Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Two types of serotonin or 5-hydroxytryptamine (5-HT) autoreceptors provide critical regulation of 5-HT release in the rat brain by supplying mechanisms for presynaptic inhibitory feedback. The 5-HT_1A autoreceptors are located in the somatodendritic neuronal region, at the site of the serotonergic cell bodies in the dorsal (DR) or median (MR) raphe nucleus, and regulate the release of 5-HT by modulating neurotransmitter synthesis, terminal release, and cell discharge rate (Hjorth et al., 1982; Sprouse and Aghajanian, 1988). In contrast, the 5-HT_1B autoreceptors regulate the release of 5-HT from nerve terminals in brain regions such as the frontal cortex, striatum, and hippocampus, and on terminals of afferents and collaterals in the DR and MR (Engel et al., 1986; Maura et al., 1986). 5-HT_1A and 5-HT_1B receptors are also located postsynaptically throughout the limbic forebrain and participate in the regulation of important behavioral and physiological functions (Boschert et al., 1994; De Vry, 1995).

The ability of autoreceptors to regulate extracellular levels of 5-HT during release has made them the focus of much interest. 5-HT autoreceptors are desensitized by the chronic administration of antidepressant drugs and this may account for the delay in appearance of therapeutic effects (Blier and de Montigny, 1994). Blockade of 5-HT_1A autoreceptors can potentiate the increase of extracellular 5-HT levels caused by selective serotonin reuptake inhibitors (SSRIs) (Malagie et al., 1996; Gobert et al., 1997; Invernizzi et al., 1997; Sharp et al., 1997; Hervas and Artigas, 1998; Knobelman et al., 2001; for review, see Hjorth et al., 2000). Blockade of 5-HT_1B autoreceptors, or both 5-HT_1A and 5-HT_1B autoreceptors, have also been shown to augment the increase of extracellular levels of 5-HT by SSRIs (Rollema et al., 1996; Gobert et al., 1997; Sharp et al., 1997). Because coadministration of 5-HT autoreceptor antagonists potentiates the increase of extracellular 5-HT in forebrain regions produced by SSRIs, clinical studies have combined drugs that block 5-HT autoreceptors, such as pindolol, with SSRIs to augment their clinical effects (Artigas et al., 1996).

Mice with genetic deletion of either the 5-HT_1A (Heisler et al., 1998; Parks et al., 1998; Ramboz et al., 1998) or 5-HT_1B (Saudou et al., 1994) receptor were generated to better study their functional roles. The 5-HT_1A receptor knockout mice demonstrated a pattern of increased anxiety-related behaviors, such as the elevated plus maze, the elevated zero maze, open field, and novel object exploration, whereas 5-HT_1B receptor knockout mice demonstrated the opposite phenotype in similar behaviors. 5-HT_1B receptor knockout mice showed additional behavioral changes, such as increased aggression and vulnerability to cocaine (Ramboz et al., 1998; Rocha et al., 1998; Brunner et al., 1999; Zhuang et al., 1999). 5-HT receptor mutant mice also demonstrate regional differences in the regulation of extracellular levels of 5-HT after the administration of SSRIs. 5-HT_1A receptor knockout mice demonstrate larger increases of 5-HT in the frontal cortex or striatum than the hippocampus (Knobelman et al., 2001; Parsons et al., 2001), whereas 5-HT_1B receptor knockout mice show larger effects of SSRIs in the hippocampus than the other two regions depending on the dose (Knobelman et al., 2001; Malagie et al., 2001). Regional differences in the regulation of 5-HT could contribute to phenotypic differences of key physiological or behavioral functions.

