Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

A wide variety of stressors can increase brain tryptophan content and subsequently affect serotonin (5-HT) metabolism (Curzon et al., 1972; Chaouloff et al., 1985; Dunn, 1988a). Stress-related elevations in brain tryptophan are normal in adrenalectomized animals (Curzon et al., 1972; Dunn and Welch, 1991), but can be blocked by ganglionic blockers and -adrenergic antagonists (Dunn and Welch, 1991), suggesting that the changes are a consequence of peripheral sympathetic activation of -adrenergic receptors. Indeed, peripheral administration of the nonselective -adrenergic agonist isoproterenol (Eriksson and Carlsson, 1988) or the ₂-selective agonist clenbuterol (Edwards et al., 1989) results in comparable increases in brain tryptophan in rats. Likewise, imipramine, which inhibits norepinephrine reuptake, elevates brain tryptophan through a -adrenergic-dependent mechanism, as evidenced by prevention of its effects by propranolol (Edwards and Sorisio, 1988).

There are three known subtypes of -adrenergic receptors, ₁, ₂, and ₃, all of which are thought to couple primarily via G_s to adenylyl cylase, leading to an increase in cyclic adenosine monophosphate (cAMP), although recent evidence suggests that ₂- and ₃-receptors can also couple to G_i (Soeder et al., 1999; Xiao et al., 1999). The receptor subtypes differ primarily in their location: ₁-adrenoceptors predominate in the heart, cerebral cortex, and kidney (Minneman et al., 1979; McPherson et al., 1984; Engel et al., 1985). The major -adrenergic subtype in the lungs, cerebellum, uterus, skeletal muscle, and blood vessels is the ₂-adrenoceptor (Minneman et al., 1979; Carswell and Nahorski, 1983; McPherson et al., 1984; O'Donnell and Wanstall, 1985; Jensen et al., 1995). The ₃-adrenoceptor is expressed at high levels in brown and white adipose tissue, but has also been detected in brain, stomach, and gall bladder (Guillaume et al., 1994; Summers et al., 1995; Evans et al., 1996).

Although Edwards et al. (1989) suggested that the effect of -agonists on brain tryptophan is selective for ₂-adrenoceptors, it remains unknown whether ₃-adrenoceptors affect brain tryptophan. The current study was designed to determine the receptor subtype-selectivity for the increase in brain tryptophan produced by -adrenergic agonists in mice. Because the rate of 5-HT synthesis is directly influenced by the availability of tryptophan to tryptophan hydroxylase (Fernstrom, 1983), the present results may have important implications for 5-HT synthesis, and potentially for depression and other disorders associated with brain 5-HT.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals. Male CD-1 virus-antigen free (VAF plus) mice weighing between 18 and 20 g were obtained from Charles River (Raleigh-Durham facility, Colony R16). Mice were group-housed at 22-23°C in an Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC)-accredited animal care facility under a 12-12 light/dark cycle with lights on at 7:00 AM. Both food and water were available ad libitum. At least 48 h before each experiment, mice were placed in individual cages to avoid problems associated with disturbing group-housed animals. For studies with ₃-adrenergic receptor (AR) null mice, 8- to 10-week-old male FVB/NJ (WT) and age-matched FVB/NJ male mice with a targeted disruption of the ₃-AR gene (₃-AR KO) (Susulic et al., 1995) were used. All procedures were approved by the Louisiana State University Health Sciences Center-Shreveport Animal Care and Use Committee.

Experimental Procedures. Mice were injected intraperitoneally (i.p.) with various -agonists dissolved in 0.9% sterile saline at a volume of 10 µl/g of body weight. In experiments to determine receptor subtype-selectivity, antagonists were administered 20 min before agonists. Mice were sacrificed by decapitation 1 h following the last injection unless noted otherwise. The brain was rapidly removed and frontal cortex, hypothalamus, and brain stem were dissected as previously described (Dunn, 1988b). Brain regions were quickly weighed in tared Eppendorf tubes and frozen on dry ice. Tryptophan, serotonin (5-HT), and its major catabolite, 5-hydroxyindoleacetic acid (5-HIAA) were analyzed by HPLC with electrochemical detection as described previously (Dunn, 1993). For some studies, a shortened HPLC run was used to measure tryptophan only (retention time 7.5 min). For this, we used a Spherisorb octadecyl silane (ODS 1) reverse-phase column (25 cm, 5 µm; Keystone Scientific, Inc., Bellefonte, PA) shortened to 12.5 cm. The mobile phase contained 0.05 M NaH₂PO₄ (pH 2.75), 0.1 mM EDTA, 0.5 mM octanesulfonic acid (sodium salt), and 5% acetonitrile.

