5-HT1A Receptor Mediated Signal Transduction: Effect of Antidepressants
5-HT1A receptor density was measured in brain regions of rats treated with imipramine (5 mg/kg body wt), desipramine (10 mg/kg body wt), clomipramine (10 mg/kg body wt) and trazodone (10 mg/kg body wt) for 40 days, using [3H]8-OH-DPAT. It was observed that chronic exposure to antidepressants (ADs) results in significant downregulation of 5-HT1A receptors (30-80%) in cortex with all the drugs. Interestingly, in hippocampus, imipramine treatment increased the 5-HT1A receptor density (48%). With trazodone, however, there was a significant decrease in the density of 5-HT1A receptors in hippocampus (79%). The affinity of [3H]8-OH-DPAT was increased with imipramine treatment both in cortex and hippocampus. With trazodone treatment the affinity of [3H]8-OH-DPAT to 5-HT1A receptors was significantly decreased only in cortex. There was no change in the basal adenylyl cyclase (AC) activity after exposure to any of the drugs. However, the 5-HT sensitive AC activity was significantly increased in cortex with imipramine (72%), clomipramine (20%) and trazodone (50%) treatment, whereas in hippocampus only imipramine (50%) and trazodone (60%) treatment significantly increased AC activity. In conclusion, chronic treatment with ADs results in downregulation of cortical 5-HT1A receptors along with concomitant increase in 5-HT stimulated AC activity suggesting the involvement of cortical 5-HT1A receptor-mediated AC responses in the mechanism of action of ADs.
Advances made in the field of Serotonin (5-HT) research during the past decade have been a direct result of more than 50 years of 5-HT research. The 5-HT neurons are located in the raphe nuclei of the brain stem but the 5-HT nerve terminals can be found throughout the brain, suggesting the involvement of 5-HT in many brain functions. The effects of 5-HT on a given neuron are dependent on the different types of receptors expressed and the location of that neuron. It is also well established that different receptor subtypes can exhibit different affinities for endogenous 5-HT and recently, neurons expressing different receptor mRNAs have been identified using various molecular biology techniques.
Considerable evidence is now available to support the pivotal role of serotonin system in exerting the antidepressant response in humans. Deficits in the function of 5-HT receptors have long been implicated in many psychiatric disorders, such as depression (Meltzer and Lowy, 1993), suicide (Cheetham et al., 1990), obsessive-compulsive disorders (OCD) (Murphy et al., 1989) and migraine (Murphy, 1990). Different types of antidepressant treatments have been shown to enhance 5-HT neurotransmission via different pre- or post-synaptic mechanisms. Chronic treatment with antidepressants (ADs), such as selective serotonin re-uptake inhibitors (SSRIs), has been shown to alter the functioning of 5-HT neurons in brain. The hypothesis that an important function of ADs is to restore a disturbed balance between different 5-HT receptor subtypes in depressed patients has been proposed. (Berendsen, 1995). The alteration in 5-HT1A receptor mediated cAMP pathway has also been shown to be involved in neuropsychiatric disorders such as affective disorders and schizophrenia (Lieberman et al., 1998).
Numerous observations support the notion that serotonin and its multiple receptor subtypes are linked not only to the biological basis of affective disorders, but also to the mechanism of action of ADs (Blier and de Montigny, 1994). At least seven types of 5-HT receptors have been identified in the central nervous system using pharmacological and molecular cloning techniques. Among 5-HT receptors, 5-HT1 and 5-HT2 receptor subtypes have received considerable attention owing to their involvement in the affective disorders (Meltzer and Lowy, 1993). 5-HT1 receptors, which constitute a major class of 5-HT receptors, are negatively coupled to adenylyl cyclase (AC) via Gi-protein. 5-HT1 receptors have been further subdivided into 5 subtypes, namely, 5-HT1A, 5-HT1B, 5-HT1D, 5-HT1E and 5-HT1F sites (Peroutka, 1986; Humphrey et al., 1993). It has also been shown that 5-HT1A receptors have two affinity sites, designated as high affinity and low affinity sites (Nenonene et al., 1994). High affinity (classical 5-HT1A) sites, but not low affinity sites, are coupled to AC via Gi-protein. Apart from 5-HT2 receptors, 5-HT1 receptor subtypes, especially 5-HT1A receptors, have been shown to be involved in the mechanism of action of various ADs (Hamon et al., 1988).