One of the problems of interpreting functional changes in mice with constitutive genetic deletions is that compensation by other genes or biological mechanisms over the course of development may restore dysfunction of the mutated gene. The influence of compensation can sometimes be assessed by comparing the effects of genetic deletion with selective pharmacological antagonists. It is unclear whether the absence of presynaptic or postsynaptic functions or the development of plasticity involving other receptors underlies the phenotypic consequences of 5-HT_1A and 5-HT_1B receptor mutant mice. Because 5-HT_1A and 5-HT_1B autoreceptors regulate the release of 5-HT at different sites, the serotonergic system provides a unique opportunity to examine compensation between somatodendritic and terminal autoreceptors that may develop from their genetic deletion. The absence of function by one of the autoreceptors that regulates 5-HT could lead to adaptive compensation by the complementary autoreceptor. If 5-HT_1A or 5-HT_1B autoreceptors play a more important role in regulating the release of 5-HT in different brain regions, compensation for the loss of 5-HT autoreceptors may also develop with a regionally specific pattern. The purpose of this study was to examine functional changes in the functional regulation of extracellular 5-HT in different brain regions in mice with genetic deletion of either 5-HT_1A or 5-HT_1B receptors. An understanding of reciprocal interactions between 5-HT_1A and 5-HT_1B receptors could provide greater insight into events that regulate the release of 5-HT and its functional consequences.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Subjects. Male homozygote 5-HT_1A receptor knockout, 5-HT_1B receptor knockout, and wild-type mice were derived and raised in a colony at the University of Pennsylvania (Philadelphia, PA). Breeding founders were obtained from established colonies derived originally from the same strain (Saudou et al., 1994; Ramboz et al., 1998) by Dr. René Hen, Columbia University (New York, NY) [see Phillips et al. (1999) for more details about the genetic background]. Mice were generated by breeding homozygote mutant or wild-type mice. Both mutants and wild-type mice were derived from the same 129/SV background. After weaning, mice were housed four per cage, given free access to standard rodent chow and water, and maintained on a 12-h alternating light/dark schedule, with lights on at 7:00 AM. Mice were 8 to 12 weeks of age when used in these studies. All studies were carried out in accordance with the Guide for the Care and Use of Laboratory Animals by the U.S. National Institutes of Health and were approved by the Institutional Animal Care and Use Committee at the University of Pennsylvania.

Surgery. Dialysate measures were obtained from separate groups of mice implanted with microdialysis probes aimed at either the striatum or the ventral hippocampus. Mice were anesthetized with chloral hydrate (400 mg/kg i.p.) and positioned in a mouse stereotaxic instrument (Kopf Instruments, Tujunga, CA). Custom-made microdialysis probes were prepared for use in the mouse using 26-gauge stain steel tubing, as described previously (Knobelman et al., 2000). The probes were aimed at the following coordinates taken from bregma according to the atlas of Franklin and Paxinos (1997): striatum, +0.6 mm AP, +1.7 mm ML, and 4.5 mm DV; ventral hippocampus, 2.8 mm AP, 3.5 mm ML, and 5.0 mm DV. A drop of cyanoacrylate was spread thinly over the exposed skull and the probe was then cemented in place. Following surgery, the mice were placed into a 21.5-cm-high clear polycarbonate cylindrical in vivo microdialysis apparatus with a counterbalance arm holding a liquid swivel (Instech Laboratories, Plymouth Meeting, PA) and allowed to recover overnight.

Dialysis Procedure. Microdialysis procedures in the mouse were performed as previously described (Knobelman et al., 2000). The probes were continuously perfused with filtered artificial cerebrospinal fluid (147 mM NaCl, 1.7 mM CaCl₂, 0.9 mM MgCl₂, and 4 mM KCl, pH 6.3-6.5) at a rate of 0.8 µl/min using a Harvard Apparatus syringe pump (Instech Laboratories). Dialysate samples were collected into polypropylene microcentrifuge tubes starting 17 to 20 h after surgery at 20-min intervals for 2 h prior to the first injection, to obtain baseline values, and for 3 h after injections to measure drug effects.

