Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Selective serotonin reuptake inhibitors (SSRIs) are the most widely prescribed class of antidepressant drugs because they are clinically effective, elicit fewer aversive side effects, and have an increased safety margin compared with other antidepressants. SSRIs given acutely block the serotonin transporter and increase extracellular levels of 5-HT in forebrain regions, including the cortex, hippocampus, and striatum (Kreiss and Lucki, 1995; Invernizzi et al., 1996; Romero et al., 1996; Hjorth et al., 2000). The ability of SSRIs to increase serotonergic neurotransmission is believed to be an important component of their clinical antidepressant activity (Heninger et al., 1996). As with other antidepressants, however, repeated administration and a substantial latency are required before therapeutic effects are typically observed, and the reasons for this are not clearly understood. It is thought that repeated administration of SSRIs may lead to a net augmentation of 5-HT transmission, at least in part, by desensitizing presynaptic inhibitory 5-HT autoreceptors, either somatodendritic 5-HT_1A receptors or 5-HT_1B and 5-HT_1D receptors located on nerve terminals (Chaput et al., 1986; Blier et al., 1987; Kreiss and Lucki, 1995; Invernizzi et al., 1996).

Based in part on this framework, recent in vivo microdialysis and electrophysiology experiments have shown that administration of 5-HT_1A receptor antagonists augment the effects of SSRIs on extracellular 5-HT (for review, see Artigas et al., 1996; Hjorth et al., 2000). This framework has also served as the basis for the development of clinical treatment strategies for either inducing a more rapid onset of clinical effects or for potentially managing treatment-resistant patients (for review, see McAskill et al., 1998; Cryan and Leonard, 2000; Nelson, 2000). For example, the combined administration of pindolol, a mixed 5-HT_1A/-adrenergic receptor partial agonist, with SSRIs has been shown to enhance the increase of extracellular levels of 5-HT in preclinical studies (Romero et al., 1996) and produce a more rapid onset of clinical antidepressant effects (Blier and Bergeron, 1997; Perez et al., 1997), although not all clinical trials with this drug combination have been positive (Berman et al., 1997).

Serotonergic projections throughout the forebrain arise primarily from two main cell body groups, the dorsal raphe (DR) and median raphe (MR) nuclei (Kosofsky and Molliver, 1987). Regions innervated by the DR (e.g., striatum and frontal cortex) have been shown to be more responsive to regulation by the 5-HT_1A autoreceptor than regions innervated by the MR (e.g., dorsal and ventral hippocampus) (Sinton and Fallon, 1988; Hjorth and Sharp, 1991; Malagie et al., 1996; Invernizzi et al., 1997; Romero and Artigas, 1997; Knobelman et al., 2001).

The present studies examined in vivo pharmacological responses evoked by the novel compound EMD 68843 (also known as SB659746-A or vilazodone) because of its dual activity at 5-HT transporters and 5-HT_1A receptors. Like SSRIs, EMD 68843 is a potent inhibitor of the reuptake of 5-HT with an IC₅₀ value of 0.2 nM (Bartoszyk et al., 1997), which is about 300-fold more selective at inhibiting 5-HT than norepinephrine reuptake (C. Seyfried, Merck, unpublished data). Unlike the prototypic SSRI fluoxetine, however, EMD 68843 exhibits high affinity for the 5-HT_1A receptor (IC₅₀ = 0.5 nM) (Bartoszyk et al., 1997) and negligible affinity for other 5-HT receptors (5-HT_1D, 5-HT_2A, and 5-HT_2C receptors; C. Seyfried, Merck, unpublished data). Although its efficacy for activating 5-HT_1A receptors was not previously determined, in vitro assays demonstrated a decrease in 5-HT release, indicating activation of presynaptic 5-HT_1A receptors (Merck, unpublished data). Initial behavioral studies showed that EMD 68843 (55 mg/kg, p.o.) inhibited ultrasonic vocalizations in rats, a behavioral model for anxiolytic effects produced by activation of presynaptic 5-HT_1A receptors. The absence of effects on core body temperature of rats was suggested to indicate that EMD 68843 lacked intrinsic activity at postsynaptic 5-HT_1A receptors (Bartoszyk et al., 1997). The goal of the present study was to examine the effects of EMD 68843 on extracellular 5-HT levels as a prototype compound expected to demonstrate mixed activity at inhibiting 5-HT reuptake and regulating 5-HT_1A receptors. The effects of EMD 68843 were compared with the SSRI fluoxetine on extracellular 5-HT levels in the ventral hippocampus (HPv) and the frontal cortex (FC) after acute, systemic administration using in vivo microdialysis. Additional studies were done to characterize the effects of EMD 68843 on responses mediated by 5-HT_1A receptors. For example, the effects of EMD 68843 were examined in the forced swimming test (FST), a behavioral test in rats and mice used to predict antidepressant potential (Borsini and Meli, 1988; Lucki, 1997). Also, the efficacy of EMD 68843 to activate postsynaptic 5-HT_1A receptors was evaluated in vitro by examining its ability to promote binding of [³⁵S]GTPS to G proteins expressed in Sf9 cells efficacy (Butkerait et al., 1995; Barr and Manning, 1997) and in vivo by observing the appearance of the 5-HT behavioral syndrome (Lucki, 1992).

