- Email: [email protected]
Consumption of a high dietary dose of monosodium glutamate fails to affect extracellular glutamate levels in the hypothalamic arcuate nucleus of adult rats
BRAIN RESEARCH ELSEVIER
Brain Research 736 (1996) 76 81
Research report
Consumption of a high dietary dose of monosodium glutamate fails to affect extracellular glutamate levels in the hypothalamic arcuate nucleus of adult rats Mikhail B. Bogdanov, Olga A. Tjurmina, Richard J. Wurtman * Department of Brain and Cognitit,e Sciences, E25-604, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, MA 02139, USA
Accepted 4 June 1996
Abstract We examined the effects of systemic or oral ad libitum monosodium glutamate (MSG) administration on glutamate levels in plasma, and on glutamate release from the arcuate nucleus of the hypothalamus (estimated using brain microdialysis). Systemic MSG administration (0.25, 0.5, 1 or 2 g / k g , i.p.) to adult rats caused dose-dependent increases in glutamate levels within arcuate nucleus dialysates. These levels increased during the initial 20 rain after systemic MSG administration, and peaked during the second 20-min interval (maximally to 116 _+ 7%, 146 + 15%, 790 _+ 191% and 1230 _+ 676% of basal values, respectively). Plasma glutamate levels, measured simultaneously, were increased maximally during the initial 20 min after MSG administration. These increases were 10-, 13-, 76- and 163-fold after doses of 0.25, 0.5, 1 and 2 g / k g , i.p., respectively. In feeding experiments, consumption of 2.3 g / k g of MSG by previously-trained rats during an 1-h period increased plasma glutamate levels to 352 _+ 61% of basal values 140 rain after the start of the feeding period. No changes were observed in glutamate levels of arcuate nucleus dialysates. These findings may explain why ad libitum dietary consumption of MSG apparently lacks neurotoxic potential. Keywords: Glutamate: Arcuate nucleus of hypothalamus; Microdialysis; Extracellular level; Plasma level; Feeding
1. Introduction
Subcutaneous administration of MSG can cause neuronal degeneration within the inner retinal layer in the neonatal mice [11] and can also produce neurodegenerative changes in the arcuate nucleus of the rodent's hypothalamus [17,18,21]. The ability of MSG to induce such changes depends upon its route of administration: neuronal lesions have never been demonstrated after voluntary ad libitum consumption of very high MSG doses [7,8]. The most probable explanation for this lack of effect is the difference between the effects of systemic vs dietary MSG on plasma and then brain glutamate levels. Numerous processes might be expected to affect the extracellular brain glutamate concentrations after MSG consumption, including the metabolism of glutamate in the liver and gastrointestinal tract, its active efflux from the brain by blood-brain trans-
* Corresponding author.
port systems, and the activity of neuronal and/or glial glutamate uptake systems [6,9,16,19]. We recently found that systemic administration of high MSG doses induced prominent increases in extracellular glutamate levels within the rat's striatum, as measured by in vivo microdialysis [3]. In contrast, a comparably high MSG dose, consumed in the diet, failed to elevate these levels. It is possible that dietary MSG might affect extracellular glutamate levels elsewhere in the brain. The circumventricular organs of the brain have high permeability to blood-derived substances [13,19]. The neurotoxic effects of systemically administered glutamate are known to be particularly prominent in the rat's hypothalamic arcuate nucleus [7]. In this study we compared directly, using brain microdialysis, the effects of systemic or ad libitum dietary MSG administration on in vivo extracellular glutamate levels within the arcuate nucleus of adult rats. Since the key question related to the potential neurotoxicity of exogenous glutamate is whether brain extracellular glutamate concentrations change in response to changes in its plasma concentrations, we monitored plasma glutamate levels to
0006-8993/96/$15.00 Copyright © 1996 Elsevier Science B.V. All rights reserved. Pll S 0 0 0 6 - 8 9 9 3 ( 9 6 ) 0 0 6 7 9 - 8
M.B. Bogdanov et al./ Brain Research 736 (1996) 76-81
determine their relationship to brain extracellular glutamate levels.
2. Materials and methods 2.1. Animals and MSG administration
Male Sprague-Dawley rats (Charles River, MA) were housed two per cage for 1 week before an experiment; food (Prolab Animal Diet 3000, Agway Inc., Syracuse, NY; 22% protein content) and water were available ad libitum. Lights were on from 0800 to 1800 h, and all experiments were performed during the light period. The effects of MSG on glutamate levels in plasma and in arcuate nucleus dialysates were studied following the systemic intraperitoneal administration of MSG or it's elective consumption in the diet. In the first study, rats (250-300 g) were implanted with an arterial catheter and a dialysis probe 18-24 h before the start of perfusion. After collection of basal dialysates and blood samples, MSG was administered i.p. at doses of 0.25, 0.5, 1 or 2 g / k g (dissolved in distilled water, 10 ml/kg), and measurements were continued for an additional 4 h. Control animals received saline (10 m l / k g , i.p.). In the second study, animals were trained for 3 - 4 weeks, in advance of experiments, to consume their total daily food intake during a 1-h period. One week after arrival, rats (200-220 g at the start of the training period) were housed one per cage with water available ad libitum. The rat diet was provided ad libitum for 1 h, from noon to 1300 h. After the last training session, the rats were implanted with the arterial catheter and the dialysis probes, and perfusion experiments were started the next morning. Basal dialysate and blood samples were obtained, and MSG-containing or control diets were provided between noon and 1300 h. Measurements were continued during the 1 h feeding period and for 4 h afterwards. The MSG-based diet contained 25.2% protein, 4.4% fat, 4.4% crude fiber, 7% ash and 6.6% MSG. In the control diet the protein content was 31.8%. Water was available ad libitum throughout the experiment. 2.2. Surgery and experimental procedure
Rats were anaesthetized with ketamine/xylazine ( 8 0 / 1 0 m g / k g , i.p.) and a polyethylene catheter (PE 10 connected to PE 50) was implanted into each animal's abdominal aorta, through the fight femoral artery, for collection of blood samples. The catheter was tunnelled under the skin and exteriorized on the back of the neck. Thereafter animals were mounted onto a stereotaxic frame (David Kopf) with an incisor bar set at - 3 . 3 mm below an interaural line. After the skull was exposed and a burr hole drilled, the probe was lowered into the brain and secured to the
77