Physiological studies have implicated 5-HT1A receptors in therapeutic mechanism of action of ADs (Chaput et al., 1991). Over the last decade, growing evidence has suggested that 5-HT1A receptor ligands represent a new class of mixed anxiolytic antidepressant drugs. Buspirone, ipsapirone and gepirone, 5-HT1A receptor partial agonists, have been reported to have antianxiety effects both in pre-clinical and clinical studies (Traber and Glaser, 1987; Eison and Eison, 1994). Evidence also indicates a significant role for 5-HT1A receptors in the mechanism of action of antidepressant (Newman et al., 1993; Stahl, 1994; Le Poul et al., 2000) and antimanic drugs (Subhash et al., 1999).
Attempts to delineate the molecular mechanisms of action of ADs on the basis of acute studies have limited value and increasing attention is being turned on to resultant adaptive changes stemming from their chronic treatment. In this, post-synaptic changes are more pertinent to the mechanism of action of ADs. The time course for the occurrence of these adaptive changes in the brain of laboratory animals is consistent with the delayed onset of the antidepressant response in humans. The elucidation of these mechanisms of action has lead to the development of new pharmacological strategies to potentiate the therapeutic effect of the currently available drugs and the identification of novel targets to accelerate and further improve the treatment response in depression (Blier and de Montigny, 1998). The best evidence that 5-HT neurons may be involved in mediating antidepressant effects is the antidepressant activity of selective 5-HT uptake blockers (Cowen, 1990).
Although SSRIs have been recently found to be more effective with lesser side effects, tricyclic antidepressants (TCAs) are still considered for treatment in many cases (Delgado et al., 1992). The TCAs, imipramine (IMI), desipramine (DMI) and clomipramine (CMI) are chemically closely related but have distinct pharmacokinetic and pharmacodynamic properties. All three drugs are known to be non-specific monoamine inhibitors that inhibit both norepinephrine (NE) and 5-HT uptake with different potencies. Both IMI and DMI have been demonstrated to be effective in treating long-term depression (Wroblewski et al., 199 Corrections required for AD ppr.docx 6). CMI, which increased a long lasting cortical suppression of firing elicited by dorsal raphe nucleus, seems to act on 5-HT1A receptor subtype (Contreras et al., 1993). Murphy et al. (1989), in their clinical study, have shown that CMI is choice of drug in treating OCD.
Trazodone (TZD), a chlorophenylpiperazine derivative, is an atypical AD drug that is commonly referred to as SSRI (Asberg et al., 1987). The most discussed mechanism of action for antidepressant effects of TZD is selective blockade of 5-HT uptake. TZD inhibits the uptake of 5-HT with high efficacy when compared to its efficacy towards NE or DA re-uptake inhibition (Richelson, 1988). Among 5-HT receptors, TZD has been shown to bind to 5-HT2 receptors with higher affinity than 5-HT1 receptors (Peroutka and Snyder, 1980). Apart from its weak, but selective inhibition of 5-HT uptake, other most likely mechanism of action of TZD is the formation of metabolite, meta-chlorophenylpiperazine (mCPP), a 5-HT1 receptor agonist (Caccia et al., 1982). Chronic administration of TZD has been shown to affect 5-HT1A mediated inhibition of forskolin stimulated AC activity, suggesting the direct involvement of 5-HT1A receptors in the mechanism of action of TZD (Tsuchiyama et al., 1991; Newman et al., 1993).