Analysis of Dialysate. Samples were automatically injected into a Bioanalytical Systems 460 high-pressure liquid chromatograph by a BAS sample sentinel refrigerated microsampler set to a 12-µl injection volume. The high-pressure liquid chromatograph mobile phase [12.42 mM citric acid, 39.85 mM NaPO₄ (monobasic), 0.25 mM EDTA, 0.737 mM 1-decanesulfonic acid, 10.0 mM NaCl, 0.2% triethylamine, 15-19% MeOH, pH 4.3] was pumped through a reverse phase 1 × 100 mm ODS 3-µm microbore column (C₁₈; BAS) with a 10-µl sample loop at a flow rate of 90 µl/min (Kreiss et al., 1993). The 5-HT from chromatographs of dialysate samples was identified by comparing their elution times with those of reference standards. The amount of 5-HT in each dialysate sample was quantified from their respective peak heights using a linear regression analysis of the peak heights obtained from a series of reference standards.

Histology. At the completion of the experiment, brains were removed and frozen at 80°C. The brains were then sectioned with a refrigerated cryostat, stained with Neutral Red, and the tissue was examined for the location of the dialysis probe. Data from animals with probes located outside of the target regions were not used. This procedure was not followed in all cases because of an experimental error that caused destruction of some of the histological samples. Histology was examined in 38 of 85 animals with placements in the striatum (62% wild-type, 40% 5-HT_1A/ and 27% 5-HT_1B/), and 28 of 85 animals with placements in the ventral hippocampus (33% wild-type, 40% 5-HT_1A/, and 26% 5-HT_1B/). Of the 66 total animals in which placements were verified it was necessary to exclude only one animal, whose target region was in the striatum, due to probe placement outside of this region.

PCR Genotyping. Because of homozygote breeding, a random sample of approximately 25% of participating mice were genotyped to verify that animals used in the study demonstrated the expected genetic deletion. This was confirmed in all cases. Mice were genotyped by PCR analysis. Briefly, tail biopsies were digested in 0.2 ml of NID-buffer (50 mM KCl, 10 mM Tris/Cl pH 8.3, 2 mM MgCl₂, 0.1 mg/ml gelatin, 0.45% Nonidet P-40, 0.45% Tween 20) and 1.2 µl proteinase K (10 mg/ml) overnight at 56°C. Tail DNA (1-3 µl) was used directly in the PCR reaction. For genotyping 5-HT_1A receptor knockout mice, the following conditions and PCR primers were used: 95°C; 60 s/65°C; 60 s/72°C; 90 s, 32 cycles. NEO primer: GCC TTC TAT CGC CTT CTT CTT GAC G; 5' 5-HT_1A receptor primer: CCA ACT ATC TCA TCG GCT CCT T; 3' 5-HT_1A receptor primer: GCT CCC TTC TTT TCC ACC TTC T. Expected sizes of PCR products were 450 bp for the wild-type allele and 750 bp for the mutant allele. For genotyping 5-HT_1B receptor knockout mice, the following conditions and PCR primers were used: 95°C; 90 s/55°C; 120 s/72°C; 120 s, 35 cycles. NEO primer: CTT CTA TCG CCT TCT TGA CG; 5' 5-HT_1B receptor primer: GAC TTG GTT CAC GTA CAC AG; 3' 5-HT_1B receptor primer: CCC ATC AGC ACC ATG TAC AC. Expected sizes of PCR products were 500 bp for the wild type allele and 680 bp for the mutant allele.

Data Analysis. The first four samples were averaged to derive the baseline value and were corrected for individual probe recovery to compare against archived values from this laboratory. Probe recovery in vitro was measured with a standard solution of artificial cerebrospinal fluid containing 5-HT (10 nM) at room temperature. Baseline values were compared between wild-type and mutants using a single-factor analysis of variance (ANOVA).