skull. Concentric dialysis probes (membrane length 0.5-0.7 mm; outer diameter 250 ~m; MW cut-off 20 kDa; in vitro recovery for glutamate at 2 ixl/min 5 - 7 % ) were used. Coordinates of the probe's tip, using the bregma as the reference point, were: AP-3.6; ML 0.4; V-10.4 [20]. Perfusion experiments were started 18-24 h after surgery. The perfusion medium (145 mM NaC1, 2.7 mM KC1, 1.2 CaCI2, 1.0 mM MgC12, pH 7.4) was delivered at 2 m l / m i n using a microperfusion pump (CMA/100, CMA), and dialysate samples were collected every 20 min. Collection of basal dialysate samples started after at least 3 - 4 h of perfusion. Blood samples (100-120 ixl), taken via the catheter implanted into the abdominal aorta, were collected at the same times as dialysates (i.e. every 20 min). Immediately after collection blood samples were centrifuged at 4000 g for 10 min, and plasma samples were collected and stored at - 2 0 ° C until assayed for glutamate content. After completion of each experiment, animals were deeply anaesthetized with ketamine, and transcardially perfused with saline followed by 4% paraformaldehyde, pH 7.4. Fixed brains were removed and placed into a 30% sucrose/4% paraformaldehyde solution (5/1). After cryostat sectioning and cresyl violet staining, brain sections (40 ixm thickness) were examined to localize probe placement within the hypothalamic arcuate nucleus. Only data from animals with verified probe locations in the arcuate nucleus were included in analyses. 2.3. Glutamate assay
Concentrations of amino acids in dialysates or plasma samples were detected by H P L C / E C after precolumn derivatization with o-phtalaldehyde (OPA), using a modification of method of Donzanti and Yamamoto [5]. Prior to assay, plasma samples were mixed 1:1 with 5% sulfosalicylic acid and incubated at room temperature for 20 min. Mixtures were centrifuged at 10000g for 15 min. Supernatants were filtered through 0.22 txm cellulose acetate filters and analyzed after appropriate dilutions. Dialysate samples were assayed directly, without previous treatment. The derivatization stock reagent contained 27 mg of OPA, 1 ml of 100% methanol, 5 ixl of 2-mercaptoethanol (2-ME) and 9 ml of 0.1 M sodium tetraborate, pH 9.3. The working solution was prepared by diluting the stock solution 1:24 with 0.1 M sodium tetraborate. Derivatization was performed by mixing 10 txl of the sample or standard with 10 ~zl of the working O P A / 2 - M E reagent for 2 min. Ten ill of the mixture were injected onto the column. The stability of the working derivatization solution was tested by injecting standards after every 10 samples. The HPLC system consisted of a dual piston pump (L-6000, Hitachi); an ESA 465 autosampler (ESA, Chelmsford, MA); a 3-~m C18 ODS 80 × 4.6 mm column (HR-80, ESA), and an ESA 5200A coulometric detector with an ESA 5014 dualelectrode analytical cell. The first electrode was set at + 200 mV, and the second at + 400 mV. The potential of
78
M.B. Bogdanou et al. / Brain Research 736 (1996) 76 81
the guard cell (model 5020, ESA) placed between the pump and the injection valve was set at + 4 0 0 inV. The mobile phase, delivered at 1.2 m l / m i n , was 0.! M sodium dibasic phosphate buffer, 25% ( v / v ) methanol and 5% ( v / v ) acetonitrile, pH 6.4 with phosphoric acid.
els were found in the second 20-min sample. After 0.25, 05, 1, and 2 g / k g of MSG, glutamate levels in the first sample were increased to 115 _+ 7%, 145 _ 13%, 689 _+ 190%, and 748 _+ 387% of basal levels, respectively. Maximal increases in glutamate levels were 116 _+ 7%, 146 _+ 15%, 7 9 0 + 191%, and 1 2 3 0 + 6 7 6 % of basal values, respectively. After 0.25 and 0.5 g / k g doses, glutamate levels returned to basal values within 40 and 60 min of MSG administration, respectively. The increases induced by 1 and 2 g / k g MSG doses lasted for 2 and 3.5 h, respectively. Systemic administration of MSG caused prominent increases in plasma glutamate levels (Fig. 2). In contrast to the effects of MSG on glutamate levels in the arcuate nucleus, maximal increases in plasma were observed already in the first sample collected 20 min following MSG administration. At 0.25, 0.5, 1 and 2 g / k g , i.p., MSG induced maximal increases in plasma glutamate levels to 992_+ 64%, 1 3 1 0 _ 121%, 7580___ 1300%, and 16,300 ± 1990% of basal values, respectively. These levels gradually declined, dose-dependently, over time.
2.4. Presentation of data The concentrations of glutamate in dialysate or plasma samples were expressed as percentages of the mean values of three consecutive samples obtained prior to systemic or dietary MSG administration (basal levels). Data are given as means ± S.E.M.. M a n n - W h i t n e y U-test was applied to the analysis.
3. Results
3.1. Effects of systemic MSG administration on glutamate levels in plasma and in dialysates,from the arcuate nucleus In control rats, basal glutamate levels in 20-min dialysate samples from the arcuate nucleus were 0.17 ___0.011 b~M (n = 9). These usually stabilized after 3 - 4 h of perfusion, and remained stable for the next 5 h. Plasma glutamate levels in these animals were 254_+ 10.7 p~M. Intraperitoneal administration of saline did not modify glutamate levels in plasmas or dialysates for at least 4 h after the injection (data not shown). All systemic MSG doses tested in this study induced significant dose-dependent increases in glutamate levels within arcuate nucleus dialysates (Fig. 1). Significant increases were observed in the first 20-min sample after MSG administration. Maximal increases in glutamate lev-
2000
3.2. Effects of dietao, MSG administration on glutamate leL,els in plasma and dialysates from the arcuate nucleus The amounts of food consumed during the 1 h feeding period of the experiment were 7.24 4- 0.97 g in rats provided with the control diet (n = 6), and 7.88 4-0.76, in those given the MSG-based diet (n = 8). After correction for body weights, the oral dose of MSG consumed by the animals during 1 h was calculated to have been 2.33 _+ 0.37 g/kg. In rats consuming the control diet no significant changes were observed in glutamate levels within plasmas or
*
1 oo 1600 -
.
1400 •
~
1200
.
MSG
doses,
1000
g/kg i p: --~-- 0.25
1/ 600
.
400
*
', T . *
o.5
,
*
*
200
0
2
3
4
5
6 Time,
7
8
20-rain
9
10
11
12
13
14
15
intervals
Fig. 1. Effects of systemic MSG administration (0.25, 0.5, 1 or 2 g / k g , i.p.) on extracellular glutamate levels in dialysates of the hypothalamic arcuate
nucleus in freely moving rats. Injection of MSG is indicated by the arrow. Data are presented as means +_S.E.M. of percent variations from mean basal glutamate levels (three consecutive samples in which glutamate levels varied by no more than 15%). Each time point represents data from 6 to 8 animals. • P < 0.05, compared at the correspondingtime points with the saline-injected control group; two-tailed Mann-Whitney U-test.
