251
Brain Research, 584 (1992) 251-256 © 1992 Elsevier Science Publishers B.V. All rights reserved 0006-8993/92/$05.00
BRES 17913
An in vitro circadian rhythm of protein synthesis in the rat suprachiasmatic nucleus under tissue culture conditions Shigenobu Shibata, Toshiyuki Hamada, Keiko Tominaga and Shigenori Watanabe Department of Pharmacology, Faculty of Pharmaceutical Sciences, Kyushu University 62, Fukuoka 812, Japan (Accepted 25 February 1992)
Key words: Circadian rhythm; Suprachiasmatic nucleus; Protein synthesis; tissue culture
Because inhibitors of protein synthesis produce changes in the circadian rhythm of wheel-running activity in rodents, we examined the circadian changes of in vitro protein synthesis by the rat suprachiasmatic nucleus (SCN). We demonstrated a robust circadian rhythm of [14C]leucine incorporation as well as [3H]2-deoxyglucose (2DG) uptake by the SCN. The peak time of 2DG uptake was around circadian time 6-9 h (CT6-CT9). In contrast to 2DG uptake, the maximum rate of leucine incorporation occurred around CT22-CT0. Thus, the leucine incorporation rate preceded 2DG uptake by 6-9 h. Leucine incorporation was inhibited by protein synthesis inhibitors, but not by tetrodotoxin. Since a robust circadian rhythm of leucine incorporation by the SCN was detected in vitro by using our method, this procedure may be useful to study the circadian clock function of the SCN.
INTRODUCTION The mammalian suprachiasmatic nucleus (SCN) has been identified as a circadian pacemaker for behavioral and physiological functions 17. Endogeneous circadian rhythms of the SCN neuronal firing rate and 2-deoxyglucose (2DG) uptake have been reported both in vivo 5'19 and in vitro 13'23, which appears to be direct evidence of the pacemaker function of this brain nucleus. Inhibitors of protein synthesis on 80 S ribosomes produce changes in the length or phase of circadian rhythms in a variety of organisms, including Acetabularia 8, Euglena 3, Gonyaulax 26, Neurospora 11, and Aplysia 7'9'16, suggesting that protein synthesis is somehow required to regulate the circadian rhythm. In addition, recent studies have demonstrated that phase changes in the circadian rhythm of hamster wheel-running activity were produced not only by the peripheral injection of cycloheximide and anisomycin 25 but also by injection of anisomycin into the SCN 6. These findings suggest that it would be of interest to know whether the overall rate of protein synthesis in the SCN varies with the time of day. Previous studies
using radiolabeled amino acids as tracers have demonstrated possible circadian variations of protein synthesis in the S C N 15'27. However, other studies did not find a circadian rhythm of protein synthesis in the rat S C N 4'18. There is a large body of literature on the role of protein synthesis in the fields of learning, sleep, feeding and sexual behavior in mammals 1. In vivo experiments cannot exclude hormonal influences or neuronal influences from other brain areas, and 2DG uptake by the golden hamster SCN shows a more robust rhythm in the in vitro 22 situation than it does in vivo 21. Accordingly, the present study determined the overall rate of SCN protein synthesis using cultured SCN slices. MATERIALS AND METHODS Slice preparation Wistar rats, weighing 200-300 g were housed under a normal (lights on at 08.00 h) or reversed (lights on at 20.00 h) 12:12 light-dark cycle for at least 2 weeks prior to the study. To eliminate the direct effects of light on protein synthesis within the SCN slices, all rats were placed in darkness for 2 days prior to killing and then decapitated under dim red light. In this experiment, the circadian time (CT) refers to the clock time, with the lights-on time being arbitrarily designated as CT0.
Correspondence: S. Shibata, Dept. of Pharmacology, Faculty of Pharmaceutical Sciences, Kyushu University 62, Fukuoka 812, Japan.