Drugs. ICI 118551 hydrochloride ((±)-1-[2,3-(dihydro-7-methyl-1H-inden-4-yl)oxyl]-3-[(1-methylethyl) amino]-2-butanol (Zeneca Pharmaceuticals, formerly ICI Pharmaceuticals, Cheshire, UK), betaxolol hydrochloride (1-[4-[2-(cyclopropylmethoxy) ethyl] phenoxy]-3-isopropylamino-2-propranol (LERS Synthelabo, Paris, France), and dobutamine hydrochloride ((±-3,4-dihydroxy-N-(3-[4-hydroxyphenyl]-1-methylpropyl)--phenethylamine; Eli Lilly, Indianapolis, IN) were obtained from Dr. James O'Donnell. Bupranolol hydrochloride was provided by Schwarz Pharma (Monheim, Germany). Clenbuterol hydrochloride (4-amino-a-(t-butylaminomethyl)-3,5-dichlorobenzyl alcohol hydrochloride), BRL 37344 sodium salt ((±)-(R*,R*)-[4-[2-[[2-(3-chlorophenyl)-2-hydroxyethyl] amino) propyl] phenoxy]acetic acid sodium), CL 316243 (disodium 5-[(2R)-2-([(2R)-2-(3-chlorophenyl)-2-hydroxyethyl] amino) propyl]-1,3-benzodioxole-2,2-dicarboxylate), SR 59230A oxalate salt (3-(2-ethylphenoxy)-1[1S)-1,2,3,4-tertahydronaphth-1-yla-mino]-(2S)-2-propanol oxalate), (S)-propranolol hydrochloride ((S)-1-isopropylamino-3-(1-naphthyoxy)-2-propanol hydrochloride), and nadolol were obtained from Sigma-Aldrich (St. Louis, MO).

Statistical Analysis. Data are represented as means ± S.E.M. Two-way ANOVA was performed to determine interactions between agonists and antagonists. Fisher's least significant difference test was used for individual comparisons. One-way ANOVA followed by Fisher's least significant difference test was used for the time course studies. Student's two-tailed t test was used for studies in which there were only two groups. Significance was accepted at p < 0.05.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Time Course Studies. Mice were injected with the ₂-selective agonist clenbuterol (0.3 mg/kg) or saline and brain samples collected 15, 30, 60, or 120 min following the injection. The tryptophan concentration in hypothalamus increased within 30 min, reached a maximum within 1 h, and returned to pretreatment levels within 2 h (Fig. 1A). Interestingly, 5-HIAA:5-HT, an index of 5-HT metabolism, was significantly elevated 1 h after the maximum tryptophan increase, perhaps reflecting increased tryptophan availability (Fig. 1B). Similar results were obtained in brain stem (data not shown).

View larger version (17K): [in this window] [in a new window]   Fig. 1.   Effect of the 2-adrenoceptor-selective agonist, clenbuterol (0.3 mg/kg), on A, hypothalamic tryptophan concentrations, and B, 5-HIAA:5-HT ratios. Mice were injected with saline (n = 5) or clenbuterol (n = 6) and brain samples collected 15, 30, 60, and 120 min later. Tryptophan, 5-HIAA, and 5-HT were determined in brain stem and hypothalamus **, significantly different from corresponding saline (p < 0.01; ***, p < 0.001).

Effects of Nonselective -Adrenoceptor Antagonists. Mice were injected with 2.5 mg/kg S-propranolol (Fig. 2A) or nadolol (Fig. 2B) 20 min before clenbuterol (0.1 mg/kg) or saline administration. Clenbuterol significantly elevated brain tryptophan in brain stem, hypothalamus, and frontal cortex, an effect that was prevented by pretreatment with either propranolol or nadolol (p < 0.01).