79
M.B. Bogdanov et a l . / Brain Research 736 (1996) 76-81 20000 18000 16000. 14O0O
~ 12000 e~
M S G d o s e s , g / k g i.p.:
"6 10000
~--I
\*
~
0.25
\ \.
8000
6000 4000 2000O~ 2
3
4
5
6
7
8
9
10
11
12
13
14
15
Time, 2 0 - m i n intervals
Fig. 2. Effects of systemic MSG administration (0.25, 0.5, 1 or 2 g/kg, i.p.) on plasma glutamate levels in freely moving rats. Injection of MSG is indicated by the arrow. Data are presented as means _+ S.E.M. of percent variations from mean basal plasma glutamate levels (three consecutive samples, corresponding to the basal dialysate samples). Each time point represents data from 6 to 8 animals. * P < 0.05, compared at the corresponding time points with saline-injected controls; two-tailed Mann-Whitney U-test.
dialysates f r o m arcuate nucleus, either during the 1 h f e e d i n g period or within the subsequent 4 h (Figs. 3 and
period), but still r e m a i n e d significantly elevated (up to 135 + 11% o f basal levels).
4). C o n s u m p t i o n o f the M S G - c o n t a i n i n g diet failed to affect extracellular g l u t a m a t e levels in the arcuate nucleus (Fig. 3). P l a s m a glutamate levels, m e a s u r e d simultaneously, w e r e significantly e l e v a t e d (Fig. 4). T h e s e levels w e r e increased (up to 152 _+ 12% o f basal values, P < 0.05) after 20 min, and increased further to a peak o f 352 _+ 61% o f basal values 140 min after the start o f the f e e d i n g period. P l a s m a g l u t a m a t e levels tended to decline by the end o f the e x p e r i m e n t (5 h after the start o f f e e d i n g
4. Discussion T h e s e data show that a high single dietary dose o f M S G , w h i c h significantly elevates p l a s m a glutamate levels, fails to affect extracellular glutamate levels in the rat's arcuate nucleus. U s i n g p r e v i o u s l y trained animals, it was possible to p r o v i d e the animals with as m u c h as 2.3 g / k g o f oral M S G during the 1-h period o f f o o d consumption.
1281 ] I _
DIET: 15
" - * - control
=, (.9
•"-e-- M S G
FEEDING 0 2
3
4
5
6
7
8
9
10
11
12
13
t4
15
16
17
18
Time, 20-min intervals Fig. 3. Effects of ad libitum consumption of control or MSG-based diet on extracellular glutamate levels in dialysates of the hypothalamic arcuate nucleus in freely moving rats. Control or MSG-based diets were present over the 1-h period during the experiment, as indicated by the bar. The amounts of food consumed in the 1 h feeding period were 7.24 + 0.97 g among animals provided with the control diet (n = 6), and 7.88 ± 0.76 g among those provided with the MSG-based diet (n = 8). The MSG dose provided over the 1-h period of feeding was 2.33 + 0.37 g/kg. Data are presented as means + S.E.M. of percent variations from mean basal glutamate levels (three consecutive samples before start of the 1 h feeding period).
M.B. Bogdanou et al. / Brain Research 736 (1996) 76-81
80 450
400
35O
3OO
-~
DIET:
.D
250 -
.
200 •
"6
~
control
~
MSG
3 150
100
FEEDING 5O
0
1. 2
3
4
s
6
t
8
o
lo
11
12
13
14
15
16
1T
18
Time, 20-rain intervals
Fig. 4. Effects of ad libitum consumption of control or MSG-based diet on the plasma glutamate levels in freely moving rats. Control or MSG-based diets were present over the 1-h period during the experiment, as indicated by the bar. The amounts of food consumed in the 1 h feeding period were 7.24 _+ 0.97 g among animals provided with the control diet (n = 6), and 7.88 _+ 0.76 g among those provided with the MSG-based diet (n = 8). The MSG dose provided over the 1-h period of feeding was 2.33 _+ 0.37 g/kg. Data are presented as means _+ S.E.M. of percent variations from mean basal glutamate levels (three consecutive samples before start of the 1 h feeding period). * P < 0.05, compared at the corresponding time points with the data obtained in animals provided with the control diet; two-tailed Mann-Whitney U-test.
The magnitude of the peak increase in plasma glutamate levels in animals that consumed this MSG-based diet, up to 3.5 times basal levels, was in the range reported previously by other authors [1,2,4,10,15]. In contrast to dietary MSG, systemic MSG administration did increase glutamate levels within arcuate nucleus dialysates and in plasma, and the increases within the arcuate nucleus were comparable to those we described previously in the rat's striatum, a brain region within the blood-brain barrier [3]. The most obvious explanation for the observed differences between the effects of systemic and dietary MSG on extracellular glutamate levels in the arcuate nucleus is the pattern and magnitude of the increases in plasma glutamate levels that follow these two routes of administration. The lowest systemic MSG dose tested in this study (0.25 g / k g , i.p.) induced a 10-fold increase in plasma glutamate levels. Extracellular glutamate levels in the arcuate nucleus increased significantly, but by only 16% above basal values. In the feeding experiments, where no overall effect was observed on the arcuate dialysate glutamate levels, a much smaller (to 3.5-fold basal levels) increase in plasma glutamate levels was observed; this apparently was insufficient to affect the arcuate nucleus. It seems unlikely that arcuate glutamate levels would have risen significantly at a latter time point, since plasma glutamate levels in the feeding studies were maximally increased 140 rain after the start of feeding and remained elevated through the end of observation period. The mechanisms that determine the magnitude of the rise in arcuate nucleus glutamate levels after systemic MSG administration are not clear. The time delay between the maximal increases in plasma and in arcuate levels, and approximately the 10-fold differences between the maxi-
mal increases observed after all systemic MSG doses tested, suggest the operation of mechanisms which efficiently buffer changes in arcuate nucleus extracellular glutamate levels. These could include a transport system in brain capillary endothelial cells that actively extrudes glutamate from the brain extracellular fluid to the blood and glial a n d / o r neuronal glutamate uptake systems [6,9,16,19]. The increase in plasma osmolarity that follows systemic MSG administration [12,22] probably increases the permeability of the BBB at the level of the arcuate nucleus. In the presence of increased plasma glutamate levels this allows more glutamate to enter this nucleus. In accordance with this hypothesis, it has recently been demonstrated that opening of the BBB by an intracarotid protamine infusion enhances effects of intracarotid glutamate on extracellular glutamate levels in parietal cortex [23]. Monno et al. recently reported that 4 g / k g of MSG given to adult rats by forced gavage induced 4.2- and 8.9-fold increases in extracellular glutamate levels within hippocampus and hypothalamus respectively; plasma glutamate levels were increased only 5.3-fold [14]. It is difficult to explain how the increase in extracellular brain glutamate levels could exceed the concurrent (40 min after MSG administration) increase in plasma glutamate levels. Basal plasma glutamate levels reported in the study of Monno et al. (261 ~M) were about the same as those found in this study (254 puM). In our experiments, we found that increases in plasma glutamate levels are generally about 10 times higher than those in arcuate extracellular glutamate levels, following all systemic MSG doses tested. The neurotoxicity of exogenous glutamate, when administered systemically to rodents at very high doses has
M.B. Bogdanov et al. / Brain Research 736 (1996) 76-81
been well documented by numerous investigators, while neurotoxic effects of MSG ingested as a dietary constituent have never been described. The mechanism of MSG's neurotoxic effects has been attributed to a sustained increase in extracellular brain glutamate concentrations [18]. Our present findings strongly suggest that different changes in plasma glutamate levels following systemic or dietary MSG may differentially affect brain glutamate levels and may in turn underlie the apparent lack of neurotoxic potential of MSG following its dietary consumption.