252 Animals were killed on day 1 at 8 different times of the day (CT 0, 3, 6, 9, 12, 15, 18 and 21), and the brains were quickly removed from their skulls. Coronal hypothalamic slices (450 > m in thickness) were prepared through the SCN and the anterior hypothalamic area (AHA) using a tissue chopper. The slices were placed between two meshes and were continuously perifused with Krebs-Ringer solution warmed to 36°C and equilibrated with 95% O 2 / 5 % CO2, using an identical chamber reported previously 12. The composition of the control Krebs solution was (in mM): NaC1, 129; MgSO4, 1.3; NaHCO3, 22.4; KH2PO4, 1.2; KCI, 4.2; glucose, 10.0; CaCI> t.5; and HEPES, 10 (0.5% gentamycin was also added). This buffer was maintained at pH 7.3-7.4 throughout the experiment. After preincubation for 24 h, slices were transfered into the incubation chamber on day 2. The incubation medium contained 1 / , C i / m l of 2DG (2-deoxy-D-[3H]glucose, specific activity 30-60 Ci/mmol; Amersham) a n d / o r 0.1-0.2 > C i / m l of leucine(L-[14C]leucine, sp. act. 55 mCi/mmol; Amersham). Incubation was carried out for 45 min at 37°C using the previously reported chamber 12, which was equipped to recirculate 14 ml of buffer at 4.4 m l / m i n , with continuous bubbling of humidified 95% 0 2 / 5 % CO 2 through the buffer as it entered the chamber. To initiate incubation, the slices held between two meshes were removed from the preincubation chamber and drained briefly. Then 1 ml of the isotope buffer was pipetted over the slices, which were then placed in the incubation chamber within 10 s of removal from the first chamber. In this manner, a square wave pulse of radioactivity was delivered to the slices. The slices were then removed from the chamber, rinsed with 20 ml of warm gassed preincubation buffer, and returned to the original preincubation chamber for 30 min. The procedure for monitoring protein synthesis in rat brain slices using leucine incorporation has been reported previously J4,2s, and we used this method with some modifications. For the [14C]leucine incorporation study, slices were incubated and labeled with [14C]leucine in the same way, but before terminating incubation the soluble radioactive leucine was 'chased' with a 30-rain wash-out using 1 mM unlabeled leucine. This reduced the level of soluble radioactive leucine to < 10% of that in slices which were incubated with [14C]leucine and not chased. This procedure preferentially removed the soluble [14C]leucine, which confirmed the finding of Vornov and Coyle 2s. In preliminary experiments, the incorporation of leucine into protein was linear for up to 60 rain. After the wash-out, slices were placed on dry ice to stop metabolism for the biochemical measurement of radio-isotopes. In some experiments, slices were fixed immediately in ice-cold 4% paraformaldehyde (for 24 h) before autoradiography. It has been reported that over 80% of the radio-isotope that remains in the tissue after this treatment is incorporated in protein 14.
autoradiogram. The optical densities of the standards were also determined and a standard curve was obtained. Using this standard curve, the measured tissue optical densities were converted to tissue radioactivity expressed as n C i / g tissue/45 rain.
Drugs The drugs used in this study were tetrodotoxin (TTX, Sankyo Inc. Japan), cycloheximide (Sigma Chemicals), and anisomycin (Sigma Chemicals). All drugs were dissolved in distilled water before use. Cycloheximide and anisomycin were added to the Krebs-Ringer solution from 30 min prior to the addition of [14C]leucine and throughout exposure to the radiolabeled amino acid to block protein synthesis. It is reported that treatment with cycloheximide blocks protein synthesis but does not affect the accumulation of precursors t4. Tetrodotoxin was also added to the slices from 30 min prior to the addition of [14C]leucine and [3H]2-deoxyglucose and throughout exposure to the radiolabeled amino acid. TTX treatment abolised firing by SCN neurons within 1 min (unpublished observation). Therefore, pretreatment with TTX for 30 min was regarded as sufficient to block all SCN neuronal activity.