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals. Male Sprague-Dawley rats (Charles River Laboratories, Wilmington, MA) weighing 175 to 200 g at the start of the experiment were housed two per cage and maintained under conditions of constant temperature (22°C) on a 12:12-h light/dark cycle (7:00 AM on and 7:00 PM off) with free access to food and water. Animal procedures were conducted in accordance with the guidelines published in the National Institutes of Health Guide for Care and Use of Laboratory Animals, and all protocols were approved by the University of Pennsylvania Institutional Animal Care and Use Committee.

Microdialysis Probe Construction. Vertical, concentric microdialysis probes were constructed as previously described (Kreiss and Lucki, 1997). A plastic screw cap secured the probe into the guide cannula on the day of the experiment. In vitro recovery rates for dialysis probes were determined by placing them in an Eppendorf tube filled with aCSF. The aCSF was continually perfused through the probe at a rate of 0.8 µl/min. After a known concentration of 5-HT standard was added to the bath, the concentration of 5-HT in the perfusate was compared with the amount in the bath. The reported values for 5-HT in dialysate samples are corrected for recovery for archival purposes. The mean recovery rate was 24 ± 1.2%.

Dialysate Analysis. The amount of 5-HT in dialysates was determined with high-pressure liquid chromatography with electrochemical detection. Briefly, 12-µl samples were injected directly onto high-pressure liquid chromatography, which consisted of a PM80 solvent delivery system (Bioanalytical Systems, West Lafayette, IN), a 10-µl sample loop, and a Sepstik microbore column (ODS 3 µm; 100 × 1 mm; Bioanalytical Systems). The mobile phase consisted of 32 mM NaH₂PO₄, 0.67 µM EDTA, 0.43 mM octyl sulfate, and 19% methanol adjusted to a pH of 4.3. The flow rate through the system was 110 µl/min, and the detector was set at a potential of +0.60 V relative to a silver/AgCl reference electrode. Standard concentrations of 5-HT were prepared prior to injection of tissue samples. Dialysate concentrations of 5-HT were determined using a linear regression analysis of the peak heights obtained from a range of standard concentrations.

Data Analysis. Microdialysis data are reported as a percentage of the control baseline values (determined from the mean of the final three samples prior to drug injection). The transformed value (percentage of baseline) was used as the dependent variable for assessment of within-group effects. Values for the mean and the nine samples collected immediately after drug injection were included in the analyses. Between-group analyses of dose differences or treatment differences (EMD versus fluoxetine) were conducted using two-way ANOVA with repeated measures over time. Fisher's post hoc test was used for comparison of group means relative to a control mean. The level of significance required for all post hoc analyses was p < 0.05. One-way ANOVA with repeated measures was used for comparison of drug values to a baseline mean. Two-way ANOVA was performed to determine the influence of 8-OH-DPAT on EMD- and fluoxetine-elicited increases in extracellular 5-HT. The new baseline value (100%) was defined by the mean of the two samples following EMD or fluoxetine injection. The values were determined from the first sample following 8-OH-DPAT administration to the end of sample collection.