A general hypothesis of 5-HT receptor dysfunction in depression suggests that 5-HT1A receptors may be downregulated, where as 5-HT2 receptor may be upregulated and display inadequate ability to convert receptor occupancy by 5-HT into adequate physiological response. Pre-clinical studies have shown that numerous ADs downregulate both 5-HT1A and 5-HT2 receptors (Stahl, 1994). Decreased 5-HT1A receptors have been shown autoradiographically in various brain regions of rats treated chronically with IMI and other drugs (Mizuta and Segawa, 1989). Similarly significantly decreased 5-HT1A receptors with concomitant increase in 5-HT stimulated AC activity has been shown after fluoxetine treatment, suggesting that fluoxetine, which has high affinity for these receptors, acts by modulating 5-HT1A receptor mediated response in brain (Subhash et al., 2000). However, there are contradicting reports suggesting no alteration in 5-HT1A receptors and also in second messenger system following chronic treatment with ADs (Odagaki et al., 1991; File et al., 1999).
Although the signal transduction pathway of 5-HT receptor subtypes is well known, the agonistic or antagonistic activity of variety of chemical agents at these receptors is not clearly understood because of lack of selective ligands to study these receptors. The availability of 8-hydroxy-2 [di-n-propylamino] tetralin (8-OH-DPAT), 5-HT1A selective agonist, has helped in characterizing the 5-HT1A receptors in various regions of brain. Radioligand binding is a technique that measures the binding of a labeled agonist or antagonist to its receptor, either in tissues or in purified form. However, membrane preparations are widely used. This technique allows number of receptors to be quantified with the determination of the affinity of ligand/ drug to the receptor, which was not possible earlier. It also helps in studying the dissociation and association kinetics and integration between ligand/drugs, effector systems and G-proteins (Challis, 1997). In our lab this radioligand binding technique is used to study these parameters of 5-HT1A receptors in regions of rat brain using highly specific radioligand [3H]8-OH-DPAT, which has very high affinity to this subtype. The detailed procedure employed and the results observed in cortex and hippocampus of rats chronically (40 days) treated with IMI, DMI, CMI and TZD on 5-HT1A receptors and linked second messenger system (cAMP) is presented. The study of long-term AD treatment provides an insight into the mechanism of actions of these drugs in modulating the serotonin signal transduction and will also help in understanding the diverse actions of ADs in different mood disorders.
Materials and Methods
[3H]8 hydroxy n dipropylaminotetralin ([3H]8- OH-DPAT, s.a. 221 Ci/mmol) was obtained from Amersham Int. (UK). Imipramine, desipramine, trazodone, 5 HT, theophylline, ATP and pargyline were obtained from Sigma Chemicals (USA). GF/B filters were obtained from Millipore (UK). Clomipramine was a gift from M/S Sun Pharmaceuticals (INDIA).
Animals and administration of drugs
Adult male Sprague Dawley rats, weighing 200-240 Gms, were used for all experiments. Animals (20 for each drug), procured from Central Animal Research Facility (CARF), were housed in cages (4 rats per cage) and exposed to regular day/night period with food and water adlibitum.
Imipramine (5 mg/kg body wt), desipramine (10 mg/kg body wt) clomipramine (10 mg/kg body wt) and trazodone (10 mg/kg body wt), were injected intraperitoneally, once daily, in the morning for a period of 40 days. Animals were sacrificed by decapitation under ether anesthesia, 24 hours after the last injection. Control rats received 0.5 ml of normal saline for the same period and were sacrificed along with the experimental rats. Brains were removed and cerebral cortex and hippocampus were dissected out on ice-cold petri dish. Tissues obtained from three rats were pooled for receptor binding and AC assay.
Crude membrane preparation was obtained by homogenizing the brain tissue in 20 volumes of Tris-HCl buffer (50 mM, pH 7.4) containing 0.32 M sucrose, following the procedure described elsewhere (Subhash et al.,1998). For [3H]8-OH-DPAT binding the pellet was resuspended in 50 mM Tris HCl buffer (pH 7.4). Protein concentration was estimated by Lowry’s method (1951) and adjusted to 1 mg/ml using same buffer.
Receptor binding assay :5-HT1A receptors were labeled with [3H]8-OH-DPAT (0.06-1.0 nM), following essentially the procedure described in detail earlier (Subhash et al., 1998). Non-specific binding was defined by using 10 µM 5-HT. After the incubation of membranes (200 µg protein) at 370C, with 6-8 concentrations of the ligand, the reaction was stopped by the addition of 2 ml ice-cold Tris HCl buffer (pH 7.4). The reaction mixture was rapidly filtered through GF/B filters under vacuum. Filters were dried and transferred to scintillation vials containing 5 ml scintillation cocktail and allowed to equilibrate overnight. Radioactivity was counted using LKB beta counter with 53% efficiency.