Acknowledgements Expert technical assistance of Jeffrey Breu is greatly acknowledged. This study was supported in part by a grant from the Center for Brain Sciences and Metabolism Chafftable Trust.
References [1] Airoldi, L., Bizzi, A., Salmona, M. and Garratini, S., Attempts to establish the safety margin for neurotoxicity of monosodium glutamate. In L.J. Filer, S. Garratini, M.R. Kare, W.A. Reynolds and R.J. Wurtman (Eds.), Glutamic Acid: Advances in Biochemistry and Physiology, Raven Press, New York, 1979, pp. 321-331. [2] Bizzi, A., Veneroni, E., Salmona M. and Garratini, S., Kinetics on monosodium glutamate in relation to its neurotoxicity, Toxicol. Lett., 1 (1977) 123-130. [3] Bogdanov, M.B. and Wurtman, R.J., Effects of systemic or oral ad libitum monosodium glutamate administration on striatal glutamate release, as measured using microdialysis in freely moving rats, Brain Res., 660 (1994) 337-340. [4] Caccia, S., Garattini, S., Ghezzi, P. and Zanini, M.G., Plasma and brain levels of glutamate and pyroglutamate after oral monosodium glutamate to rats, Toxicol. Lett., 10 (1982) 169-175. [5] Donzanti, B.A. and Yamamoto, B.K., An improved and rapid HPLC-EC metod for the isocratic separation of amino acid neurotransmitters from brain tissue and microdialysis perfusates, Life Sci., 43 (1988) 913-922. [6] Fonnum, F., Glutamate: a neurotransmitter in mammalian brain, J. Neurochem., 42 (1984) 1-11. [7] Garattini, S., Evaluation of the neurotoxic effects of glutamic acid, In R.J. Wurtman and J.J. Wurtman (Eds.), Nutrition and the Brain, Raven Press, New York, 1979, pp. 79-124.
81
[8] Heywood, R., James, R.W. and Worden, A.N., The ad libitum feeding of monosodium glutamate to wealning mice, Toxicol. Lett., 1 (1977) 151-155. [9] Hutchison, H.T., Eisenberg, H.M. and Haber, B., High-affinity transport of glutamate in rat brain microvessels, Exp. Neurol., 87 (1985) 260-269. [10] Liebshutz, J., Airoldi, L., Brownstein, M.J., Chinn, N.G. and Wurtman, R.J., Regional distribution of endogenous and parenteral glutamate, aspartate and glutamine in rat brain, Biochem. Pharmacol., 26 (1977) 443-446. [11] Lucas, D.R. and Newhouse, J.P., The toxic effect of sodium Lglutamate on the inner layers of the retina, AMA Arch. Opthalmol., 58 (1957) 193-201. [12] McCall, A., Glaeser, B.S., Millington, W. and Wurtman, R.J., Monosodium glutamate neurotoxicity, hyperosmolarity and bloodbarrier disfunction, Neurobehav. Toxicol., 1 (1979)279-283. [13] Merchenthaler, I., Neurons with access to the general circulation in the central nervous system of the rat: a retrograde tracing study with fluoro-gold, Neuroscience, 44 (1991) 655-662. [14] Monno, A., Vezzani, A., Bastine, A., Salmona, M. and Garattini, S., Extracellular glutamate levels in the hypothalamus and hippocampus of rats after acute or chronic oral intake of rnonosodium glutamate, Neurosci. Lett., 193 (1995) 45-48. [15] O'Hara Y. and Takasaki, Y., Pelatiobship between plasma glutamate levels and hypothalamic lesions in rodents, Toxicol. Lett., 4 (1979) 499-505. [16] Oldendorf, W. and Szabo, J., Amino acid assignment to one of three blood-brain barrier amino acid carriers, Am. Z Physiol., 230 (1976) 94-98. [17] Olney, J.W., Brain lesions, obesity and other disturbances in mice treated with monosodium glutamate, Science, 164 (1969) 719-721. [18] Olney, J.W., Excitotoxic amino acids: Research applications and safety implications, In L.J. Filer, S. Garratini, M.R. Kare, W.A. Reynolds and R.J. Wurtman (Eds.), Glutamic Acid: Advances in Biochemistry and Physiology, Raven Press, New York, 1979, pp. 287-319. [19] Pardridge, W.M., Regulation of amino acids availability to brain: Selective control mechanisms for glutamate, In L.J. Filer, S. Garratini, M.R. Kare, W.A. Reynolds and R.J. Wurtman (Eds.), Glutamic Acid: Advances in Biochemistry and Physiology, Raven Press, New York, 1979, pp. 125-138. [20] Paxinos, G. and Watson, C., The Rat Brain in Stereotaxic Coordinates, Academic Press, New York, 1986. [21] Peng, Y., Gubin, J., Harper, A.E., Vavich, M.G. and Kemmer, A.R., Food intake regulation: Amino acid toxicity and changes in rat brain and plasma amino acids, J. Nutr., 103 (1973) 608-617. [22] Torii, K., Takasaki, Y., Iwata, S. and Wurtman, R.J., Changes in blood osmolarity, electrolytes and metabolites among adult rats treated with a neurotoxic dose of MSG, Life Sci., 28 (1981) 2855-2864. [23] Westergen, I., Nystrom, B., Hamberger, A. and Johansson, B., Intracerebrai dialysis and blood-brain barrier, J. Neurochem., 64 (1995) 229-234.
