsubtypes. Dexfenfluramine has recently been shown toinhibit the Kv2.1 channel (25), which may contribute toRMP (6). Because fluoxetine, venlafaxine, and phenter-mine do not appear to inhibit the subtypes that setRMP, this may account for their lack of effect onpulmonary pressure at lower concentrations. As dis-cussed above, fluoxetine causes a large depolarizationof membrane potential if the cell is predepolarized to 230 mV. This further suggests that its inhibitoryeffects may be primarily on channels that open at morepositive membrane potentials. Alternatively, the inter-action of fluoxetine with KV channels to cause IKinhibition may itself be voltage dependent.
Although the mortality and morbidity rates associ- ated with PPH are high (9), the annual incidence of thedisease is low (1). This suggests that there is a geneticpredisposition to its development. The nature of thispredisposition is unknown. It may involve alteredexpression of ion channels, decreased production ofendogenous vasodilators, or increased production of endogenous vasoconstrictors. We have shown that inhi-bition of endogenous NO with L-NAME dramaticallyincreases the pulmonary vasoconstrictor responses todexfenfluramine, with constrictions seen at concentra-tions as low as 0.1 µM (31). This raises the possibilitythat low NO production may increase patient suscepti-bility to PPH. Indeed, patients with anorexigen-induced PPH appear to have an NO deficiency yearsafter discontinuing anorexigen treatment (3). Alterna- Fig. 3. Membrane potential (Em) data recorded in single PASMCs at tively, a difference in potassium-channel expression resting membrane potential (A) or at a potential of 230 mV (B). C, may increase susceptibility similar to the ATP-depen- control data for each group measured before 2-min application of dent potassium-channel dysfunction found in hyperin- either Dex, fluoxetine (Fluox), venlafaxine (Ven), or phentermine(Phen). Values are means sulinemic hypoglycemia of infancy (10, 18). Indeed, 6 SE; nos. in parentheses, no. of cells. *P , 0.05 compared with control value.
smooth muscle from PAs of PPH patients (unrelated toanorexic agents) has been shown to have decreased IK 230 mV, 100 µM fluoxetine causes a significant further values and depolarized membrane potentials compareddepolarization. The pathophysiological significance of with control subjects (34).
this observation is that if the membrane potential is Fluoxetine has previously been shown to inhibit IK in already partially depolarized, fluoxetine might then human and canine jejunal smooth muscle through a cause vasoconstriction. By comparison, dexfenflur- protein kinase C-dependent mechanism (11). In jejunal amine at a concentration of 100 µM is able to elicit a smooth muscle, one determinant of the role of a diffus- depolarization from RMP. This concentration applied ible second messenger in the inhibitory effect of fluox- acutely to the PASMCs is considerably higher than the etine on IK was that it could only be demonstrated with plasma level measured in patients treated chronically the perforated-patch clamp configuration where the (,1 µM). It should be remembered, however, that the cytosol of the cell remains intact. With the use of the cells in this study are from rats that have not been same rationale, the effects of fluoxetine observed in selected for any genetic susceptibility to PPH and also PASMCs may be independent of a cytosolic second that it is possible that dexfenfluramine may be concen- messenger because inhibition of IK was observed with trated in the cell over time.
the conventional whole cell patch-clamp configuration.
4-AP has been shown to inhibit most of the IK and to Because the inhibition of IK by dexfenfluramine was the depolarize PASMCs from their RMP (see RESULTS), same with the whole cell and perforated-patch tech- suggesting that 1) the outward current in resistance niques, we cannot confirm the necessity for a cytosolic PASMCs is predominantly due to activation of KV second messenger, at least in the SMC.
channels and 2) that KV channels are open and, at least Dexfenfluramine induces the release of serotonin in part, control RMP. Indeed, the data presented here from neurons and inhibits its reuptake (21), whereas show that 4-AP causes a significant membrane depolar- fluoxetine and venlafaxine only inhibit reuptake (28).
ization after administration of noneffective doses of Because serotonin itself is known to cause pulmonary fluoxetine, venlafaxine, or phentermine. The whole cell vasoconstriction (20, 23) and inhibition of potassium IK in PASMCs is likely to be due to current flowing channels (16, 24), the effects reported here could be ELECTROPHYSIOLOGY AND PULMONARY VASOCONSTRICTION attributed to increased levels of serotonin caused by the Hampl. Molecular identification of the role of voltage-gated K1
drugs. This is unlikely to explain the electrophysiologi- channels, Kv1.5 and 2.1, in hypoxic pulmonary vasoconstriction cal changes because the patch-clamp studies were done and control of resting membrane potentials in rat arterialmyocytes. J. Clin. Invest. 10: 1–12, 1998.
in single PASMCs that have no serotonergic innerva- 7. Brenot, F., P. Herve, P. Petitpretz, F. Parent, P. Duroux,
tion, and smooth muscle is not a recognized site of and G. Simonneau. Primary pulmonary hypertension and
serotonin production (12). Furthermore, inhibition of fenfluramine use. Br. Heart J. 70: 537–541, 1993.