Adenylyl cyclase assay
Tissues, obtained from cerebral cortex and hippocampus, were homogenized in 10 volumes of Tris HCl buffer (50 mM, pH 7.4) and centrifuged at 3800 rpm to remove the cell debris. The supernatant obtained was used for AC assay. Both basal and 5-HT (10 µM) stimulated AC activity was estimated using [3H]ATP, following essentially the procedure described by Malnoe et al. (1990) and employed earlier (Subhash et al., 1998). AC activity was expressed as the amount of [3H]cAMP (pmoles) formed per milligram of protein. Protein content was estimated by Lowry’s method.
The data from the binding experiments was analyzed using computerized program “LIGAND” (McPherson, 1983) to obtain the equilibrium dissociation constant (Kd), the density of receptors (Bmax) and the Hill co efficient. The Bmax values were expressed in fmol/mg protein and Kd in nM. Student’s “t” test was used for statistical analysis.
In vivo effect of ADs on 5 HT1A receptors
The density of cortical 5-HT1A receptors was decreased significantly with IMI (60%; p<0.0001), DMI (32%; p<0.0001), CMI (36%; p<0.0001) and TZD (70%; p<0.0001) treatment. The affinity of [3H]8-OH-DPAT to cortical 5 HT1A receptors, as reflected by the Kd values, was significantly increased (p<0.01) in IMI treated rats (Kd=0.26+0.02 nM) and was significantly decreased (p<0.001) in TZD treated rats (Kd=0.60+0.06 nM) as compared to control (Kd = 0.47+0.02) rats (Table 1).
The density of 5-HT1A receptors was decreased significantly in hippocampus (79%; p<0.0001) after chronic treatment with trazodone.
However, there was no significant change in 5HT1A receptor density in hippocampus of DMI and CMI treated rats. The receptor density was, however, increased significantly (48%,;p<0.001) in hippocampus of IMI treated rats with a significant decrease in Kd (p<0.01) values. The affinity of [3H]8-OH-DPAT to 5-HT1A receptors in hippocampus was not altered in rats treated with other drugs (Table 1).
Hill co-efficient values were near to unity in both the regions of experimental rats, suggesting that at this concentration, [3H]8-OH-DPAT binds to a single class of high affinity 5-HT1A receptors, as in control rats.
Adenylyl cyclase activity
Both basal and 5 HT (10 µM) stimulated AC activity was assayed in cerebral cortex and hippocampus of experimental and control rats. In control cortex the basal AC activity (7.2+0.9 pmol/mg protein) was stimulated by nearly 75% with 10 µM 5-HT. A similar increase in AC activity was observed in the hippocampus (Table 2).
After chronic AD treatment, the basal AC activity remained unaltered in both the regions. However, the 5-HT sensitive AC activity was significantly increased by 72% (p<0.0001) in cortex and by 50% (p<0.001) in hippocampus after IMI treatment. After CMI treatment, the 5-HT stimulated AC activity was significantly increased only in cortex by 20% (p<0.001). The 5-HTstimulated AC activity was significantly (p<0.0001) increased in cortex (50%) and hippocampus (60%) after TZD treatment. However, no significant alterations were observed in 5-HT sensitive AC after DMI treatment, in both regions (Table 2).
Effects of chronic treatment of IMI, DMI, CMI and TZD (ADs) on cortical and hippocampal 5-HT1A receptors and linked second messenger system, AC were studied using radioligand binding technique and enzymatic assay, respectively. The results demonstrate that chronic administration of these ADs has differential effect on 5-HT1A receptors in cortex and hippocampus. The ADs used are chemically closely related and are non-specific monoamine uptake blockers that block uptake of 5-HT and NE. IMI and CMI are equipotent in blocking 5-HT and NE uptake and DMI has been shown to be approximately 40 fold more potent in blocking NE than 5-HT uptake (Auerbach et al., 1995). Trazodone, an atypical antidepressant, inhibits the reuptake of 5-HT and this relative specificity of TZD in blocking the 5-HT uptake was the first suggested mechanism of action for its antidepressant effects (Richelson and Pfenning, 1984).