Brain Research 736 (1996) 76 81
Research report
Consumption of a high dietary dose of monosodium glutamate fails to affect extracellular glutamate levels in the hypothalamic arcuate nucleus of adult rats Mikhail B. Bogdanov, Olga A. Tjurmina, Richard J. Wurtman * Department of Brain and Cognitit,e Sciences, E25-604, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, MA 02139, USA
Accepted 4 June 1996
Abstract We examined the effects of systemic or oral ad libitum monosodium glutamate (MSG) administration on glutamate levels in plasma, and on glutamate release from the arcuate nucleus of the hypothalamus (estimated using brain microdialysis). Systemic MSG administration (0.25, 0.5, 1 or 2 g / k g , i.p.) to adult rats caused dose-dependent increases in glutamate levels within arcuate nucleus dialysates. These levels increased during the initial 20 rain after systemic MSG administration, and peaked during the second 20-min interval (maximally to 116 _+ 7%, 146 + 15%, 790 _+ 191% and 1230 _+ 676% of basal values, respectively). Plasma glutamate levels, measured simultaneously, were increased maximally during the initial 20 min after MSG administration. These increases were 10-, 13-, 76- and 163-fold after doses of 0.25, 0.5, 1 and 2 g / k g , i.p., respectively. In feeding experiments, consumption of 2.3 g / k g of MSG by previously-trained rats during an 1-h period increased plasma glutamate levels to 352 _+ 61% of basal values 140 rain after the start of the feeding period. No changes were observed in glutamate levels of arcuate nucleus dialysates. These findings may explain why ad libitum dietary consumption of MSG apparently lacks neurotoxic potential. Keywords: Glutamate: Arcuate nucleus of hypothalamus; Microdialysis; Extracellular level; Plasma level; Feeding
1. Introduction
Subcutaneous administration of MSG can cause neuronal degeneration within the inner retinal layer in the neonatal mice [11] and can also produce neurodegenerative changes in the arcuate nucleus of the rodent's hypothalamus [17,18,21]. The ability of MSG to induce such changes depends upon its route of administration: neuronal lesions have never been demonstrated after voluntary ad libitum consumption of very high MSG doses [7,8]. The most probable explanation for this lack of effect is the difference between the effects of systemic vs dietary MSG on plasma and then brain glutamate levels. Numerous processes might be expected to affect the extracellular brain glutamate concentrations after MSG consumption, including the metabolism of glutamate in the liver and gastrointestinal tract, its active efflux from the brain by blood-brain trans-
* Corresponding author.
port systems, and the activity of neuronal and/or glial glutamate uptake systems [6,9,16,19]. We recently found that systemic administration of high MSG doses induced prominent increases in extracellular glutamate levels within the rat's striatum, as measured by in vivo microdialysis [3]. In contrast, a comparably high MSG dose, consumed in the diet, failed to elevate these levels. It is possible that dietary MSG might affect extracellular glutamate levels elsewhere in the brain. The circumventricular organs of the brain have high permeability to blood-derived substances [13,19]. The neurotoxic effects of systemically administered glutamate are known to be particularly prominent in the rat's hypothalamic arcuate nucleus [7]. In this study we compared directly, using brain microdialysis, the effects of systemic or ad libitum dietary MSG administration on in vivo extracellular glutamate levels within the arcuate nucleus of adult rats. Since the key question related to the potential neurotoxicity of exogenous glutamate is whether brain extracellular glutamate concentrations change in response to changes in its plasma concentrations, we monitored plasma glutamate levels to
0006-8993/96/$15.00 Copyright © 1996 Elsevier Science B.V. All rights reserved. Pll S 0 0 0 6 - 8 9 9 3 ( 9 6 ) 0 0 6 7 9 - 8
M.B. Bogdanov et al./ Brain Research 736 (1996) 76-81
determine their relationship to brain extracellular glutamate levels.
2. Materials and methods 2.1. Animals and MSG administration
Male Sprague-Dawley rats (Charles River, MA) were housed two per cage for 1 week before an experiment; food (Prolab Animal Diet 3000, Agway Inc., Syracuse, NY; 22% protein content) and water were available ad libitum. Lights were on from 0800 to 1800 h, and all experiments were performed during the light period. The effects of MSG on glutamate levels in plasma and in arcuate nucleus dialysates were studied following the systemic intraperitoneal administration of MSG or it's elective consumption in the diet. In the first study, rats (250-300 g) were implanted with an arterial catheter and a dialysis probe 18-24 h before the start of perfusion. After collection of basal dialysates and blood samples, MSG was administered i.p. at doses of 0.25, 0.5, 1 or 2 g / k g (dissolved in distilled water, 10 ml/kg), and measurements were continued for an additional 4 h. Control animals received saline (10 m l / k g , i.p.). In the second study, animals were trained for 3 - 4 weeks, in advance of experiments, to consume their total daily food intake during a 1-h period. One week after arrival, rats (200-220 g at the start of the training period) were housed one per cage with water available ad libitum. The rat diet was provided ad libitum for 1 h, from noon to 1300 h. After the last training session, the rats were implanted with the arterial catheter and the dialysis probes, and perfusion experiments were started the next morning. Basal dialysate and blood samples were obtained, and MSG-containing or control diets were provided between noon and 1300 h. Measurements were continued during the 1 h feeding period and for 4 h afterwards. The MSG-based diet contained 25.2% protein, 4.4% fat, 4.4% crude fiber, 7% ash and 6.6% MSG. In the control diet the protein content was 31.8%. Water was available ad libitum throughout the experiment. 2.2. Surgery and experimental procedure
Rats were anaesthetized with ketamine/xylazine ( 8 0 / 1 0 m g / k g , i.p.) and a polyethylene catheter (PE 10 connected to PE 50) was implanted into each animal's abdominal aorta, through the fight femoral artery, for collection of blood samples. The catheter was tunnelled under the skin and exteriorized on the back of the neck. Thereafter animals were mounted onto a stereotaxic frame (David Kopf) with an incisor bar set at - 3 . 3 mm below an interaural line. After the skull was exposed and a burr hole drilled, the probe was lowered into the brain and secured to the
77