Behavioral Testing. To determine whether EMD 68843 displayed antidepressant-like activity similar to that observed for fluoxetine and other SSRIs, we subjected rats to the FST, a test frequently used to reliably screen for effective antidepressants. Sprague-Dawley rats (Charles River Laboratories Inc., Wilmington, MA; 175-200 g at the start of the experiment) were housed in groups of two in polycarbonate cages and maintained on a 12:12-h light/dark cycle in a temperature-controlled (22°C) colony with free access to food and water. Animals were handled for 5 min daily, beginning 4 days prior to the onset of the experiment. Animals were randomly assigned to four groups (n = 7-8 rats/group) and were exposed to a pretest swim session for 15 min 24 h prior to the 5-min swim test. Rats were placed into a glass cylinder (46 cm high × 20 cm wide) filled to a 30-cm depth with water at room temperature for 15 min. Animals were removed from the water, dried by the experimenter, and placed into plastic cages with a heating pad. EMD 68843 (1.0, 3.0, and 10.0 mg/kg) or vehicle (6% DMSO) was administered s.c. three times following the initial 15-min pretest exposure to the swim chamber. Injections were given 23.5, 4.5, and 1 h prior to the swim test. The 5-min swim session was videotaped from above for subsequent analysis of behavioral responses by a trained observer (Detke et al., 1995a). The rater was blind with respect to treatment group. Swimming was defined as horizontal movement through the swim chamber, which included crossing into another quadrant. Climbing activity consisted of upward-directed movements of the forepaws along the side of the swim chamber, and immobility was assigned when no activity was observed other than that necessary to keep the rat's head above the water.

5-HT Behavioral Syndrome. Rats were observed for 15 min for the appearance of the 5-HT behavioral syndrome immediately after the administration of 8-OH-DPAT (1.0 mg/kg, i.p.) or EMD 68843 (3.0 mg/kg, i.p). Varying doses of EMD 68843 (0.3-3.0 mg/kg i.p.) were also administered 15 min prior to a challenge dose of 8-OH-DPAT (1.0 mg/kg, i.p.) to determine whether EMD 68843 could block the 5-HT syndrome. The rats were observed in individual polycarbonate cages with the floor covered with bedding material. The cage was similar to their home cage. Rats were rated for the occurrence of the following symptoms following drug injection: 1) flat body posture; 2) hindlimb abduction; 3) lateral head weaving; 4) resting tremor; 5) forepaw treading; and 6) Straub tail. Each symptom was rated on an intensity scale with 0 = absent, 1 = weak, 2 = moderate, and 3 = intense. An intensity score of 2 or greater was required for the symptom to be considered as prominently presented. Rats were scored as showing the 5-HT behavioral syndrome if at least four of the six symptoms were prominently presented during the observation period. These procedures have previously been used by this laboratory to rate the 5-HT behavioral syndrome produced by full agonists (Lucki, 1992; Singh and Lucki, 1993).

[³⁵S]GTPS Binding. Ligand-promoted binding of [³⁵S]GTPS to G proteins with the 5-HT_1A receptor expressed in Sf9 cells was used to compare the relative efficacy of compounds and was evaluated as previously described (Butkerait et al., 1995; Barr and Manning, 1997). 8-OH-PIPAT defines the effects of a full agonist, and vehicle (or DMSO) defines receptor constitutive activity. In the present study, [³⁵S]GTPS binding achieved with 8-OH-PIPAT was compared with that achieved with EMD 68843, fluoxetine, and vehicle.

Drugs. All drugs were prepared freshly on the day of use. EMD 68843 (Merck, Darmstadt, Germany) was dissolved in 6% DMSO and injected i.p. in a volume of 2 ml/kg in the rat studies. Fluoxetine hydrochloride (Lilly Research Laboratories, Indianapolis, IN) was dissolved in deionized water and injected i.p. in a volume of 2 ml/kg. R-8-OH-DPAT hydrobromide (Sigma/RBI, Natick, MA) was dissolved in deionized water. Fluoxetine and R-8-OH-DPAT doses were calculated as the base weight. EMD 68843 doses were calculated without correction. Drugs were administered in a volume of 10 ml/kg in the mouse studies.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Microdialysis Studies

Extracellular 5-HT Levels in the Ventral Hippocampus. Administration of fluoxetine (10 mg/kg) increased extracellular levels of 5-HT to a maximum of 192.2 ± 37.2% above basal values in the HPv. A higher dose of fluoxetine (20 mg/kg) increased extracellular 5-HT by 273.9 ± 77.1%. Two-way ANOVA revealed a significant increase in extracellular 5-HT in the HPv at both doses but no significant difference between dose [dose: F(1,13) = 0.99, N.S.; time: F(9,117) = 10.52, p < 0.0001; dose × time interaction: F(9,117) = 1.00, N.S.; see Fig. 1A]. In contrast, EMD 68843 dose dependently increased extracellular 5-HT in the HPv [dose: F(4,26) = 14.8, p < 0.0001; time: F(9,234) = 20.75, p < 0.0001; dose × time interaction: F(36,234) = 3.36, p < 0.0001; see Fig. 1B]. The largest increase was observed following the 3 mg/kg dose of EMD 68843 (558.3 ± 52.5% above baseline).