Statistical analysis All data were analyzed with A N O V A and significant differences between groups were determined using Duncan's test. RESULTS
The present experiment demonstrated a robust circadian rhythm of leucine uptake as well as 2DG uptake
_= E 10 o
Biochemical analysis The SCN was punched out from the slices which were placed on dry ice according to method reported prviously m, and the remaining hypothalamic regions in the slices were designated as the anterior hypothalamic area (AHA). The SCN and A H A containing [3H]2DG and [14C]leucine were homogenized in 1 ml of phosphate buffer containing 0.5% perchloric acid, and 450/,1 of the homogenate was used to determine the total protein content. The radioactivity in another 450/,1 of homogenate was determined with a liquid scintilation counter after sufficient solubilization and normalization for the protein content of the sample. Averages for groups of animals are expressed as the mean _+S.E. in d p m / m g total protein/45 min.
Autoradiography After fixation with paraformaldehyde for 24 h, tissue slices containing [14C]leucine were frozen at - 8 0 ° C for autoradiography. A previous study showed that over 80% of the 14C that remains in the tissue after this treatment is incorporated into protein 14. Frozen tissue slices were cut into 2 0 / , m sections on a cryostat mounted with uniform pressure in a casette with 14C-labeled standards, and exposed to X-ray film for 7 days. After exposure, the same sections were stained with Cresyl violet for morphological identification. The optical density of each region was measured with an image analyzer (MCID, Muromachi, Japan) that permits quantitative analysis of the
E= -o v
(7)
17)
(7)
2'1
0'
3.
(10)
(7)
(7)
17)
(10)
(7)
(7)
12 Time
1's
1'8
21 '
0'
5
01: . . 6.
9 Circadian
•
•
2-deoxyglucoae uptake
©
,~
leucine incorporation
Fig. 1. Circadian rhythm of [14C]leucine uptake and [3H]2-deoxyglucose uptake by the rat hypothalamic slices. After a 45-min exposure to [lac]leucine and [3H]2-deoxyglucose simultaneously, the SCN was punched out from the slices and the remaining hypothalamic regions were designated as the AHA. Then the radioactivity in this nucleus was measured and the protein content was determined. Each point shows the mean _+S.E.M. of normalized radioactivity (dpm per mg of protein). The numbers in parentheses are the number of animals. The slices measured at each CT (0, 3, 6, 9, 12, 15, 18 and 21) on day 2 were prepared at the corresponding CT (0, 3, 6, 9, 12, 15, 18 and 21) on day 1. SCN, suprachiasmatic nucleus; AHA, anterior hypothalamic area. Circadian changes of leucine incorporation into protein and 2-deoxyglucose uptake were significant ( P < 0.01) for the SCN, but they were not significant for the A H A ( P > 0.05) (ANOVA).
253 by the SCN, while the AHA showed no circadian rhythm when incubated simultaneously with [3H]2DG and [laC]leucine (Fig. 1). In this experiment, we simul-
taneously examined glucose and leucine uptake. The peak time of 2DG uptake in the SCN was around CT6-CT9, and thus the present results coincided well
Fig. 2. Autoradiographic analysis of [ 14C]leucine incorporation into protein by the slices of rat suprachiasmatic nucleus. Slices were prepared and incubated with [14C]ieucine as described in Materials and Methods and [~4Clleucine incorporation was determined by autoradiography, b and f, control slices exposed to [14C]leucine for 45 min. b, CT9; f, CT0. d and h, slices treated with cycloheximide (1 /.LM) from 30 rain prior to the addition of radiolabeled chemicals, and also throughout the exposure to tracers for 45 min as in b and f. d, CT9; h, CT0. a, c, e, g, Cresyl violet stain. Bar = 500 ~tm. A high rate of [14C]leucine incorporation can be seen as heavily labeled areas in the SCN at CT0. At CT9, the SCN shows little [14C]leucine incorporation.