skull. Concentric dialysis probes (membrane length 0.5-0.7 mm; outer diameter 250 ~m; MW cut-off 20 kDa; in vitro recovery for glutamate at 2 ixl/min 5 - 7 % ) were used. Coordinates of the probe's tip, using the bregma as the reference point, were: AP-3.6; ML 0.4; V-10.4 [20]. Perfusion experiments were started 18-24 h after surgery. The perfusion medium (145 mM NaC1, 2.7 mM KC1, 1.2 CaCI2, 1.0 mM MgC12, pH 7.4) was delivered at 2 m l / m i n using a microperfusion pump (CMA/100, CMA), and dialysate samples were collected every 20 min. Collection of basal dialysate samples started after at least 3 - 4 h of perfusion. Blood samples (100-120 ixl), taken via the catheter implanted into the abdominal aorta, were collected at the same times as dialysates (i.e. every 20 min). Immediately after collection blood samples were centrifuged at 4000 g for 10 min, and plasma samples were collected and stored at - 2 0 ° C until assayed for glutamate content. After completion of each experiment, animals were deeply anaesthetized with ketamine, and transcardially perfused with saline followed by 4% paraformaldehyde, pH 7.4. Fixed brains were removed and placed into a 30% sucrose/4% paraformaldehyde solution (5/1). After cryostat sectioning and cresyl violet staining, brain sections (40 ixm thickness) were examined to localize probe placement within the hypothalamic arcuate nucleus. Only data from animals with verified probe locations in the arcuate nucleus were included in analyses. 2.3. Glutamate assay
Concentrations of amino acids in dialysates or plasma samples were detected by H P L C / E C after precolumn derivatization with o-phtalaldehyde (OPA), using a modification of method of Donzanti and Yamamoto [5]. Prior to assay, plasma samples were mixed 1:1 with 5% sulfosalicylic acid and incubated at room temperature for 20 min. Mixtures were centrifuged at 10000g for 15 min. Supernatants were filtered through 0.22 txm cellulose acetate filters and analyzed after appropriate dilutions. Dialysate samples were assayed directly, without previous treatment. The derivatization stock reagent contained 27 mg of OPA, 1 ml of 100% methanol, 5 ixl of 2-mercaptoethanol (2-ME) and 9 ml of 0.1 M sodium tetraborate, pH 9.3. The working solution was prepared by diluting the stock solution 1:24 with 0.1 M sodium tetraborate. Derivatization was performed by mixing 10 txl of the sample or standard with 10 ~zl of the working O P A / 2 - M E reagent for 2 min. Ten ill of the mixture were injected onto the column. The stability of the working derivatization solution was tested by injecting standards after every 10 samples. The HPLC system consisted of a dual piston pump (L-6000, Hitachi); an ESA 465 autosampler (ESA, Chelmsford, MA); a 3-~m C18 ODS 80 × 4.6 mm column (HR-80, ESA), and an ESA 5200A coulometric detector with an ESA 5014 dualelectrode analytical cell. The first electrode was set at + 200 mV, and the second at + 400 mV. The potential of
78
M.B. Bogdanou et al. / Brain Research 736 (1996) 76 81
the guard cell (model 5020, ESA) placed between the pump and the injection valve was set at + 4 0 0 inV. The mobile phase, delivered at 1.2 m l / m i n , was 0.! M sodium dibasic phosphate buffer, 25% ( v / v ) methanol and 5% ( v / v ) acetonitrile, pH 6.4 with phosphoric acid.
els were found in the second 20-min sample. After 0.25, 05, 1, and 2 g / k g of MSG, glutamate levels in the first sample were increased to 115 _+ 7%, 145 _ 13%, 689 _+ 190%, and 748 _+ 387% of basal levels, respectively. Maximal increases in glutamate levels were 116 _+ 7%, 146 _+ 15%, 7 9 0 + 191%, and 1 2 3 0 + 6 7 6 % of basal values, respectively. After 0.25 and 0.5 g / k g doses, glutamate levels returned to basal values within 40 and 60 min of MSG administration, respectively. The increases induced by 1 and 2 g / k g MSG doses lasted for 2 and 3.5 h, respectively. Systemic administration of MSG caused prominent increases in plasma glutamate levels (Fig. 2). In contrast to the effects of MSG on glutamate levels in the arcuate nucleus, maximal increases in plasma were observed already in the first sample collected 20 min following MSG administration. At 0.25, 0.5, 1 and 2 g / k g , i.p., MSG induced maximal increases in plasma glutamate levels to 992_+ 64%, 1 3 1 0 _ 121%, 7580___ 1300%, and 16,300 ± 1990% of basal values, respectively. These levels gradually declined, dose-dependently, over time.
2.4. Presentation of data The concentrations of glutamate in dialysate or plasma samples were expressed as percentages of the mean values of three consecutive samples obtained prior to systemic or dietary MSG administration (basal levels). Data are given as means ± S.E.M.. M a n n - W h i t n e y U-test was applied to the analysis.
3. Results
3.1. Effects of systemic MSG administration on glutamate levels in plasma and in dialysates,from the arcuate nucleus In control rats, basal glutamate levels in 20-min dialysate samples from the arcuate nucleus were 0.17 ___0.011 b~M (n = 9). These usually stabilized after 3 - 4 h of perfusion, and remained stable for the next 5 h. Plasma glutamate levels in these animals were 254_+ 10.7 p~M. Intraperitoneal administration of saline did not modify glutamate levels in plasmas or dialysates for at least 4 h after the injection (data not shown). All systemic MSG doses tested in this study induced significant dose-dependent increases in glutamate levels within arcuate nucleus dialysates (Fig. 1). Significant increases were observed in the first 20-min sample after MSG administration. Maximal increases in glutamate lev-
2000
3.2. Effects of dietao, MSG administration on glutamate leL,els in plasma and dialysates from the arcuate nucleus The amounts of food consumed during the 1 h feeding period of the experiment were 7.24 4- 0.97 g in rats provided with the control diet (n = 6), and 7.88 4-0.76, in those given the MSG-based diet (n = 8). After correction for body weights, the oral dose of MSG consumed by the animals during 1 h was calculated to have been 2.33 _+ 0.37 g/kg. In rats consuming the control diet no significant changes were observed in glutamate levels within plasmas or
*
1 oo 1600 -
.
1400 •
~
1200
.
MSG
doses,
1000
g/kg i p: --~-- 0.25
1/ 600
.
400
*
', T . *
o.5
,
*
*
200
0
2
3
4
5
6 Time,
7
8
20-rain
9
10
11
12
13
14
15
intervals
Fig. 1. Effects of systemic MSG administration (0.25, 0.5, 1 or 2 g / k g , i.p.) on extracellular glutamate levels in dialysates of the hypothalamic arcuate
nucleus in freely moving rats. Injection of MSG is indicated by the arrow. Data are presented as means +_S.E.M. of percent variations from mean basal glutamate levels (three consecutive samples in which glutamate levels varied by no more than 15%). Each time point represents data from 6 to 8 animals. • P < 0.05, compared at the correspondingtime points with the saline-injected control group; two-tailed Mann-Whitney U-test.