Drugs. All drugs were prepared fresh before use. Fluoxetine hydrochloride (Eli Lilly, Indianapolis, IN), WAY 100635 maleate (Wyeth Ayerst, Philadelphia, PA), GR 127935 hydrochloride (Glaxo Wellcome, Hertfordshire, UK), CP 94,253 (Pfizer, Groton, CT) and R-8-OH-DPAT hydrobromide (Sigma/RBI, Natick, MA) were dissolved in deionized water and administered in a volume of 8 ml/kg i.p. Doses were calculated as the weight of the base.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Effects of R-8-OH-DPAT in Wild-Type, 5-HT_1A, and 5-HT_1B Receptor Knockout Mice. The effects of R-8-OH-DPAT differed significantly across genotype (Table 1). Acute administration of the 5-HT_1A receptor agonist R-8-OH-DPAT (1.0 mg/kg) produced a decrease of 5-HT from basal levels in the striatum of wild-type mice to a minimum of 39 ± 6% (Fig. 1A). In comparison, follow-up tests showed that the overall effect of R-8-OH-DPAT on extracellular 5-HT was significantly blunted in the 5-HT_1A receptor knockout mice (P > 0.05). In contrast, there was no significant difference between the response of wild-type and 5-HT_1B receptor knockout mice.

View larger version (22K): [in this window] [in a new window]   Fig. 1.   Effects of R-8-OH-DPAT on extracellular 5-HT levels in the 5-HT1A receptor knockout mouse, 5-HT1B receptor knockout mouse, and wild-type mouse. A, effects of systemic administration of 1.0 mg/kg R-8-OH-DPAT in the striatum of wild-type mice (; n = 7; baseline = 6.21 ± 0.28 fmol/10 µl), 5-HT1A receptor knockout mice (; n = 7; baseline = 8.07 ± 1.10 fmol/10 µl), and 5-HT1B receptor knockout mice (; n = 7; baseline = 6.55 ± 0.53 fmol/10 µl) on extracellular 5-HT. B, effects of systemic administration of 1.0 mg/kg R-8-OH-DPAT in the ventral hippocampus of wild-type mice (; n = 7; baseline = 7.83 ± 1.33 fmol/10 µl), 5-HT1A receptor knockout mice (; n = 6; baseline = 7.91 ± 1.67 fmol/10 µl), and 5-HT1B receptor knockout mice (; n = 7; baseline = 7.25 ± 1.26 fmol/10 µl) on extracellular 5-HT. Values represent mean changes in 5-HT content, expressed as percentage of baseline values. Vertical lines indicate 1 S.E.M.

Effects of CP 94,253 in Wild-Type, 5-HT_1A, and 5-HT_1B Receptor Knockout Mice. Acute administration of the selective 5-HT_1B receptor agonist CP 94,253 (1.0 mg/kg) reduced extracellular levels of 5-HT differently in 5-HT_1A receptor knockout, 5-HT_1B receptor knockout, and wild-type mice (Table 1). As shown in Fig. 2A, CP 94,253 significantly decreased extracellular levels of 5-HT in the striatum of the wild-type mice to a minimum of 66 ± 9%. Striatal 5-HT levels were decreased significantly more in the 5-HT_1A receptor knockouts than wild-type mice (P < 0.05), to a minimum of 51 ± 11%. In contrast, the effect of CP 94,253 was blunted in 5-HT_1B receptor knockout mice (minimum 5-HT levels were 92 ± 8%).

View larger version (22K): [in this window] [in a new window]   Fig. 2.   Effects of CP 94,253 on extracellular 5-HT levels in wild-type mice, 5-HT1A receptor knockout mice, and 5-HT1B receptor knockout mice. A, effects of systemic administration of 1.0 mg/kg CP 94,253 in the striatum of wild-type mice (; n = 7; baseline = 6.58 ± 0.50 fmol/10 µl), 5-HT1A receptor knockout mice (; n = 6; baseline = 6.39 ± 1.09 fmol/10 µl), and 5-HT1B receptor knockout mice (; n = 7; baseline = 5.27 ± 0.50 fmol/10 µl) on extracellular 5-HT. B, effects of systemic administration of 1.0 mg/kg CP 94,253 in the ventral hippocampus of wild-type mice (; n = 6; baseline = 7.27 ± 0.90 fmol/10 µl), 5-HT1A receptor knockout mice (; n = 6; baseline = 7.67 ± 0.60 fmol/10 µl), and 5-HT1B receptor knockout mice (; n = 6; baseline = 6.11 ± 0.79 fmol/10 µl) on extracellular 5-HT. Values represent mean changes in 5-HT content, expressed as percentage of baseline values. Vertical lines indicate 1 S.E.M.