View larger version (42K): [in this window] [in a new window]   Fig. 2.   Effects of nonselective -adrenoceptor antagonists on the clenbuterol-induced increase in brain tryptophan. Mice were injected with propranolol (2.5 mg/kg) (A); or nadolol (2.5 mg/kg) (B) 20 min before clenbuterol (0.1 mg/kg) or saline (n = 6) and sacrificed 1 h later. Tryptophan was determined in brain stem, hypothalamus, and frontal cortex. Two-way ANOVA indicated significant agonist-antagonist interactions for propranolol in brain stem (F1,20 = 9.04, p < 0.01), hypothalamus (F1,19 = 10.2, p < 0.01), and frontal cortex (F1,20 = 7.73, p < 0.02), and for nadolol in brain stem (F1,20 = 29.6, p < 0.001), hypothalamus (F1,20 = 14.1, p < 0.01), and frontal cortex (F1,19 = 12.5, p < 0.01). *, significantly different from saline controls (p < 0.05; **, p < 0.01). , significantly different from saline-clenbuterol (p < 0.01).

Effects of Subtype-Selective -Adrenoceptor Antagonists. Figure 3 illustrates the effects of selective -adrenergic antagonists on the clenbuterol-induced increase in brain tryptophan. The ₂-selective antagonist, ICI 118551 (0.5 mg/kg, Fig. 3A), and the ₁-selective antagonist, betaxolol (1 mg/kg, Fig. 3B), were administered 20 min before clenbuterol or saline. A significant interaction was detected between clenbuterol and ICI 118551 in all brain regions (p < 0.01), but not between clenbuterol and betaxolol.

View larger version (41K): [in this window] [in a new window]   Fig. 3.   Effects of subtype-selective -adrenoceptor antagonists on the clenbuterol-induced increase in tryptophan. Mice (n = 6) were injected with the 2-selective ICI 118551 (0.5 mg/kg) (A); or the 1-selective betaxolol (1 mg/kg) (B) 20 min before clenbuterol (0.1 mg/kg) and brain samples collected 1 h later. Two-way ANOVA indicated significant agonist-antagonist interactions for ICI 118551 in brain stem (F1,21 = 19.3, p < 0.001), hypothalamus (F1,21 = 21.5, p < 0.001), and frontal cortex (F1,21 = 12.4, p < 0.001), but nonsignificant interactions for betaxolol in brain stem (F1,17 = 0.906, p = 0.36), hypothalamus (F1,18 = 2.87, p = 0.11), and frontal cortex (F1,18 = 0.823, p = 0.38). *, significantly different from saline controls (p < 0.05; **, p < 0.01). , significantly different from saline-clenbuterol (p < 0.01).

Interactions between Subtype-Selective -Adrenoceptor Antagonists and the ₁-Adrenoceptor-Selective Agonist Dobutamine. Mice were treated with the ₁-selective antagonist, atenolol (1 mg/kg, Fig. 4A), or the ₂-selective antagonist, ICI 118551 (0.5 mg/kg, Fig. 4B), 20 min before 10 mg/kg dobutamine, a ₁-selective agonist. Dobutamine significantly increased tryptophan in all three brain regions. ICI 118551 attenuated the dobutamine-induced increases in brain tryptophan, whereas atenolol did not. No interaction between dobutamine and atenolol was detected by two-way ANOVA. ICI 118551 treatment significantly attenuated the dobutamine-induced increased increase in brain stem tryptophan (p < 0.01), whereas significance was approached in hypothalamus (p = 0.09) and frontal cortex (p = 0.08).

View larger version (44K): [in this window] [in a new window]   Fig. 4.   Effect of subtype-selective adrenoceptor antagonists on the increases in brain tryptophan induced by dobutamine. Mice were injected with: A, the 1-selective antagonist atenolol (1 mg/kg); or B, the 2-selective antagonist ICI 118551 (0.5 mg/kg) 20 min before the 1-selective agonist dobutamine (10 mg/kg) or saline (n = 6). Brain samples were collected 1 h later. Two-way ANOVA indicated nonsignificant agonist-antagonist interactions for atenolol in brain stem (F1,20 = 0.127, p = 0.73), hypothalamus (F1,20 = 0.236, p = 0.63), and frontal cortex (F1,20 = 0.286, p = 0.60), but a significant interaction for ICI 118551 in brain stem (F1,18 = 9.16, p < 0.01), and interactions that approached significance in hypothalamus (F1,18 = 3.32, p = 0.09) and frontal cortex (F1,17 = 3.56, p = 0.08). *, significantly different from saline controls (p < 0.05; **, p < 0.01). , significantly different from saline-dobutamine (p < 0.01).