View larger version (16K): [in this window] [in a new window]   Fig. 1.   The effects of fluoxetine (panel A) and EMD 68843 (panel B) on extracellular 5-HT levels in the ventral hippocampus. Panel A shows that both the 10 mg/kg [squares (n = 7); baseline = 4.2 ± 0.8 fmol/10 µl] and 20 mg/kg [circles (n = 8); baseline = 6.9 ± 1.3] dose of FLX increased extracellular levels of 5-HT in the ventral hippocampus. However, there was no significant difference between the two doses. Panel B shows the effect of several doses of EMD 68843: 0.3 mg/kg [squares (n = 6); baseline = 7.1 ± 1.3 fmol/10 µl]; 1.0 mg/kg [diamonds (n = 7); baseline = 5.9 ± 1.1 fmol/10 µl]; 3.0 mg/kg [triangles (n = 6); baseline = 6.3 ± 1.4 fmol/10 µl]; and 10.0 mg/kg [circles (n = 6); baseline = 4.7 ± 0.7 fmol/10 µl]. Post hoc analysis showed a significant dose-related increase in extracellular 5-HT levels elicited by all doses of EMD 68843, except 0.3 mg/kg (Fisher's PLSD, p < 0.05).

Extracellular 5-HT Levels in the Frontal Cortex. Both doses of fluoxetine resulted in significantly elevated levels of extracellular 5-HT compared with saline injection as revealed by two-way ANOVA [dose: F(2,16) = 15.07, p < 0.0002; time: F(9,144) = 7.02, p < 0.0001; dose × time interaction: F(18,144) = 1.87, p = 0.001; see Fig. 2A]. The peak increase (95.7 ± 26.2%) observed at the 20 mg/kg dose was less than that seen with a 10 mg/kg dose, which elicited a peak increase of 164.9 ± 28.3% above baseline.

View larger version (17K): [in this window] [in a new window]   Fig. 2.   The effects of fluoxetine (panel A) and EMD 68843 (panel B) on extracellular 5-HT levels in the frontal cortex. Panel A illustrates the significant increase of extracellular 5-HT in the frontal cortex elicited by 10 mg/kg [squares (n = 7); baseline = 8.9 ± 1.6 fmol/10 µl] or 20 mg/kg [circles (n = 6); baseline = 11.4 ± 1.5 fmol/10 µl] FLX. The effects of administering saline [triangles (n = 6); baseline = 9.3 ± 1.5 fmol/10 µl] are also shown. Panel B shows the effect of EMD 68843: 0.3 mg/kg [squares (n = 6); baseline = 8.4 ± 1.7 fmol/10 µl]; 1.0 mg/kg [circles (n = 6); baseline = 4.1 ± 1.2 fmol/10 µl]; and 3.0 mg/kg [triangles; (n = 7); baseline = 5.3 ± 1.0 fmol/10 µl] on extracellular 5-HT levels in the frontal cortex. Post hoc analysis showed a significant difference between the effect of 1.0 and 3.0 mg/kg EMD 68843 when compared with saline.

8-OH-DPAT Combined with Fluoxetine or EMD 68843. The ability of the 5-HT_1A receptor agonist, 8-OH-DPAT to reduce extracellular levels of 5-HT following treatment with either fluoxetine or EMD 68843 was examined in the FC and the HPv. In the HPv, 8-OH-DPAT reduced extracellular 5-HT to the same extent in fluoxetine- and EMD-treated animals [drug: F(1,12) = 1.66, N.S.; time: F(6,72) = 7.55, p < 0.0001; drug × time interaction: F(6,72) = 0.89, p < N.S., two-way ANOVA; see Fig. 3A]. In contrast, 8-OH-DPAT reduced extracellular 5-HT in the FC of fluoxetine-treated animals to near basal levels, whereas only a brief reduction of 5-HT was noted in the EMD-treated animals [drug: F(1,12) = 2.90, N.S.; time: F(6,72) = 3.31, p = 0.006; drug × time interaction: F(6, 72) = 2.84, p = 0.016; see Fig. 3B].