Effects of Fluoxetine and WAY 100635 in Wild-Type and 5-HT_1B Receptor Knockout Mice. The effects of fluoxetine (20 mg/kg), given alone or with WAY 100635 (0.1 mg/kg), on extracellular levels of 5-HT in the striatum (Fig. 3A) differed significantly in wild-type and 5-HT_1B receptor knockout mice (Table 1). In wild-type mice, fluoxetine alone elicited a maximum increase of 349 ± 46% at peak, and WAY 100635 augmented fluoxetine's effects to a maximum of 683 ± 84% (P < 0.05). In the 5-HT_1B receptor knockout mice, fluoxetine alone elicited a maximum increase of 319 ± 43% of baseline. The additional administration of WAY 100635 augmented fluoxetine's effects to 619 ± 83% (P < 0.05). The AUC values (Fig. 3B) show that WAY 100635 significantly increased the effects of fluoxetine in both wild-type mice and 5-HT_1B receptor knockout mice (P < 0.05), but there was no difference in drug effects between genotype.

View larger version (41K): [in this window] [in a new window]   Fig. 3.   Effect of fluoxetine (20 mg/kg) with and without concurrent administration of WAY 100635 (0.1 mg/kg) on extracellular 5-HT in wild-type and 5-HT1B receptor knockout mice. A, effects of systemic administration of either 20 mg/kg fluoxetine alone in the wild-type mice (; n = 7; baseline = 8.41 ± 1.06 fmol/10 µl) and 5-HT1B receptor knockout mice (; n = 6; baseline = 9.22 ± 0.80 fmol/10 µl) or fluoxetine followed 80 min later by systemic administration of 0.1 mg/kg WAY 100635 in the wild-type mice (; n = 6; baseline = 6.67 ± 1.27 fmol/10 µl) and 5-HT1B receptor knockout mice (; n = 6; baseline = 9.32 ± 0.75 fmol/10 µl) on extracellular 5-HT levels in the striatum. Values represent mean changes in 5-HT content, expressed as a percentage of baseline values. Vertical lines indicate 1 S.E.M. Bars in B represent mean AUC values summed from the effects measured 80 to 200 min after administration of fluoxetine and immediately following acute systemic administration of WAY 100635, as indicated by the arrow in A. Vertical lines indicate 1 S.E.M. C, effects of systemic administration of either 20 mg/kg fluoxetine alone in the wild-type mice (; n = 7; baseline = 6.69 ± 0.64 fmol/10 µl) and 5-HT1B receptor knockout mice (; n = 7; baseline = 4.78 ± 0.36 fmol/10 µl) or fluoxetine followed 80 min later by systemic administration of 0.1 mg/kg WAY 100635 in the wild-type mice (; n = 7; baseline = 8.33 ± 1.24 fmol/10 µl) and 5-HT1B receptor knockout mice (; n = 7; baseline = 8.48 ± 1.49 fmol/10 µl) on extracellular 5-HT levels in the ventral hippocampus. Values represent mean changes in 5-HT content, expressed as a percentage of baseline values. Vertical lines indicate 1 S.E.M. Bars in D represent mean AUC values summed from the effects measured 80 to 200 min after administration of fluoxetine and immediately following acute systemic administration of WAY 100635, as indicated by the arrow in C. Vertical lines indicate 1 S.E.M.