79
M.B. Bogdanov et a l . / Brain Research 736 (1996) 76-81 20000 18000 16000. 14O0O
~ 12000 e~
M S G d o s e s , g / k g i.p.:
"6 10000
~--I
\*
~
0.25
\ \.
8000
6000 4000 2000O~ 2
3
4
5
6
7
8
9
10
11
12
13
14
15
Time, 2 0 - m i n intervals
Fig. 2. Effects of systemic MSG administration (0.25, 0.5, 1 or 2 g/kg, i.p.) on plasma glutamate levels in freely moving rats. Injection of MSG is indicated by the arrow. Data are presented as means _+ S.E.M. of percent variations from mean basal plasma glutamate levels (three consecutive samples, corresponding to the basal dialysate samples). Each time point represents data from 6 to 8 animals. * P < 0.05, compared at the corresponding time points with saline-injected controls; two-tailed Mann-Whitney U-test.
dialysates f r o m arcuate nucleus, either during the 1 h f e e d i n g period or within the subsequent 4 h (Figs. 3 and
period), but still r e m a i n e d significantly elevated (up to 135 + 11% o f basal levels).
4). C o n s u m p t i o n o f the M S G - c o n t a i n i n g diet failed to affect extracellular g l u t a m a t e levels in the arcuate nucleus (Fig. 3). P l a s m a glutamate levels, m e a s u r e d simultaneously, w e r e significantly e l e v a t e d (Fig. 4). T h e s e levels w e r e increased (up to 152 _+ 12% o f basal values, P < 0.05) after 20 min, and increased further to a peak o f 352 _+ 61% o f basal values 140 min after the start o f the f e e d i n g period. P l a s m a g l u t a m a t e levels tended to decline by the end o f the e x p e r i m e n t (5 h after the start o f f e e d i n g
4. Discussion T h e s e data show that a high single dietary dose o f M S G , w h i c h significantly elevates p l a s m a glutamate levels, fails to affect extracellular glutamate levels in the rat's arcuate nucleus. U s i n g p r e v i o u s l y trained animals, it was possible to p r o v i d e the animals with as m u c h as 2.3 g / k g o f oral M S G during the 1-h period o f f o o d consumption.
1281 ] I _
DIET: 15
" - * - control
=, (.9
•"-e-- M S G
FEEDING 0 2
3
4
5
6
7
8
9
10
11
12
13
t4
15
16
17
18
Time, 20-min intervals Fig. 3. Effects of ad libitum consumption of control or MSG-based diet on extracellular glutamate levels in dialysates of the hypothalamic arcuate nucleus in freely moving rats. Control or MSG-based diets were present over the 1-h period during the experiment, as indicated by the bar. The amounts of food consumed in the 1 h feeding period were 7.24 + 0.97 g among animals provided with the control diet (n = 6), and 7.88 ± 0.76 g among those provided with the MSG-based diet (n = 8). The MSG dose provided over the 1-h period of feeding was 2.33 + 0.37 g/kg. Data are presented as means + S.E.M. of percent variations from mean basal glutamate levels (three consecutive samples before start of the 1 h feeding period).
M.B. Bogdanou et al. / Brain Research 736 (1996) 76-81
80 450
400
35O
3OO
-~
DIET:
.D
250 -
.
200 •
"6
~
control
~
MSG
3 150
100
FEEDING 5O
0
1. 2
3
4
s
6
t
8
o
lo
11
12
13
14
15
16
1T
18
Time, 20-rain intervals
Fig. 4. Effects of ad libitum consumption of control or MSG-based diet on the plasma glutamate levels in freely moving rats. Control or MSG-based diets were present over the 1-h period during the experiment, as indicated by the bar. The amounts of food consumed in the 1 h feeding period were 7.24 _+ 0.97 g among animals provided with the control diet (n = 6), and 7.88 _+ 0.76 g among those provided with the MSG-based diet (n = 8). The MSG dose provided over the 1-h period of feeding was 2.33 _+ 0.37 g/kg. Data are presented as means _+ S.E.M. of percent variations from mean basal glutamate levels (three consecutive samples before start of the 1 h feeding period). * P < 0.05, compared at the corresponding time points with the data obtained in animals provided with the control diet; two-tailed Mann-Whitney U-test.
The magnitude of the peak increase in plasma glutamate levels in animals that consumed this MSG-based diet, up to 3.5 times basal levels, was in the range reported previously by other authors [1,2,4,10,15]. In contrast to dietary MSG, systemic MSG administration did increase glutamate levels within arcuate nucleus dialysates and in plasma, and the increases within the arcuate nucleus were comparable to those we described previously in the rat's striatum, a brain region within the blood-brain barrier [3]. The most obvious explanation for the observed differences between the effects of systemic and dietary MSG on extracellular glutamate levels in the arcuate nucleus is the pattern and magnitude of the increases in plasma glutamate levels that follow these two routes of administration. The lowest systemic MSG dose tested in this study (0.25 g / k g , i.p.) induced a 10-fold increase in plasma glutamate levels. Extracellular glutamate levels in the arcuate nucleus increased significantly, but by only 16% above basal values. In the feeding experiments, where no overall effect was observed on the arcuate dialysate glutamate levels, a much smaller (to 3.5-fold basal levels) increase in plasma glutamate levels was observed; this apparently was insufficient to affect the arcuate nucleus. It seems unlikely that arcuate glutamate levels would have risen significantly at a latter time point, since plasma glutamate levels in the feeding studies were maximally increased 140 rain after the start of feeding and remained elevated through the end of observation period. The mechanisms that determine the magnitude of the rise in arcuate nucleus glutamate levels after systemic MSG administration are not clear. The time delay between the maximal increases in plasma and in arcuate levels, and approximately the 10-fold differences between the maxi-
mal increases observed after all systemic MSG doses tested, suggest the operation of mechanisms which efficiently buffer changes in arcuate nucleus extracellular glutamate levels. These could include a transport system in brain capillary endothelial cells that actively extrudes glutamate from the brain extracellular fluid to the blood and glial a n d / o r neuronal glutamate uptake systems [6,9,16,19]. The increase in plasma osmolarity that follows systemic MSG administration [12,22] probably increases the permeability of the BBB at the level of the arcuate nucleus. In the presence of increased plasma glutamate levels this allows more glutamate to enter this nucleus. In accordance with this hypothesis, it has recently been demonstrated that opening of the BBB by an intracarotid protamine infusion enhances effects of intracarotid glutamate on extracellular glutamate levels in parietal cortex [23]. Monno et al. recently reported that 4 g / k g of MSG given to adult rats by forced gavage induced 4.2- and 8.9-fold increases in extracellular glutamate levels within hippocampus and hypothalamus respectively; plasma glutamate levels were increased only 5.3-fold [14]. It is difficult to explain how the increase in extracellular brain glutamate levels could exceed the concurrent (40 min after MSG administration) increase in plasma glutamate levels. Basal plasma glutamate levels reported in the study of Monno et al. (261 ~M) were about the same as those found in this study (254 puM). In our experiments, we found that increases in plasma glutamate levels are generally about 10 times higher than those in arcuate extracellular glutamate levels, following all systemic MSG doses tested. The neurotoxicity of exogenous glutamate, when administered systemically to rodents at very high doses has
M.B. Bogdanov et al. / Brain Research 736 (1996) 76-81
been well documented by numerous investigators, while neurotoxic effects of MSG ingested as a dietary constituent have never been described. The mechanism of MSG's neurotoxic effects has been attributed to a sustained increase in extracellular brain glutamate concentrations [18]. Our present findings strongly suggest that different changes in plasma glutamate levels following systemic or dietary MSG may differentially affect brain glutamate levels and may in turn underlie the apparent lack of neurotoxic potential of MSG following its dietary consumption.