Interactions between Nonselective -Adrenoceptor Antagonists and the ₃-Adrenoceptor-Selective Agonist BRL 37344. Propranolol (2.5 mg/kg) (Fig. 5A), a _1/2-selective antagonist, did not alter the response to the ₃-selective agonist BRL 37344 (0.2 mg/kg). However, bupranolol (10 mg/kg, Fig. 5B), a _1/2/3-antagonist, attenuated the effects of BRL 37344 on brain tryptophan in hypothalamus (p < 0.01) and frontal cortex (p < 0.05), but not in brain stem (p = 0.27).

View larger version (41K): [in this window] [in a new window]   Fig. 5.   Effects of -adrenoceptor antagonists on BRL 37344-induced increases in brain tryptophan. Mice (n = 6) were injected with the 1/2-selective propranolol (2.5 mg/kg) (A); or the 1/2/3-antagonist bupranolol (B) 20 min before the 3-agonist BRL 37344 (0.2 mg/kg) and brain samples collected 1 h later. Tryptophan concentrations were determined in brain stem, hypothalamus, and frontal cortex. Two-way ANOVA indicated nonsignificant agonist-antagonist interactions for propranolol in brain stem (F1,30 = 0.010, p = 0.99), hypothalamus (F1,30 = 0.167, p = 0.85), and frontal cortex (F1,30 = 0.792, p = 0.46), and for bupranolol in brain stem (F1,21 = 1.27, p = 0.27), but significant interactions for bupranolol in hypothalamus (F1,21 = 10.5, p < 0.01) and frontal cortex (F1,21 = 5.33, p < 0.05). *, significantly different from saline controls (p < 0.05; **, p < 0.01). , significantly different from saline-BRL 37344 (p < 0.05; , p < 0.01).

Interactions between Subtype-Selective -Adrenoceptor Antagonists and BRL 37344. Mice were treated with the ₃-selective antagonist SR 59230A (1.5, 2.5, 5, or 20 mg/kg) (5 mg/kg shown, Fig. 6A) or ICI 118551 (0.5 mg/kg, Fig. 6B) 20 min before BRL 37344 (0.2 mg/kg) and brain samples collected 1 h later. Neither antagonist prevented the BRL 37344-induced increases in brain tryptophan. No interaction between SR 59230A and BRL 37344 was detected by two-way ANOVA. It was considered that BRL 37344 might stimulate ₂-receptors. However, there was no interaction between BRL 37344 and ICI 118551 in brain stem, hypothalamus, or frontal cortex, suggesting that BRL 37344 does not stimulate ₂-receptors to increase brain tryptophan.

View larger version (41K): [in this window] [in a new window]   Fig. 6.   Effects of subtype-selective -adrenoceptor antagonists on BRL 37344-induced increase in brain tryptophan. Mice (n = 6) were injected with the 3-selective antagonist SR 59230A (5 mg/kg) (A); or the 2-selective antagonist ICI 118551 (0.5 mg/kg) (B) 20 min before the 3-agonist BRL 37344 (0.2 mg/kg) and brain samples collected 1 h later. Tryptophan concentrations were determined in brain stem and hypothalamus in A, and brain stem, hypothalamus, and frontal cortex in B. Two-way ANOVA indicated nonsignificant agonist-antagonist interactions for SR 59230A in brain stem (F1,20 = 0.435, p = 0.52) and hypothalamus (F1,20 = 0.774, p = 0.39), and for ICI 118551 in brain stem (F1,20 = 0.057, p = 0.81), hypothalamus (F1,21 = 0.106, p = 0.75), and frontal cortex (F1,21 = 0.092, p = 0.77). **, significantly different from saline controls (p < 0.01).

₃-Adrenergic Receptor-Deficient Mice. To confirm the involvement of ₃-adrenergic receptors, we tested the effects of the ₃-selective agonist, CL 316243 (0.1 mg/kg), in ₃-adrenergic receptor knockout (₃AR KO mice) (Fig. 7A). Because the knockouts were based on an FVB background, FVB mice were used as age-matched controls. Mice were treated with CL 316243 1 h before brain samples were collected. CL 316243 significantly increased brain tryptophan in the wild-type mice, but had no effect in the ₃-receptor knockout mice. Two-way ANOVA indicated an interaction between agonist and genotype in brain stem, hypothalamus, and frontal cortex (p < 0.001). To determine whether ₃-receptor knockout mice would respond to clenbuterol, the mice were treated with clenbuterol (0.1 mg/kg) 1 h before brain and blood samples were collected (Fig. 7B). The knockout mice exhibited a normal clenbuterol-induced increase in tryptophan, indicating that the knockout mice could show a normal tryptophan response, and suggesting that clenbuterol did not exert its actions via ₃-adrenoceptors. Two-way ANOVA detected no interaction between genotype and agonist.