Effects of Fluoxetine and GR 127935 in the Wild-Type and 5-HT_1A Receptor Knockout Mice. The effects of fluoxetine (20 mg/kg) given alone or with GR 127935 (0.056 mg/kg) on extracellular levels of 5-HT in the striatum (Fig. 4A) differed significantly between 5-HT_1A receptor knockout and wild-type mice (Table 1). In wild-type mice, extracellular 5-HT levels were increased to a maximum of 349 ± 46% of baseline by fluoxetine alone and to 440 ± 27% of baseline by GR 127935 plus fluoxetine, but this difference was not statistically significant (P > 0.05). In the 5-HT_1A receptor knockout mice, fluoxetine alone elicited a maximum increase of 877 ± 103% of baseline. The additional administration of GR 127935 augmented the effects of fluoxetine to 1275 ± 181% of baseline (P < 0.05). The AUC values (Fig. 4B) showed that GR 127935 augmented the effects of fluoxetine only in 5-HT_1A receptor knockout mice (P < 0.05).

View larger version (41K): [in this window] [in a new window]   Fig. 4.   Effect of fluoxetine (20 mg/kg) with and without concurrent administration of GR 127935 (0.056 mg/kg) on extracellular 5-HT in wild-type and 5-HT1A receptor knockout mice. A, effects of systemic administration of either 20 mg/kg fluoxetine alone in the wild-type mice (; n = 7; baseline = 8.41 ± 1.06 fmol/10 µl; same as Fig. 3) and 5-HT1A receptor knockout mice (; n = 6; baseline = 8.56 ± 1.21 fmol/10 µl) or fluoxetine followed 80 min later by systemic administration of 0.056 mg/kg GR 127935 in the wild-type mice (; n = 7; baseline = 8.51 ± 0.56 fmol/10 µl) and 5-HT1A receptor knockout mice (; n = 6; baseline = 6.05 ± 0.34 fmol/10 µl) on extracellular 5-HT levels in the striatum. Values represent mean changes in 5-HT content, expressed as a percentage of baseline values. Vertical lines indicate 1 S.E.M. Bars in B represent mean AUC values summed from the effects measured 80 to 200 min after administration of fluoxetine and immediately following acute systemic administration of GR 127935, as indicated by the arrow in A. Vertical lines indicate 1 S.E.M. C, effects of systemic administration of either 20 mg/kg fluoxetine alone in the wild-type mice (; n = 7; baseline = 6.69 ± 0.64 fmol/10 µl; same as Fig. 3) and 5-HT1A receptor knockout mice (; n = 6; baseline = 7.51 ± 1.34 fmol/10 µl) or fluoxetine followed 80 min later by systemic administration of 0.056 mg/kg GR 127935 in the wild-type mice (; n = 6; baseline = 9.42 ± 0.65 fmol/10 µl) and 5-HT1A receptor knockout mice (; n = 7; baseline = 6.12 ± 0.64 fmol/10 µl) on extracellular 5-HT levels in the ventral hippocampus. Values represent mean changes in 5-HT content, expressed as a percentage of baseline values. Vertical lines indicate 1 S.E.M. Bars in D represent mean AUC values summed from the effects measured 80 to 200 min after administration of fluoxetine and immediately following acute systemic administration of GR 127935, as indicated by the arrow in C. Vertical lines indicate 1 S.E.M.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

In a constitutive deletion, genes are disabled permanently from conception and compensation by other genes and biological mechanisms may develop to restore the function of the mutated gene. Since 5-HT_1A and 5-HT_1B autoreceptors regulate the release of 5-HT from different neuronal locations, the serotonergic system provides a unique opportunity to examine compensation between somatodendritic and terminal autoreceptors that may develop from their genetic deletion. The effectiveness of the genetic deletions was shown by the inability of selective 5-HT autoreceptor agonists to reduce extracellular 5-HT levels of mice with a corresponding absence of either 5-HT_1A or 5-HT_1B receptors. These studies confirmed functional deficits of 5-HT_1A or 5-HT_1B autoreceptors in 5-HT_1A receptor or 5-HT_1B receptor knockout mice (Trillat et al., 1997), respectively. The absence of a response to CP 94,253 in 5-HT_1B receptor knockout mice also indicated that 5-HT_1D autoreceptors would not appear to be activated by the challenge dose. Selective 5-HT_1A or 5-HT_1B receptor antagonists have been shown similarly to block the effects of corresponding agonists in wild-type mice (Knobelman et al., 2000).