View larger version (17K): [in this window] [in a new window]   Fig. 3.   The effect of 8-OH-DPAT administration on increased levels of 5-HT produced either by fluoxetine or EMD 68843. Panel A illustrates the effect of 8-OH-DPAT administration on extracellular levels of 5-HT in the ventral hippocampus elicited by 10 mg/kg FLX [circles (n = 6); baseline = 4.2 ± 0.4 fmol/10 µl] or 3 mg/kg EMD 68843 [squares (n = 7); baseline = 5.7 ± 1.4 fmol/10 µl]. FLX or EMD 68843 were administered at timepoint 0 (left arrow). A subsequent i.p. injection of 1 mg/kg 8-OH-DPAT was administered 40 min later (right arrow). Samples were collected for an additional 160 min. Panel B illustrates the effect of 8-OH-DPAT administration on extracellular levels of 5-HT in the FC elicited by 10 mg/kg FLX [circles (n = 7); baseline = 3.8 ± 0.4 fmol/10 µl] or 3 mg/kg EMD 68843 [squares (n = 7); baseline = 2.9 ± 0.6 fmol/10 µl]. The effect of 8-OH-DPAT measured in the FC was greater in animals treated with fluoxetine than with EMD 68843 (see Results, p = 0.016).

Behavioral Studies

Forced Swimming Test in Rats. EMD 68843 (1.0 mg/kg) significantly reduced the amount of time spent immobile [F(3,27) = 3.67, p < 0.03]. A concomitant increase was evident in the amount of time spent swimming [F(3,27) = 4.68, p < 0.01], but no significant change was noted in climbing [F(3,27) = 0.66, p > 0.05]. Follow-up tests confirmed that only the 1.0 mg/kg dose produced significant changes of immobility and swimming when compared with vehicle (see Fig. 4B).

View larger version (38K): [in this window] [in a new window]   Fig. 4.   Antidepressant-like activity of EMD 68843 examined in the rat and mouse FST. Panel A, the mouse FST measured the amount of time spent immobile (+1 S.E.) in the last 4 min of a 6-min swim session (n = 9-10 mice/group). EMD 68843 (1.0 mg/kg, i.p.) significantly decreased the amount of time spent immobile of DBA/2J mice when compared with vehicle (, p < 0.05, Fisher's PLSD test). Panel B, behavioral activity in the rat FST was characterized as immobility, swimming, or climbing during each 5-sec interval of the 5-min test session (n = 7-8 rats/group). The mean counts (+1 S.E.) are displayed for each behavioral category. EMD 68843 (1.0 mg/kg, i.p.) significantly reduced the frequency of immobility counts and increased swimming compared with vehicle (, p < 0.05, Fisher's PLSD test). No significant change in climbing was measured at any dose.

Forced Swimming Test in Mice. EMD 68843 displayed antidepressant-like activity in the mouse FST at the 1.0 mg/kg dose. A significant decrease in immobility was revealed by one-way ANOVA [F(3,32) = 3.36, p < 0.03]. No significant changes were observed with the 0.3 or 3.0 mg/kg dose (see Fig. 4A).

Evaluation of Efficacy at 5-HT_1A Receptors

Sf9 Cells. Ligand-facilitated binding of [³⁵S]GTPS to Gi proteins together with the 5-HT_1A receptor expressed in Sf9 cells was used to evaluate relative efficacy (Butkerait et al., 1995; Barr and Manning, 1997). As shown in Fig. 5, 5-HT ligands differentially increased [³⁵S]GTPS binding [F(4,10) = 249.9, p < 0.001]. The full agonist 8-OH-PIPAT increased [³⁵S]GTPS binding about 6-fold over that observed with vehicle (p < 0.001). EMD 68843 increased binding by about 4-fold, demonstrating activation of 5-HT_1A receptors (p < 0.001). However, the increase was approximately 30% less than that obtained with 8-OH-PIPAT (p < 0.001), consistent with the idea that EMD 68843 is a 5-HT_1A receptor partial agonist. The SSRI fluoxetine had no effect relative to vehicle (p > 0.05).