The major finding of the present study is the significant downregulation of cortical 5-HT1A receptors in rat brain after chronic treatment (40 days) with ADs. However, only TZD treatment showed significant decrease in hippocampal 5-HT1A receptor density. Interestingly, IMI treatment produced a significant increase in the density of 5-HT1A receptors in hippocampus. The affinity of the ligand ([3H]8-OH-DPAT) to both cortical and hippocampal 5-HT1A receptors was altered only in IMI treated rats. The affinity however, was decreased only in cortex of TZD treated rats, as seen by increased Kd values. These results are in agreement with earlier reports (Pandey et al., 1991; Lund et al., 1992; Wieland et al., 1993). However, Odagaki et al. (1991) have observed no significant changes in 5-HT1A receptors after chronic treatment with ADs and Papp et al. (1994) failed to show any correlation between 5-HT1A receptor density and unpredictable stress after chronic treatment with IMI.
The AC activity in cortex and hippocampus was affected in differential manner after AD treatment. The Basal AC activity was unaltered after treatment with ADs in both the regions. The 5-HT sensitive AC activity was significantly increased in cortex and hippocampus of IMI and TZD treated rats and only in cortex of CMI treated rats. Variable effects of AD treatment on AC activity have been reported. Newman et al. (1990) reported reduced degree of inhibition of forskolin stimulated AC activity by 5-HT after chronic treatment with DMI. However, Odagaki et al. (1991) have observed no significant changes in AC activity in rat hippocampus after chronic treatment with ADs. Similarly, Sapena et al. (1994) reported that DMI treatment did not alter the AC activity in rat cortical slices. In this study also we observed no alterations in hippocampal 5-HT sensitive AC activity after DMI and CMI treatment, which might be due to no change in the 5-HT1A receptor density in this region.
Basic and clinical research has indicated that the sensitivity of 5-HT1A receptor may be reduced in depression and that agonists acting at this receptor subtype may have antidepressant properties (Cheetham et al., 1993; Thielen and Frazer, 1995 ; deVry, 1995). Reduced hypothermic response in depressed patients has been demonstrated following administration of buspirone and ipsapirone indicating a sub sensitivity of 5-HT1A receptors (Lesch et al.,1990). These studies have implicated that 5-HT1A receptors play an important role in the pathophysiology of depression and in the mechanism of action of ADs.
In this study, the density of 5-HT1A receptors has decreased only in cortex after chronic treatment with ADs, where as in hippocampus the density remained unchanged except in IMI and TZD treated rats. Functionally also, only in IMI and TZD treated rats, the 5-HT stimulated AC activity was significantly altered indicating the sensitivity of hippocampal 5-HT1A receptors to IMI and TZD. This differential effect of ADs on 5-HT1A receptors might be due to the direct interaction of the ADs with 5-HT1A receptors or due to the complex mutual inhibitory effects of 5-HT receptor subtypes, which respond differentially to the individual drugs (Berendsen, 1995). The significant downregulation of nearly 75% of 5-HT1A receptors observed in both cortex and hippocampus with TZD treatment could be due to the agonistic action of the metabolite mCPP formed in vivo from trazodone. This agonistic effect is also suggested from the study of linked AC response. The basal AC activity did not alter in TZD treated rats. However, there was a significant increase in the 5-HT stimulated AC activity in both cortex and hippocampus after TZD treatment. This alteration in AC response could be suggestive of downregulation of 5-HT1A receptors.
Few studies have been undertaken to study the effect of TZD on 5-HT1A receptor-mediated response (Tsuchiyama et al., 1991; Newman et al., 1993). However, in majority of these studies the effect of chronic administration (30-40 days) has not been studied. Selective blockade of 5-HT2 receptors, relative to 5-HT1 receptors may be another potential mechanism of AD drug effects. Lakoski and Aghajanian (1985) found that the 5-HT2 selective antagonist, ketanserin, enhanced the inhibitory effects of 5-HT in forebrain areas. Subsequent work has suggested that stimulation of 5-HT2 receptors may functionally oppose the effects of 5-HT1A receptors in postsynaptic areas (Sheldon and Aghajanian, 1990). These electrophysiological findings raise the possibility that the clinical effects of ADs could be related to antagonist action at 5-HT2/2C receptors and agonist action at 5-HT1 receptors.