Doses of R-8-OH-DPAT and CP 94,253 were selected to produce large effects to measure deficits produced by genetic deletions, but there were no apparent differences in effects between the two regions for either drug. Previous studies in rats have suggested that DR-innervated regions may more sensitive to the effects of 5-HT_1A receptor agonists than MR-innervated regions (Blier et al., 1990; Casanovas and Artigas, 1996; Casanovas et al., 1997), although these regional differences for 8-OH-DPAT have not been shown consistently (Sinton and Fallon, 1988; Hajos et al., 1995). However, the issue of regional differences for 5-HT_1A and 5-HT_1B receptor agonists in the mouse remains unsettled because differences in sensitivity were not examined. The present determinations differ from prior studies because the effects of the agonists were measured in the absence of anesthesia and without SSRIs added to the perfusion media to increase basal levels.

Topographical differences between forebrain regions have been suggested to exist for regulating the effects of SSRIs. 5-HT_1A receptor antagonists augment the effects of SSRIs in areas with predominant DR innervation, such as the frontal cortex and striatum, but are less effective (or not at all) in the MR-innervated dorsal hippocampus (Malagie et al., 1996; Invernizzi et al., 1997; Romero and Artigas, 1997; Sharp et al., 1997; Hervas and Artigas, 1998; Hjorth et al., 2000). Recent microdialysis studies in 5-HT_1A and 5-HT_1B receptor mutant mice also demonstrated that the effects of SSRIs vary between forebrain regions. In 5-HT_1A receptor knockout mice, the effects of fluoxetine were augmented more in the frontal cortex and striatum than in the ventral hippocampus (Knobelman et al., 2001; Parsons et al., 2001), whereas the converse pattern of effects occurred for the effects of fluoxetine and paroxetine in 5-HT_1B receptor knockout mice (Knobelman et al., 2001; Malagie et al., 2001). Because similar regional differences were also produced by treating wild-type mice with a corresponding 5-HT receptor antagonist, the augmented effects of SSRIs appear due to deletion of the targeted receptor. The reason for the existence of such topographical differences is unresolved. The differences may involve intrinsic differences involving the anatomical location of their cells of origin, regional differences in the factors regulating the balance between release and autoreceptor inhibition, or different patterns of afferent innervation (Hervas et al., 2000; Hjorth et al., 2000).

One instance of compensation identified in 5-HT_1A receptor knockout mice was that terminal 5-HT_1B autoreceptors in the striatum appeared to increase their responsiveness. The regulation of extracellular 5-HT levels by the 5-HT_1B receptor agonist CP 94,253 was significantly enhanced when measured in the striatum but was unaltered when measured in the ventral hippocampus. A second challenge, measuring the augmentation of fluoxetine's effects by the 5-HT_1B/1D receptor antagonist GR 127935, showed that the antagonist significantly increased fluoxetine's effects in the striatum more in 5-HT_1A receptor mutants than in wild-type mice. The altered effects of CP 94,253 or GR 127935 plus fluoxetine were confined to the ventral hippocampus and did not occur in the striatum. This pattern is consistent with the development of increased regulation of extracellular 5-HT in the striatum by terminal 5-HT_1B autoreceptors in compensation for the absence of 5-HT_1A receptors. The compensatory change may have been restricted to the striatum because of the important role of the 5-HT_1A autoreceptor in regulating 5-HT release in that region (Knobelman et al., 2001). The mechanism by which striatal 5-HT_1B autoreceptors developed greater sensitivity in the absence of 5-HT_1A receptors is not clear from these studies. Autoradiography studies have revealed no changes in the number of 5-HT_1B receptors in the 5-HT_1A receptor knockout mice compared with wild-type mice (R. Hen, personal communication), but receptor autoradiography could not discriminate presynaptic from postsynaptic 5-HT_1B receptors. Altered function of terminal 5-HT_1B receptors could be caused by changes in the signaling pathway of the 5-HT_1B autoreceptor. In contrast, a small increase in activity of terminal 5-HT_1B autoreceptors was reported in hippocampal slices taken from 5-HT_1A receptor knockout mice (Ramboz et al., 1998), justifying examination of more widespread distribution of this pattern of compensation. The slight discrepancy between studies may be due to some of the many differences between techniques.