254 TABLE I
Effects of protein synthesis inhibitors and tetrodotoxin on the rhythm of leucine incorporation and 2-deoxyglucose uptake in the rat suprachiasmatic nucleus Treatment
Circadian time (CT)
Control
Cycloheximide (0.1 ixM)
Leucine incorporation
CT0 CT9
18.2-+2.3 11.9 _+1.2 '~
11.8_+1.2 * 9.5 _+ 1.1
2-Deoxyglucose uptake
CT0 CT9
7.2 -+ 1.1 12.8 + 1.0 ~#
7.5 _+3.0 11.1 + 1.7
Anisomycin (1 IzM) 4.4:t:1.1 ** 4.3 + 1.0 * * 5.8 -+ 7.0 10.5 ___1.8
TTX (1 txM) 18.2+2.0 12.0 _+ 1.2 5.9 + 1.2 6.4 + 6.0 * *
[14C]leucine incorporation and [3H]2-deoxyglucose uptake was calculated as normalized radioactivity (dpm per mg of protein) after a 45-min exposure to [14C]leucine and [3H]2-deoxyglucose simultaneously. Cyloheximide, anisomycin and T T X were added to the slices for 30 min prior to the addition of radiolabeled chemicals and throughout the exposure to tracers. After SCN was punched out, the radioactivity in this nucleus was measured and the protein content was determined by Bio-Rad protein assay. T h e data are given as d p m / m g p r o t e i n / 4 5 min (mean -+ S.E. from 6 to 9 animals). * P < 0.05, ** P < 0.01 vs. control group (Duncan's multiple-range test) and # P < 0.05, ~ ' P < 0.01 vs. CT0 group (Duncan's multiple-range test).
with previous in vivo data ~9. The maximal rate of leucine incorporation occurred at around CT22-CT0, so leucine incorporation preceded that of 2DG by 6-9 h. The leucine incorporation rate by the SCN was unaffected by coincubation with TTX (1 t~M), while 2DG uptake was reduced by this treatment (Table I). In contrast, cycloheximide (0.1/~M) and anisomycin (1 ~M) significantly reduced leucine incorporation by the SCN, while 2DG uptake was unaffected by these compounds. The reduction of leucine uptake by cycloheximide and anisomycin was more marked at CT0 than at CT9. Autoradiographic analysis of the leucine uptake rate was performed. Autoradiographs of the SCN and AHA obtained from slices incubated in [~4C]leucine-containing solution at CT9 and CT0 are presented in Fig. 2. The SCN appears as a pair of dark regions above the optic chiasm at CT0, but it is not evident at CT9. In contrast, the incorporation of leucine by the AHA was very low. Autoradiographs of the SCN from slices incubated with 1 /~M cycloheximide are shown in Fig. 2. The incorporation of leucine by the SCN was high in a normal culture medium, but was reduced by incubation with cycloheximide at CT9 and CT0. When slices were incubated in cycloheximide to inhibit protein synT A B L E II
Effect of cycloheximide on the rhythm of [14C]leucine incorporation in the rat suprachiasmatic nucleus Circadian time (CT)
Control
Cycloheximide (1 Iz M)
CT0 CT9
238+ 12 1 3 9 + 3 0 **
67-+ 19 * 70+ 5 *
[14C]Leucine incorporation was calculated using the M C I D system for the control and cycioheximide-treated groups shown in the autoradiograph in Fig. 2. Optical densities were converted to n C i / g tissue using the m e a n + S.E. from 3 animals, * P < 0.01 vs. control group (Duncan's multiple-range test), * * P < 0.05 vs. CT0 group (Duncan's multiple-range test).