Acknowledgements Expert technical assistance of Jeffrey Breu is greatly acknowledged. This study was supported in part by a grant from the Center for Brain Sciences and Metabolism Chafftable Trust.
References [1] Airoldi, L., Bizzi, A., Salmona, M. and Garratini, S., Attempts to establish the safety margin for neurotoxicity of monosodium glutamate. In L.J. Filer, S. Garratini, M.R. Kare, W.A. Reynolds and R.J. Wurtman (Eds.), Glutamic Acid: Advances in Biochemistry and Physiology, Raven Press, New York, 1979, pp. 321-331. [2] Bizzi, A., Veneroni, E., Salmona M. and Garratini, S., Kinetics on monosodium glutamate in relation to its neurotoxicity, Toxicol. Lett., 1 (1977) 123-130. [3] Bogdanov, M.B. and Wurtman, R.J., Effects of systemic or oral ad libitum monosodium glutamate administration on striatal glutamate release, as measured using microdialysis in freely moving rats, Brain Res., 660 (1994) 337-340. [4] Caccia, S., Garattini, S., Ghezzi, P. and Zanini, M.G., Plasma and brain levels of glutamate and pyroglutamate after oral monosodium glutamate to rats, Toxicol. Lett., 10 (1982) 169-175. [5] Donzanti, B.A. and Yamamoto, B.K., An improved and rapid HPLC-EC metod for the isocratic separation of amino acid neurotransmitters from brain tissue and microdialysis perfusates, Life Sci., 43 (1988) 913-922. [6] Fonnum, F., Glutamate: a neurotransmitter in mammalian brain, J. Neurochem., 42 (1984) 1-11. [7] Garattini, S., Evaluation of the neurotoxic effects of glutamic acid, In R.J. Wurtman and J.J. Wurtman (Eds.), Nutrition and the Brain, Raven Press, New York, 1979, pp. 79-124.
81
[8] Heywood, R., James, R.W. and Worden, A.N., The ad libitum feeding of monosodium glutamate to wealning mice, Toxicol. Lett., 1 (1977) 151-155. [9] Hutchison, H.T., Eisenberg, H.M. and Haber, B., High-affinity transport of glutamate in rat brain microvessels, Exp. Neurol., 87 (1985) 260-269. [10] Liebshutz, J., Airoldi, L., Brownstein, M.J., Chinn, N.G. and Wurtman, R.J., Regional distribution of endogenous and parenteral glutamate, aspartate and glutamine in rat brain, Biochem. Pharmacol., 26 (1977) 443-446. [11] Lucas, D.R. and Newhouse, J.P., The toxic effect of sodium Lglutamate on the inner layers of the retina, AMA Arch. Opthalmol., 58 (1957) 193-201. [12] McCall, A., Glaeser, B.S., Millington, W. and Wurtman, R.J., Monosodium glutamate neurotoxicity, hyperosmolarity and bloodbarrier disfunction, Neurobehav. Toxicol., 1 (1979)279-283. [13] Merchenthaler, I., Neurons with access to the general circulation in the central nervous system of the rat: a retrograde tracing study with fluoro-gold, Neuroscience, 44 (1991) 655-662. [14] Monno, A., Vezzani, A., Bastine, A., Salmona, M. and Garattini, S., Extracellular glutamate levels in the hypothalamus and hippocampus of rats after acute or chronic oral intake of rnonosodium glutamate, Neurosci. Lett., 193 (1995) 45-48. [15] O'Hara Y. and Takasaki, Y., Pelatiobship between plasma glutamate levels and hypothalamic lesions in rodents, Toxicol. Lett., 4 (1979) 499-505. [16] Oldendorf, W. and Szabo, J., Amino acid assignment to one of three blood-brain barrier amino acid carriers, Am. Z Physiol., 230 (1976) 94-98. [17] Olney, J.W., Brain lesions, obesity and other disturbances in mice treated with monosodium glutamate, Science, 164 (1969) 719-721. [18] Olney, J.W., Excitotoxic amino acids: Research applications and safety implications, In L.J. Filer, S. Garratini, M.R. Kare, W.A. Reynolds and R.J. Wurtman (Eds.), Glutamic Acid: Advances in Biochemistry and Physiology, Raven Press, New York, 1979, pp. 287-319. [19] Pardridge, W.M., Regulation of amino acids availability to brain: Selective control mechanisms for glutamate, In L.J. Filer, S. Garratini, M.R. Kare, W.A. Reynolds and R.J. Wurtman (Eds.), Glutamic Acid: Advances in Biochemistry and Physiology, Raven Press, New York, 1979, pp. 125-138. [20] Paxinos, G. and Watson, C., The Rat Brain in Stereotaxic Coordinates, Academic Press, New York, 1986. [21] Peng, Y., Gubin, J., Harper, A.E., Vavich, M.G. and Kemmer, A.R., Food intake regulation: Amino acid toxicity and changes in rat brain and plasma amino acids, J. Nutr., 103 (1973) 608-617. [22] Torii, K., Takasaki, Y., Iwata, S. and Wurtman, R.J., Changes in blood osmolarity, electrolytes and metabolites among adult rats treated with a neurotoxic dose of MSG, Life Sci., 28 (1981) 2855-2864. [23] Westergen, I., Nystrom, B., Hamberger, A. and Johansson, B., Intracerebrai dialysis and blood-brain barrier, J. Neurochem., 64 (1995) 229-234.