Evidence also was obtained from 5-HT_1B receptor knockout mice that the response to the 5-HT_1A receptor agonist R-8-OH-DPAT was blunted when measured in the ventral hippocampus. This was confirmed by studies that measured a significantly smaller augmentation of the effects of fluoxetine by the 5-HT_1A receptor antagonist WAY 100635 in the ventral hippocampus of 5-HT_1B receptor knockout mice than wild-type mice. No significant change in response to either pharmacological challenge involving 5-HT_1A receptors was measured in the striatum of 5-HT_1B receptor knockout mice. The decreased response to 5-HT_1A receptor activation in mice absent 5-HT_1B receptors developed in a brain region where the 5-HT_1B autoreceptors have been shown to be more important in regulating 5-HT release (Knobelman et al., 2001). Decreased autoregulation of hippocampal 5-HT by the somatodendritic 5-HT_1A autoreceptor could develop from the traditional "desensitization" of the autoreceptor that has been documented following chronic administration of SSRIs or the 5-HT_1A receptor agonist 8-OH-DPAT (Kreiss and Lucki, 1992, 1995; Rutter et al., 1994; Invernizzi et al., 1996). Although necessarily speculative, the absence of 5-HT_1B autoreceptors on 5-HT reciprocal collaterals in the MR, or 5-HT heteroreceptors on afferent innervation to the MR, could chronically activate and desensitize 5-HT_1A autoreceptors in the MR (Boschert et al., 1994; Pineyro and Blier, 1999). Altered neuronal activity of mouse DR neurons and a higher density of 5-HT transporter binding could provide the basis for further investigating physiological mechanisms in the MR underlying compensation in the 5-HT_1B receptor knockout mouse (Evrard et al., 1999).

Based on their altered response to microdialysis studies, we propose that striatal 5-HT_1B receptors increase their sensitivity in 5-HT_1A receptor knockout mice and MR 5-HT_1A receptors are desensitized in 5-HT_1B receptor knockout mice. These proposals are based on the current view of how 5-HT_1A and 5-HT_1B receptors regulate extracellular 5-HT in the striatum and ventral hippocampus (Kreiss and Lucki, 1994; Hjorth et al., 1997; Romero and Artigas, 1997). However, endogenous 5-HT in some regions can regulate its release by activating inhibitory feedback loops traversing back to the DR (Bosker et al., 1997; Casanovas et al., 1999) or by interactions with other neurotransmitters (Boschert et al., 1994). Because the present studies involved systemic administration of pharmacological challenges, they did not determine the location of the autoreceptors responsible for augmenting the effects of fluoxetine in 5-HT receptor mutants. Verification of the proposed mechanisms regulating autoreceptor compensation requires local administration of 5-HT_1A and 5-HT_1B receptor agonists into target regions of the mouse brain.

The influence of the interplay between somatodendritic and terminal autoreceptors on regional patterns of 5-HT release in 5-HT receptor mutant mice has important implications for genetic, pharmacological, and clinical studies. First, differences in 5-HT transmission produced by genetic mutation can involve compensatory processes rather than the direct absence of the receptor. Second, because of the permanent receptor loss, neuronal compensations in knockout mice could suggest additional targets that are regulated by chronic drug treatments or brain lesions. For this reason, constitutive 5-HT_1A and 5-HT_1B receptor knockout mice may be better suited as models of disease states than a direct archetype of 5-HT receptor function (Scearce-Levie et al., 1999). Finally, regional variations in 5-HT transmission could contribute to different expression of behavioral phenotypes or altered drug responses in 5-HT receptor mutant mice (Zhuang et al., 1999) or in humans with genetic polymorphisms that produce alterations of these receptors (Veenstra-VanderWeele et al., 2000).