thesis, the optical density of the autoradiographs was reduced, indicating that the radioactivity detected by autoradiography was incorporated into protein. The data are summarized in Table II. In control groups, the incorporation rate was 238 ± 12 (n = 3) nCi/g tissue/ 45 min in the SCN at CT0 and 139 ± 30 (n = 3) at CT9. In cycloheximide-treated cultures, autoradiographs of the SCN showed no significant difference between CT9 and CT0. The incorporation rate of leucine by the SCN was 67 + 19 (n = 3) at CT0 and 70 ± 5.0 (n = 3) at CT9. DISCUSSION The present experiment demonstrated a robust circadian rhythm of protein synthesis in the rat SCN when it was cultured for 24 h. The maximum rate of leucine incorporation occurred around CT22-CT0, while the peak 2DG uptake was around CT6-CT9. Thus, incorporation of leucine preceded 2DG uptake by 6-9 h and the spontaneous discharge of SCN neurons 5'23. Recently, Shinohara et al. 24 reported that somatostatin (SST) and SST mRNA were present in the SCN and that SST and SST mRNA levels fluctuated significantly with the circadian cycle, so that the peak mRNA level preceded spontaneous neuronal discharge by 6-8 h. This result is consistent with our present findings and suggests that the circadian rhythm of the leucine incorporation rate reflects the circadian changes of synthesis of peptides like SST, since the phase of leucine incorporation coincides with the expression of SST mRNA. However, at present we do not know the precise mechanisms that cause the peak of leucine incorporation to precede the peaks of spontaneous discharge or 2DG uptake. The rate of leucine incorporation was high at CT0, but decreased rapidly at CT3 and CT6. The shape of the phase-response curve (PRC) to the
255 direct microinjection of anisomycin into the SCN was very similar in shape to the PRCs obtained when protein synthesis inhibitors were applied to Aplysia and Neurospora. The 'breakpoint' (the transition from phase delay to phase advance) of the PRCs occur at around CT4 in the hamster 6'25. This breakpoint (CT4) may be related to the rapid decrease of protein synthesis between CT0 and CT6. Treatment with cycloheximide and anisomycin strongly inhibited the circadian changes of leucine incorporation, but did not affect 2DG uptake. This result strongly suggests that circadian changes of leucine incorporation by the SCN may reflect its clock function or some related function. It is not clear whether the phase shifts induced by the microinjections of protein synthesis inhibitors into the SCN are due to a general inhibition of protein synthesis or to the inhibition of specific protein(s) that may be related to SCN oscillatory functions. The effects of cycloheximide and anisomycin on the incorporation of leucine were dependent upon the circadian time (CT), since the reduction of leucine incorporation by these chemicals was more evident at CT0 than at CT9. In addition, the phase change of hamster activity induced by inhibitors of protein synthesis depends upon the time of injection6'25. These findings suggest that circadian changes of leucine incorporation may be tied to a component of the SCN circadian clock mechanism rather than the result of some non-specific events. Whether leucine incorporation is a direct component of the pacemaker mechanism or just another 'hand of clock' remains to be elucidated. TTX treatment inhibited 2DG uptake, but did not affect leucine incorporation. Recent studies have suggested that TTX blocks the appearance of the overt rhythm, vasopressin release 2, and locomotor activity2°, but does not change the circadian clock mechanism itself. Therefore, the present result is consistent with these above-mentioned reports. The present findings taken together with previous reports have suggested that circadian changes of 2DG uptake or firing activity may be related to the overt rhythm rather than the clock mechanism. In previous in vivo experiments, despite the observed robust circadian rhythms of electrical activity and energy metabolism 5A3,19,23, the overall protein synthesis in the SCN appeared to remain constant 4A8. Therefore, the data presented here disagree with these previous in vivo studies. The reason for the difference is uncertain, but there are three possible explanations: (i) the in vivo data are incorrect; (ii) the in vitro situation unmasks a rhythm not evident in vivo; and (iii) the in vitro situation induces a rhythm not present in vivo. It is not possible to distinguish between these