These observations support the hypothesis that the important function of ADs is to restore a disturbed balance between 5-HT1 and 5-HT2 receptors (Stahl, 1994). Another explanation for the selective desensitization of 5-HT1A receptors could be the differential coupling of 5-HT1A receptors to different G-proteins, leading to a decreased efficiency in triggering the second messenger system and therefore cause a desensitization of 5-HT1A receptors (Li et al., 1997). The same theory can also explain the increase in the density of 5-HT1A receptors in hippocampus after treatment with IMI. Evidence for this hypothesis comes from the study of Lason and Przewocki (1993) where, they have shown that long-term treatment with IMI decreased the GO mRNA without affecting GS mRNA levels. Thus suggesting that the effect of AD treatment can also be seen at intracellular signal transduction level and their gene expression.
The results of this study suggest that chronic treatment with ADs such as IMI, DMI, CMI and TZD significantly downregulates cortical 5-HT1A receptors and the effect on AC activity is region specific. This suggests the involvement of cortical 5-HT1A receptor-mediated actions in the mechanism of action of ADs. The possibility of significant downregulation of 5-HT1A receptors by TZD observed in this study might be due to the direct action of TZD on these receptors as an agonist, apart from its 5-HT uptake inhibition, which cannot be ignored. Thus TZD seems to produce its therapeutic effects via downregulation of 5-HT1A receptors as produced by other classes of ADs.
The effect of ADs on the second messenger system linked to 5-HT1 receptor subtypes may be understood well by studying the forskolin stimulated AC system in presence of GTP and 5-HT1A agonists, and attempts in this regard are underway.
1 Asberg C, Eriksson B, Marensson B, Traskman-1. Asberg C, Eriksson B, Marensson B, Traskman-Bendz L and Wagner A (1987) Jr Clini Psychiat 47: 23-35.
- Auerbach SB, Lundberg JF and Hjorth S (1995) Neuropharmacol 34 : 89 96.
- Berendsen HH (1995). Pharmacol Ther 66 : 17-37.
- Blier P and de Montigny C (1994) Trends Pharmacol Sci 15: 220-227.
- Blier P and de Montigny C (1998) Biol Psychiat 44 : 313-23
- Caccia S, Fong MH, Garrattini S and Zanini MG (1982) Jr Pharmacol 34: 605-606.
- Challis RAJ (1997) In: Methods in molecular biology (ed), 83, pp 1-24. Humana Press, New Jersey,
- Chaput Y, de Montigney C and Blier P (1991) Neuropsychopharmacol 5 : 219-229.
- Cheetham SC, Crompton MR, Katona CLE and Burton RW (1990) Psychopharmacol 102 : 544-548.
- Cheetham SC, Viggers JA, Slater NA, Heal DJ and Buckett WR (1993) Neuropharmacol 32 : 737 743.
- Contreras CM, Marvan ML and Munoz- Mendez A (1993) Neuropsychobiol 27 :199 203.
- Cowen PJ (1990) Pharmacol Ther 46 :43-51
- deVry J (1995) Psychpharmacol 121: 1-26.
- Delgado PL, Price LH, Heninger GR and Charney DS (1992) In: Handbook of affective disorders (Paytkel ES, ed.) 83, pp 1-24,Churchill Livingstone, New York.
- Eison AS and Eison MS (1994) Prog Neruopsychopharmacol Biol Psych 18 : 47-62.
- File SE, Ouagazzal AM, Gonzalez LE and Overstreet DH (1999) Pharmacol Biochem Behav 62 : 695-701.
- Hamon M, Fattaccini CH, Adrien J, Gallisot MC, Martin P and Gozlan H (1988) J Pharmacol Expt Therp 24 : 745-752.