8. Connolly, H., J. Crary, M. McGoon, D. Hensrud, B. Ed-
2 receptors by a concentration of ketanserin suffi- cient to prevent serotonin-induced pulmonary vasocon- wards, W. Edwards, and H. Schaff. Valvular heart disease is
associated with fenfluramine-phentermine. N. Engl. J. Med. 337:
striction had no effect on dexfenfluramine- or fluoxetine- 581–588, 1997.
induced vasoconstriction (although it is acknowledged 9. D'Alonzo, G. E., R. J. Barst, and S. M. Ayres. Survival in
that this conclusion relies on a complete and specific patients with primary pulmonary hypertension: results from a block of all 5-HT national prospective registry. Ann. Intern. Med. 115: 343–349, 2 receptors by ketanserin).
In the case of dexfenfluramine, inhibition of I membrane potential depolarization, and increases in 10. Dunne, M., C. Kane, R. Shepherd, and J. Sanchez. Familial
persistent hyperinsulinemic hypoglycemia of infancy and muta- intracellular calcium may play a significant role in the tions in the sulfonylurea receptor. N. Engl. J. Med. 336: 703–706, mechanism of pulmonary hypertension. It is possible that dexfenfluramine may also release sarcoplasmic 11. Farrugia, G. Modulation of ionic currents in isolated canine and
reticulum calcium directly. Fluoxetine is the most human jejunal circular smooth muscle cells by fluoxetine. Gastro- widely prescribed antidepressant in the world (32) and, enterology 110: 1438–1445, 1996.
12. Frishman, W. H., S. Huberfield, S. Okin, Y. Wang, A. Kumar,
despite its potent inhibition of IK at positive potentials, and B. Shareef. Serotonin and serotonin antagonism in cardio-
has not been associated with pulmonary hypertension.
vascular and non-cardiovascular disease. J. Clin. Pharmacol. 35: This is consistent with its lack of effect on RMP in 541–572, 1995.
PASMCs. Interestingly, at more depolarized membrane 13. Gurtner, H. Atiologie and haufigkeit der primer vaskularen
potentials, high concentrations of fluoxetine caused formen des chronischen cor pulmonale. Dtsch. Med. Wochenschr.
further depolarization, perhaps indicating a potential 94: 850–852, 1969.
14. Hamill, O. P., A. Marty, E. Neher, B. Sakmann, and F. J.
to enhance pulmonary vasoconstriction at supraphar- Sigworth. Improved patch-clamp techniques for high resolution
macological doses in susceptible patients. By contrast, recording from cells and cell-free membrane patches. Pflu venlafaxine and phentermine had minimal effects on Arch. 391: 85–89, 1981.
membrane potential or pulmonary vasoconstriction 15. Herve, P., J.-M. Launay, M.-L. Scrobohaci, F. Brenot, G.
even at high concentrations, suggesting that they may Simonneau, P. Petitpretz, P. Poubeau, J. Cerrina, P. Dur-
oux, and L. Drouet. Increased plasma serotonin in primary
have a lower risk of initiating pulmonary hypertension pulmonary hypertension. Am. J. Med. 99: 249–254, 1995.
according to this model.
16. Hevers, W., and R. C. Hardie. Serotonin modulates the voltage-
dependence of delayed rectifier and Shaker potassium channels We thank Knoll Pharmaceutical (Mt. Olive, NJ) for its interest and in Drosophila photoreceptors. Neuron 14: 845–856, 1995.
support of this project.
17. Holliday, S. M., and P. Benfield. Venlafaxine. A review of its
H. L. Reeve was supported by National Heart, Lung, and Blood pharmacology and therapeutic potential in depression. Drugs 49: Institute Grant R29-HL-59182 and was the 1997 recipient of the 280–294, 1995.
Giles F. Filley Award. E. K. Weir and S. L. Archer were supported by 18. Kane, C., R. Shepherd, P. Squires, P. Johnson, R. James, P.
Department of Veterans Affairs Merit Review Funding.
Milla, A. Aynsley-Green, K. Lindley, and M. Dunne. Loss of
Address for reprint requests: H. L. Reeve, Research 151, VA functional KATP channels in b cells causes persistent hyperinsu- Medical Center, Minneapolis, MN 55417.
linaemic hypoglycemia of infancy. Nat. Med. 2: 1344–1347, 1996.
Received 29 May 1998; accepted in final form 16 October 1998.
19. McGinnis, J., and W. Foege. Actual causes of death in the
United States. JAMA 270: 2207–2212, 1994.
20. McGoon, M., and P. Vanhoutte. Aggregating platelets contract
isolated canine pulmonary arteries by releasing 5-hydroxytrypta- 1. Abenhaim, L., Y. Moride, F. Brenot, S. Rich, J. Benichou, X.
mine. J. Clin. Invest. 74: 823–833, 1984.