possibilities with the information currently available. Regarding possibility (i); although there were no differences at the two time points sampled in vivo, it remains possible that the use of other sampling times may yet disclose a rhythm like that demonstrated in the present results. Possibility (ii) may have arisen due to differences of experimental conditions. We previously reported that 2DG uptake by the golden hamster SCN appeared to show a more robust rhythm in vitro 22 than in vivo 21. Therefore, culture for 24 h may unmask a rhythm not evident in vivo. At present, the possibility that option (iii) is correct remains unknown. In summary, we elucidated a clear circadian change of in vitro leucine incorporation by the SCN suggesting that our method may be useful for studying the circadian clock function of the SCN. REFERENCES 1 Deutsch, J.A., Physiological Basis of Memory, 2nd edn., Academic Press, New York, 1983. 2 Earnest, D.J., Digiorgio, S.M. and Sladek, C.D., Effects of tetrodotoxin on the circadian pacemaker mechanism in suprachiasmatic explants in vitro, Brain Res. Bull., 26 (1991) 677-680. 3 Feldman, J.F., Lengthening the period of a biological clock in Euglena by cycloheximide, an inhibitor of protein synthesis., Proc. Natl. Acad. Sci. USA, 57 (1967) 1080-1087. 4 GIotzbach, S.F., Randall, T.L., Radeke, C.M. and Heller, H.C., Absence of a circadian rhythm of protein synthesis in the rat suprachiasmatic nucleus, Neurosci. Left., 76 (1987) 113-118. 5 Inouye, S.I. and Kawamura, H., Characteristics of a circadian pacemaker in the suprachiasmatic nucleus, J. Comp. Physiol., 146 (1982) 153-160. 6 Inouye, S.I., Takahashi, J., Wollnik, F. and Turek, F.W., Inhibitor of protein synthesis phase shifts a circadian pacemaker in mammalian SCN, Am. J. Physiol., 255 (1988) R1055-R1058. 7 Jacklet, J.W., Neuronal circadian rhythms; phase-shifting by a protein synthesis inhibitor, Science, 198 (1977) 69-71. 8 Karakashian, M.W. and Schweiger, H.G., Evidence for a cycloheximide-sensitive component in the biological clock of Acetabularia, Exp. Cell Res., 98 (1976) 303-312. 9 Lotshaw, D.P. and Jacklet, J.W., Involvement of protein synthesis in circadian clock of Aplysia eye, Am Z Physiol., 250 (1986) R5-R1. 10 Murakami, N. and Takahashi, K., Cicadian rhythm of adenosine 3',5'-monophosphate content in suprachiasmatic nucleus (SCN) and ventromedial hypothalamus (VMH) in the rat, Brain Res., 276 (1983) 297-304. 11 Nakashima, H., Perlman, J. and Feldman, J.F., Cycleheximide-induced phase-shifting of circadian clock of Neurospora, Am. J. Physiol., R31-R35. 12 Newman, G.C., Hospod, F.E. and Patlak, C.S., Kinetic model of 2-deoxygluxose metabolism using brain slices, Z Cereb. Blood Flow Metab., 10 (1990) 510-526. 13 Newman, G.C. and Hospod, F.E., Rhythm of suprachiasmatic nucleus 2-deoxyglucose uptake in vitro, Brain Res., 381 (1986) 345-350. 14 Raley-Susman, K.M. and Lipton, P., In vitro ischemia and protein synthesis in the rat hippocampal slice: the role of calcium and NMDA receptor activation, Brain Res., 515 (1990) 27-35. 15 Roberts, M.H. and Moore, R.Y., Diurnal changes of protein synthesis in the suprachiasmatic nucleus of the rat, Soc. Neurosci. Abstr., 12 (1986) 211. 16 Rothman, B.S. and Strumwasser, F., Phase-shifting the circadian rhythm of neural activity in the isolated Aplysia eye with
256
17 18
19
20
21
22
puromycin and cycloheximide: electrophysical and biochemical studies, ,L Gen. Physiol., 68 (1976) 359-384. Rusak, B. and Bina, K.G., Neurotransmitters in the mammalian circadian system, Annu. Rez~'. Neurosci., 13 (1990) 387-401. Scammell, T.E., Schwartz, W.J. and Smith, C.B., No evidence for a circadian rhythm of protein sythesis in the rat suprachiasmatic nuclei, Brain Res., 494 (1989) 155-158. Schwartz, W.J., Davidsen, L.C. and Smith, C.B., In vivo metabolic activity of a putative circadian oscillator, the rat suprachiasmatic nucleus, J. Comp. Neuro/., 189 (1980) 157-167. Schwartz, W.J., Gross, R.A., Morton, M.T., The suprachiasmatic nuclei contain a tetrodotoxin-resistant circadian pacemaker, Proc. NatL Acad. Sci. USA, 84 (1987) 1694-1698. Schwartz, W.J., Different in vivo metabolic activities of suprachiasmatic nuclei of Turkish and golden hamsters, Am. J. Physiol., (Regulatory, integrative and comparative physiology), 28 (1990) R1083-R1085. Shibata, S. and Moore, R.Y., Electrical and metabolic activity of suprachiasmatic nucleus neurons in hamster hypothalamic slices., Brain Res. 438 (1988) 374-378.