View larger version (38K): [in this window] [in a new window]   Fig. 5.   Promotion of the binding of [35S]GTPS to G proteins expressed in Sf9 cells by 5-HT ligands. Membranes from Sf9 cells expressing Gi2 (Gi2, G1, and G2) and 5-HT1A receptor were assayed for [35S]GTPS binding to Gi2 with 8-OH-PIPAT (600 nM), fluoxetine (100 nM), EMD 68843 (100 nM), or vehicle (striped bars). The separate vehicle (unfilled bars) for EMD 68843 consisted of 0.01% DMSO. Immunoprecipitation of Gi2 was performed with C terminus-directed antibodies. A control using nonimmune serum for untreated cells yielded negligible binding (75.0 ± 5.1 cpm). Data from a single experiment performed in triplicate are shown. The experiment shown is representative of four experiments yielding equivalent results. Bars represent mean GTP binding (cpm +1 S.E.). *, drug differed significantly from the corresponding control (H2O for PIPAT and fluoxetine, and DMSO for EMD), p < 0.001 using Fisher's PLSD test. #, values differed significantly from each other, p < 0.001 according to Fisher's PLSD test.

5-HT Syndrome in Rats. The administration of the 5-HT_1A receptor agonist 8-OH-DPAT produced all of the symptoms of the 5-HT behavioral syndrome. In contrast, EMD 68843 when tested alone at the highest dose used (3.0 mg/kg, i.p.) produced none of the symptoms. Pretreatment with EMD 68843 dose dependently blocked the ability of 8-OH-DPAT to evoke the 5-HT behavioral syndrome at doses of 1.0 and 3.0 mg/kg (see Table 1). Pretreatment with fluoxetine did not alter the 5-HT syndrome produced by 8-OH-DPAT.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

As a ligand that potently combines blockade of 5-HT reuptake with high affinity for 5-HT_1A receptors, the partial agonist EMD 68843 may be a prototype for a more effective type of SSRI. The additional regulation of 5-HT_1A autoreceptors by a 5-HT transporter inhibitor may allow a single molecule to evoke responses that are similar to those of drug combinations that could augment the clinical effects of conventional SSRIs in treating depression, such as the combination of fluoxetine with pindolol. This hypothesis was tested using in vivo microdialysis to show that systemic administration of EMD 68843 increased extracellular levels of 5-HT in the FC (527 versus 165%) and the HPv (558 versus 274%) of the rat to a greater extent than the conventional SSRI fluoxetine. Although the two compounds differ in potency at 5-HT transporters, the effects produced by systemic fluoxetine are representative maximal effects for SSRIs, as shown for other more potent SSRIs (reviewed by Fuller, 1994; Hjorth et al., 2000) or for fluoxetine tested at higher doses (up to 40 mg/kg, i.p.; Lucki, unpublished data). This finding provides an important proof of principle for the conceptual development of novel compounds with dual actions on the 5-HT transporter and 5-HT_1A receptors that may elicit a faster onset or more therapeutically effective antidepressant response.