Kurz, T. Higgenbottom, C. Oakley, E. Wouters, M. Aubier,
21. McTavish, D., and R. Heel. Dexfenfluramine. A review of its
G. Simonneau, and B. Begaud. Appetite-suppressant drugs
pharmacological properties and therapeutic potential in obesity.
and the risk of primary pulmonary hypertension. N. Engl. J.
Med. 335: 609–616, 1996.
Drugs 43: 713–733, 1992.
2. Anchors, M. Safer Than Fen-Phen. Rocklin, CA: Prima Publish-
22. Mlczoch, J. Drug and dietary induced pulmonary hypertension.
In: Pulmonary Hypertension, edited by E. K. Weir and J. T.
3. Archer, S. L., K. Djaballah, M. Humbert, E. K. Weir, M.
Reeves. New York: Futura, 1984, p. 341–359.
Fartoukh, J. Dall'Ava-Santucci, J.-C. Mercier, G. Simmon-
23. Neely, C., D. Haile, and I. Matot. Tone-dependent responses of
neau, and A. T. Dinh-Xuan. Nitric oxide deficiency in pulmo-
5-hydroxytryptamine in the feline pulmonary vascular bed are nary hypertension associated with the anorectic agents fenflur- mediated by two different 5-hydroxytryptamine receptors. J. amine and dexfenfluramine. Am. J. Respir. Crit. Care Med. 158: Pharmacol. Exp. Ther. 264: 1315–1326, 1993.
1061–1067, 1998.
24. Parker, I., M. M. Panicket, and R. Miledi. Serotonin receptors
4. Archer, S. L., J. Huang, T. Henry, D. Peterson, and E. K.
expressed in Xenopus oocytes by mRNA from brain mediate a Weir. A redox-based O
closing of K1 membrane channels. Brain Res. 7: 31–38, 1990.
2 sensor in rat pulmonary vasculature.
Circ. Res. 73: 1100–1112, 1993.
25. Patel, A., M. Lazdunski, and E. Honore. Kv2.1/Kv9.3, a novel
5. Archer, S. L., J. M. C. Huang, H. L. Reeve, V. Hampl, S.
ATP-dependent delayed rectifier K1 channel in oxygen-sensitive Tolarova, E. Michelakis, and E. K. Weir. The differential
pulmonary artery myocytes. EMBO J. 16: 6615–6625, 1997.
distribution of electrophysiologically distinct myocytes in conduit 26. Rae, J., K. Cooper, G. Gates, and M. Watsky. Low access
and resistance arteries determines their response to nitric oxide resistance perforated patch recordings using amphotericin B. J. and hypoxia. Circ. Res. 78: 431–442, 1996.
Neurosci. Methods 37: 15–26, 1991.
6. Archer, S. L., E. Souil, A. T. Dinh-Xuan, B. Schremmer, J.-C.
27. Silverstone, T. Appetite suppressants. Drugs 43: 820–836,
Mercier, A. El Yaagoubi, L. Nguyen-Hu, H. L. Reeve, and V.
ELECTROPHYSIOLOGY AND PULMONARY VASOCONSTRICTION 28. Stark, P., R. Fuller, and D. Wong. The pharmacological profile
vascular smooth muscle and cause pulmonary vasoconstriction.
of fluoxetine. J. Clin. Psychiatry 46: 7–13, 1985.
Circulation 94: 2216–2220, 1996.
29. Wang, J., M. Juhaszova, L. J. Rubin, and X.-J. Yuan.
32. Wong, D., F. Bymaster, and E. Engleman. Prozac (fluoxetine,
Hypoxia inhibits expression of voltage-gated K1 channel a Lilly 110140), the first selective serotonin uptake inhibitor and subunits in pulmonary artery smooth muscle cells. J. Clin. an antidepressant drug: twenty years since its first publication.
Invest. 9: 2347–2353, 1997.
Life Sci. 57: 411–441, 1995.
30. Weir, E. K., and S. L. Archer. The mechanism of acute hypoxic
33. Yuan, X.-Y. Voltage-gated K1 currents regulate resting mem-
pulmonary vasoconstriction: the tale of two channels. FASEB J.
brane potential and [Ca21]i in pulmonary arterial myocytes. Circ. 9: 183–189, 1995.
Res. 77: 370–378, 1995.
31. Weir, E. K., H. L. Reeve, J. Huang, E. Michelakis, D. Nelson,
34. Yuan, X.-J., J. Wang, M. Juhasova, S. P. Gaine, and L. J.
V. Hampl, and S. L. Archer. Anorexic agents aminorex, fenflur-
Rubin. Attenuated K1 channel gene transcription in primary
amine and dexfenfluramine inhibit potassium current in rat pulmonary hypertension. Lancet 351: 726–727, 1998.

Source: http://ajplung.physiology.org/content/ajplung/276/2/L213.full.pdf