- Humphrey PPA, Hartig P and Hoyer D (1993) Trend Pharmacol Scien 14 : 233-236.
- Lakoski J and Aghajanian GK (1985) Neuropharmacology 24: 265-273.
- Lason W and Przewocki R (1993) Pol J Pharmacol 45 : 219 226.
- Le Poul E, Boni C, Hanoun N, Laporte AM, Laaris N, Chauvean J, Hamon M and Lanfumey L (2000) Neuropharmacol 39 : 110-122.
- Lesch KP, Mayer S, Disselkamp-Tietz J, Hoh A, Schoellnhammer G and Schulte HM (1990) Life Sci 46: 1271-1277.
- Li Q, Battaglia G, Louis D and Van deKar LD (1997) Brain Res 769: 141-151.
- Lieberman JA, Mailman RB, Duncan G, Sikich L, Chakoc M, Nichols DE and Kraus JE (1998) Biol Psychiat 44 : 1099-1117.
- Lowry OH, Rosebrough NJ, Farr AL and Randall RJ (1951) Jr Biol Chem 193 : 265 273.
- Lund A, Mjellem-Joly N and Hole K (1992) Neuropharmacol 31 : 25-32.
- Malnoe A, Milon H and Reme C (1990) Jr Neurochem 55 : 1480-1484.
- McPherson GA (1983) Comput Prog Biomedic 17: 107-114.
- Meltzer HY and Lowy T (1993) In: Psychopharmacology: The third generation of progress. (Herbert HY ed) pp.513-525, Raven Press, New York.
- Mizuta T and Segawa T (1989) Jpn Jr Pharmacol 50 : 315-326.
- Murphy DL, Zohar J, Benkelfat C, Pato MT, Pigott TA and Insel TR (1989) Brit J Psychiat 155 :15 24.
- Murphy DL (1990) . Neuropsychopharmacol 3: 457 471.
- Nenonene EK, Radja F, Carli M, Grondin L and Reader TA (1994) J Neurochem 62 : 1822-1834.
- Newman ME, Drummer D and Lerer B (1990) J Pharmacol Expt Ther 252: 826 831.
- Newman ME, Lerer B and Shapira B (1993) Biol Psychiat 17: 1-19.
- Odagaki Y, Koyama T and Yamashita I (1991) J Neur Transm 86 : 85-96.
- Pandey SC, Isaac L, Davis JN and Pandey GN (1991) Eur J Pharmacol 202 : 221 225.
38 Papp M, Klimek V and Willner P (1994) Psychopharmacol 114: 309 314.
- Peroutka SJ (1986) Jr Neurochem 57: 529-540.
- Peroutka SJ and Snyder SH (1980) Science 210: 88-90.
- Richelson E. (1988) McLean Hosp Jour 13 : 67-88.
- Richelson E and Pfenning M (1984) Eur Jr Pharmacol 104: 277-286.
- Sapena R, Morin D, Zini R and Tillement JP (1994) Drug Expt Clin Res 20: 93-102.
- Sheldon PW and Aghajanian GK (1990) Brain Res 506: 62-69.
- Subhash MN, Srinivas BN, Vinod KY and Jagadeesh S (1998) Neurochemic Resear 23 : 1321-1326.
- Subhash MN, Vinod KY and Srinivas BN (1999) Neurochem Int 35 : 247-253.
- Subhash MN, Srinivas BN, Vinod KY and Jagadeesh S (2000) J Neural Transm 107 :377-87.
- Stahl S (1994) Psychopharmacol Bull 30 : 39-43.
- Thielen RJ and Frazer A (1995) Life Sci 56 :163-168.
- Traber J and Glaser T (1987) TrendPharmacol Sci 8 : 431-437.
- Tsuchiyama K, Nagayama H, Akiyoshi J, Yamada K, Miyamoto T, Yamauchi C, Oba A, Ito Y, Okada K and Kojima K (1991) Jap Jr Psychiat Neurol 45 : 139-140.
- Wieland S, Fischette CT and Lucki I (1993) Neruophar¬macol 32: 561 573.
- Wroblewski BA, Joseph AB and Cornblatt RR (1996) J Clin Psychiat 57 : 582 587.