The augmentation of extracellular 5-HT achieved by acute administration of EMD 68843 is likely due to the combined inhibition of the 5-HT transporter with partial agonist effects at the 5-HT_1A autoreceptor. Sf9 cells provide an intact cell setting for reconstitution of the human 5-HT_1A receptor with G protein subunits that can assess the relative efficacy of 5-HT_1A receptor agonists and antagonists under controlled conditions (Butkerait et al., 1995; Barr and Manning, 1997). As shown in the present study, EMD 68843 increased [³⁵S]GTPS binding to 69% of the magnitude of the full 5-HT_1A receptor agonist 8-OH-PIPAT in Sf9 cells, indicating that it is a partial agonist at 5-HT_1A receptors. In general, conventional SSRIs have been shown to increase extracellular levels 2- to 4-fold at maximum effective dose, whereas drugs that increase release, like fenfluramine, produce an increase in extracellular levels of 5-HT in the 8- to 12-fold range (Fuller, 1994). Although enhancement of 5-HT release cannot be ruled out, EMD 68843 does not produce a reduction of 5-HT tissue content as does fenfluramine (C. Seyfried, Merck, personal communication). Previous studies have shown that combining SSRIs with 5-HT_1A receptor antagonists, such as WAY 100635, augment extracellular levels of 5-HT in the cortex to a range above levels achieved by SSRIs given alone but less than fenfluramine (Artigas et al., 1996). The blockade of 5-HT_1A receptors limits the ability of endogenous 5-HT to interact with autoreceptors that decrease cell firing and synthesis of 5-HT and, consequently, enhances extracellular 5-HT levels in terminal regions (Artigas et al., 1996; Hjorth et al., 2000). Coadministration of 5-HT_1A receptor partial agonists, such as pindolol, WAY 100135, or tertatolol, also augments the effects of SSRIs on extracellular 5-HT levels (Fletcher et al., 1993; Romero et al., 1996; Gartside et al., 1999; Hjorth et al., 2000). However, the mechanism underlying this effect is not easily understood because 5-HT_1A receptor partial agonists produce complex effects. Partial agonists activate presynaptic 5-HT_1A receptors but block the effects of 5-HT_1A receptor agonists at postsynaptic receptors because of the different receptor reserve of 5-HT_1A receptors at these anatomical locations (Cox et al., 1993). Although 5-HT_1A receptor partial agonists inhibit 5-HT neuronal activity and reduce 5-HT synthesis, 5-HT_1A autoreceptor activation by a full agonist, such as the endogenous neurotransmitter, can be diminished by pretreatment with a partial agonist (Arborelius et al., 2000). Nevertheless, additional pharmacological mechanisms may contribute to augmenting extracellular 5-HT levels in terminal regions when 5-HT_1A receptor partial agonists are combined with SSRIs. One possible mechanism may be that partial agonists rapidly desensitize 5-HT_1A autoreceptors, thereby inhibiting their functional activity and reducing their ability to inhibit 5-HT transmission (Romero et al., 1996). Other possible mechanisms could involve postsynaptic 5-HT_1A receptors in terminal regions that participate in an inhibitory feedback loop to negatively regulate 5-HT transmission (Bosker et al., 1997; Casanovas et al., 1999) or direct effects on 5-HT release.

An important determinant of the effects of fluoxetine and EMD 68843 was the brain region being sampled. Previous microdialysis studies have described differential regulation of 5-HT transmission by the 5-HT_1A autoreceptor in distinct brain regions (Kreiss and Lucki, 1994, 1997; Malagie et al., 1996; Invernizzi et al., 1997; Romero and Artigas, 1997; Knobelman et al., 2001). Extracellular 5-HT levels demonstrate regional differences in sensitivity to 5-HT_1A autoreceptor regulation, as shown previously when the 5-HT_1A receptor antagonist WAY 100635 potentiated the effects of fluoxetine to a greater extent in the FC than in the dorsal hippocampus (Hervas and Artigas, 1998). Preferential innervation of the FC by serotonergic fibers from the DR and the innervation of the HPv by MR fibers (Kosofsky and Molliver, 1987) and greater responsiveness of DR neurons to the autoinhibitory actions of 5-HT on 5-HT_1A receptors (Sinton and Fallon, 1988; Blier et al., 1990; Romero and Artigas, 1997) may contribute to regional differences in the response to SSRIs. Also, fluoxetine appeared to induce a smaller increase of extracellular 5-HT in the frontal cortex than in the HPv, which is comparable with findings from other laboratories (Fuller et al., 1994; Kreiss and Lucki, 1995; Invernizzi et al., 1996; Hervas and Artigas, 1998).

EMD 68843 also elicited regional changes in extracellular 5-HT. The greater increase in the HPv than the FC with the low (0.3 mg/kg) dose reflected regional differences in the response to systemic administration of SSRIs. The 3.0 mg/kg dose increased extracellular 5-HT levels in the FC for a longer time than the HPv, possibly reflecting the influence of blockade of 5-HT_1A receptors in a region under greater control by the 5-HT_1A autoreceptor.

The involvement of 5-HT_1A autoreceptor mechanisms was addressed by examining the ability of the 5-HT_1A receptor agonist, 8-OH-DPAT, to reduce the elevated 5-HT levels achieved by EMD 68843 or fluoxetine administration. In the FC, 8-OH-DPAT reduced extracellular 5-HT levels to near baseline levels in fluoxetine-treated animals only, whereas the EMD 68843-treated animals demonstrated only a transient reduction of extracellular 5-HT in response to 8-OH-DPAT. This result is consistent with EMD 68843 interfering with the 5-HT_1A autoreceptor to limit the actions of 8-OH-DPAT as an autoreceptor agonist. Thus, the augmented levels of extracellular 5-HT produced by 3 mg/kg EMD 68843 probably involve the combination of 5-HT reuptake inhibition and interference with 5-HT_1A autoreceptors regulating 5-HT transmission in the FC. In contrast, in the HPv, fluoxetine- or EMD 68843-induced increases in extracellular 5-HT levels were attenuated in a similar fashion by subsequent systemic administration of 8-OH-DPAT, suggesting that 5-HT_1A autoreceptor function may exert less regulation of extracellular 5-HT and the effects of EMD 68843 in this region.