View larger version (38K): [in this window] [in a new window]   Fig. 7.   Effect of subtype-selective -adrenoceptor agonists on brain tryptophan in wild-type and 3-receptor homozygous knockout mice. Mice (n = 6) were injected with CL 316243 (0.1 mg/kg) (A); or clenbuterol (0.1 mg/kg) (B) and brain samples collected 1 h later. Tryptophan concentrations were determined in brain stem, hypothalamus, and frontal cortex. Two-way ANOVA indicated significant agonist-genotype interactions for CL 316243 in brain stem (F1,23 = 40.6, p < 0.001), hypothalamus (F1,23 = 44.6, p < 0.001), and frontal cortex (F1,23 = 27.0, p < 0.001), but nonsignificant interactions for clenbuterol in brain stem (F1,14 = 0.104, p = 0.75), hypothalamus (F1,14 = 0.380, p = 0.55), and frontal cortex (F1,14 = 1.06, p = 0.33). *, significantly different from saline controls (p < 0.05; **, p < 0.01).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The present study extends earlier findings that activation of -adrenergic receptors induced robust increases in brain tryptophan (Eriksson and Carlsson, 1988; Edwards et al., 1989; Takao et al., 1992). Consistent with data from Edwards et al. (1989) using Sprague-Dawley rats, we observed that the activation of the ₂-, but not the ₁-subtype of adrenergic receptors increased brain tryptophan in male CD-1 mice. However, we also observed that activation of ₃-adrenergic receptors induced similar but more robust increases in brain tryptophan.

Although the nonselective antagonist propranolol is lipophilic and readily crosses the blood-brain barrier, the hydrophilic antagonist nadolol has only limited access to the central nervous system (Schiff and Saxey, 1984). This suggests that the effect of clenbuterol was exerted peripherally. Pretreatment with the ₂-selective antagonist ICI 118551 (0.5 mg/kg), but not the ₁-selective antagonist betaxolol (1 mg/kg), attenuated the clenbuterol-induced increases in brain tryptophan in mice sacrificed 60 min after clenbuterol. ICI 118551 and betaxolol were used at doses considered to be sufficient to block -receptors, but still selective for the intended receptor subtype (Crissman et al., 2001). The elevation of brain tryptophan induced by the moderately selective ₁-agonist dobutamine was probably caused by activation of ₂-adrenergic receptors, because the ₂-selective antagonist ICI 118551 attenuated the increase. The ₁-selective antagonist atenolol, again, at a dose selective for the ₁-adrenoceptor subtype (Ben-Eliyahu et al., 2000; Zhang et al., 2000), had no effect on the dobutamine-induced increases in brain tryptophan, providing stronger evidence that ₁-adrenergic receptors were not involved. Indeed, although binding studies indicate that dobutamine is modestly ₁-selective (Williams and Bishop, 1981), it is reported to activate ₁-, ₂-, and ₁-adrenoceptors at doses used clinically (Ruffolo, 1987). These data suggest that ₂-, but not ₁, -adrenergic receptors can increase brain tryptophan.