EMD 68843 was examined for its ability to produce the 5-HT syndrome and its interaction with the full 5-HT_1A receptor agonist 8-OH-DPAT. The 5-HT behavioral syndrome is a series of unconditioned behavioral responses in rats evoked to study the activation of postsynaptic 5-HT_1A receptors in vivo (Lucki, 1992). The behavioral syndrome is also an animal model for a serious side effect produced in humans by a number of antidepressant drugs (Gillman, 1998). EMD 68843 did not produce the 5-HT syndrome and completely prevented the actions of the more efficacious 5-HT_1A receptor agonist 8-OH-DPAT on the 5-HT behavioral syndrome. These results are similar to effects shown previously by other 5-HT_1A receptor partial agonists and antagonists (Singh and Lucki, 1993; Detke et al., 1995b) because only high-efficacy agonists can produce the 5-HT behavioral syndrome. Although a recent finding suggested that the full 5-HT_1A agonist flibanserin might block certain components of the 5-HT syndrome induced by 8-OH-DPAT (Borsini et al., 2001), this interference was attributed to the likely involvement of flibanserin with 1 or 5-HT_2A receptors. This is not the case for EMD 68843.

The behavioral results in the mouse and rat FST provide some support for the antidepressant potential of EMD 68843 but over a limited dose range. In the rat FST, EMD 68843 reduced immobility and increased swimming behavior. This response pattern is similar to that shown many times to be produced by conventional SSRIs, such as fluoxetine (10-20 mg/kg, s.c.), and is mediated by 5-HT (Lucki, 1997; Page et al., 1999). In the mouse FST, EMD 68843 reduced immobility in DBA/2J mice similar to fluoxetine (5-20 mg/kg, i.p.; Lucki et al., 2001). These findings are consistent with the idea that antidepressant-like actions are mediated through the activation of postsynaptic 5-HT receptors. However, doses of EMD 68843 higher than 1 mg/kg were ineffective in both the mouse and the rat FST. Although multiple 5-HT receptors may be involved in antidepressant behavioral responses, activation of postsynaptic 5-HT_1A receptors appears to be important for producing behavioral effects of SSRIs in antidepressant tests (Lucki et al., 1994; De Vry, 1995). Pretreatment with 5-HT_1A receptor partial agonists or antagonists, or genetic deletion of 5-HT_1A receptors can prevent the behavioral effects of SSRIs (Singh and Lucki, 1993; Redrobe and Bourin, 1998; Mayorga et al., 2001; Reneric et al., 2001). Thus, although the 5-HT_1A receptor partial agonist component of EMD 68843 may enhance extracellular 5-HT levels, at high doses it could also restrain expression of antidepressant-like behaviors in the FST, which predict clinical activity.

These findings have direct clinical implications for the development of faster-acting antidepressant compounds. The greater elevation of extracellular 5-HT levels achieved by EMD 68843 compared with fluoxetine can improve SSRI response and is qualitatively similar to augmentation effects achieved by the combination of SSRIs with pindolol (Perez, 1997; McAskill et al., 1998; Olver et al., 2000). A single molecule producing more efficacious enhancement of extracellular 5-HT may also be more advantageous for other indications of SSRIs and circumvent potential pharmacokinetic interactions caused by the use of drug combinations (Olver et al., 2000). However, preclinical studies implicate the activation of postsynaptic 5-HT_1A receptors as important for antidepressant responses to SSRIs (see Lucki et al., 1994; Blier et al., 1997; Cryan and Leonard, 2000). Whether the 5-HT_1A receptor partial agonist component of EMD 68843 would impede or facilitate antidepressant responses when tested clinically is unknown. In conclusion, the present findings demonstrate that a single compound with dual effects on 5-HT transporters and 5-HT_1A receptors can produce a greater impact on extracellular 5-HT levels than conventional SSRIs. These intrinsically augmented effects may contribute to increasing the therapeutic effects of serotonergic antidepressants.