Whereas bupranolol, a _1/2/3-receptor antagonist, attenuated the effects of the ₃-agonist, BRL 37344, propranolol had no effect at a dose that blocks _1/2-receptors (Yang and Dunn, 1990). This suggests that ₃-receptors can mediate elevations of brain tryptophan independent of _1/2-receptors. The attenuation by bupranolol was not complete, but this may be because bupranolol has a short half-life (Coruzzi and Bertaccini, 1997). Because CL 316243, a ₃-receptor agonist 10,000 times more selective for ₃-receptors than ₂-receptors and having little, if any, activity at ₁-receptors (Bloom et al., 1992), increased brain tryptophan provides further support for the role of ₃-adrenergic receptors. Moreover, the lack of effect of CL 316243 in ₃-AR KO mice clearly indicates that ₃-adrenergic receptors can mediate increases in brain tryptophan independent of other -adrenergic subtypes. Thus, the surprising inability of the ₃-selective antagonist, SR 59230A, to attenuate the effects of either BRL 37344 or CL 316243, may not be ascribable to agonist effects on _1/2-adrenergic receptors. One explanation is a strain difference (FVB versus CD-1) in the response to SR 59230A, although this cannot be confirmed because the administration of SR 59230A to FVB mice has not been reported in the literature. It is also possible that the doses of SR 59230A tested were not sufficient to block ₃-receptors, the route of administration (i.p. injection) may not have been optimal for this antagonist, or the drug is not as effective in male CD-1 mice as it is in rats (Manara et al., 1996). Indeed, in previous studies conducted to demonstrate the ₃-receptor selectivity of SR 59230A, it was given by gavage to rats (Manara et al., 1996). Very few studies have used SR 59230A in vivo, perhaps because binding to plastics and proteins such as bovine serum albumin must be considered when using SR 59230A (Nisoli and Carruba, 1997). However, propranolol administered i.p. with SR 59230A at doses similar to those used in these studies reportedly blocked all three -adrenoceptor subtypes in C57BL/6 mice (Evans et al., 1999). Taken together, the above results suggest that stimulation of ₃-receptors, like ₂-receptors, can increase brain tryptophan.

Activation of ₃-adrenergic receptors increases energy expenditure in brown adipose tissue and lipolysis in white adipose tissue, making ₃-receptors potential anti-obesity targets (Arch et al., 1984; Bloom et al., 1992; Sakura et al., 2002). The present study indicates another effect of -adrenergic stimulation: increased brain tryptophan leading to elevations in 5-HT metabolism. Because 5-HT appears to be involved in appetite regulation, increased brain tryptophan could be seen as a useful side effect of the ₃-receptor agonists which have been proposed for the treatment of obesity (Arch, 2002). Indeed, acute administration of CL 316243 in mice decreases food intake by 45% (Susulic et al., 1995) by an as yet unknown mechanism, but which may involve central serotonergic systems.

At least three explanations for the mechanism of the increase in brain tryptophan have been proposed. First, -adrenergic receptor activation may induce lipolysis, releasing free fatty acids that displace bound tryptophan from albumin in the bloodstream, leading to enhanced transport of free tryptophan into the brain (Wurtman and Fernstrom, 1976; Curzon, 1979). Whether plasma-total or -free tryptophan is most critical for brain tryptophan uptake remains controversial, but there is evidence that plasma-free tryptophan is more important (Curzon, 1979). Indeed, preliminary studies in our laboratory indicate that ₃-agonists increase free fatty acids and plasma-free tryptophan, whereas ₂-selective and nonselective agonists do not (N. Lenard and A. Dunn, unpublished observations), which may explain the seemingly more robust increases in brain tryptophan induced by ₃-agonists. Second, the administration of -adrenergic agonists may stimulate insulin release (Atef et al., 1996; Yajima et al., 1999), which would enhance muscle uptake of branched-chain amino acids and reduce competition of tryptophan for the common brain neutral amino acid transporter (Fernstrom and Wurtman, 1972; Fernstrom, 1976). It has been shown that clenbuterol administration results in robust increases in both plasma glucose and insulin (Edwards and Virji, 1990). Studies are currently being conducted in our laboratory to further address these questions. Third, it has been speculated that brain endothelial cell -adrenergic receptors in some way regulate the transport of amino acids into the brain (Edwards et al., 1989; Takao et al., 1992). Preliminary evidence obtained in our laboratory by measuring the uptake of radiolabeled tryptophan into immortalized human brain endothelial cells (Muruganandam et al., 1997) and the transport of radiolabeled tryptophan into the brain using an in vivo perfusion model (Banks et al., 2000) did not support this hypothesis, but further experimentation will be necessary to exclude this possibility.

In summary, evidence is presented that activation of either ₂- or ₃-adrenergic receptors can increase mouse brain tryptophan by as yet unknown mechanisms. Moreover, an increase in 5-HT metabolism, suggesting an alteration of central serotonergic systems, may be secondary to this increase in brain tryptophan. Because 5-HT is involved in mood and appetite regulation, among numerous other functions (Rueter et al., 1997; Jacobs and Fornal, 1999), these results indicate the need for further study of the -adrenergic influence on the brain concentration of its precursor, tryptophan. Thus, the present study may have important implications for the benefits and/or side effects of ₂- and ₃-adrenergic agonists.