23 Shibata, S., Oomura, Y., Kita, H. and Hattori, K., Circadian rhythmic changes of neuronal activity in the suprachiasmatic nucleus of the rat hypothalamic slice, Brain Res., 247 (1982) 154-158. 24 Shinohara, K., Isobe, Y., Takeuchi, J. and Inouye, S.i., Circadian rhythms of somatostatin-immunoreactivity in the suprachiasmatic nucleus of the rat, Neurosci. Lett., 127 (1991) 9-12. 25 Takahashi, J.S and Turek, F.W., Anisomycin, an inhibitor of protein synthesis, perturbs the phase of a mammalian circadian pacemaker, Brain Res., 405 (1987) 199-203. 26 Taylor, W.R., Dunlap, J.C. and Hastings, J.W., Inhibitors of protein synthesis on 80 S ribosomes phase shift the Gonyaulax clock, J. Exp. Biol., 97 (1982) 121-126. 27 Van den Pol, A.M., Amino acid incorporation in medial hypothalamic nuclei; circadian, perikarya, and neuropil variations, Am. ,L Physiol., 240 (1981) RI6-R22. 28 Vornov, J.J. and Coyle, J.T., Glutamate neurotoxicity and inhibition of protein synthesis in the hippocampal slice, J. Neurochem., 56 (1991) 996-1006.
Brain Research, 584 (1992) 251-256 © 1992 Elsevier Science Publishers B.V. All rights reserved 0006-8993/92/$05.00
BRES 17913
An in vitro circadian rhythm of protein synthesis in the rat suprachiasmatic nucleus under tissue culture conditions Shigenobu Shibata, Toshiyuki Hamada, Keiko Tominaga and Shigenori Watanabe Department of Pharmacology, Faculty of Pharmaceutical Sciences, Kyushu University 62, Fukuoka 812, Japan (Accepted 25 February 1992)
Key words: Circadian rhythm; Suprachiasmatic nucleus; Protein synthesis; tissue culture
Because inhibitors of protein synthesis produce changes in the circadian rhythm of wheel-running activity in rodents, we examined the circadian changes of in vitro protein synthesis by the rat suprachiasmatic nucleus (SCN). We demonstrated a robust circadian rhythm of [14C]leucine incorporation as well as [3H]2-deoxyglucose (2DG) uptake by the SCN. The peak time of 2DG uptake was around circadian time 6-9 h (CT6-CT9). In contrast to 2DG uptake, the maximum rate of leucine incorporation occurred around CT22-CT0. Thus, the leucine incorporation rate preceded 2DG uptake by 6-9 h. Leucine incorporation was inhibited by protein synthesis inhibitors, but not by tetrodotoxin. Since a robust circadian rhythm of leucine incorporation by the SCN was detected in vitro by using our method, this procedure may be useful to study the circadian clock function of the SCN.
INTRODUCTION The mammalian suprachiasmatic nucleus (SCN) has been identified as a circadian pacemaker for behavioral and physiological functions 17. Endogeneous circadian rhythms of the SCN neuronal firing rate and 2-deoxyglucose (2DG) uptake have been reported both in vivo 5'19 and in vitro 13'23, which appears to be direct evidence of the pacemaker function of this brain nucleus. Inhibitors of protein synthesis on 80 S ribosomes produce changes in the length or phase of circadian rhythms in a variety of organisms, including Acetabularia 8, Euglena 3, Gonyaulax 26, Neurospora 11, and Aplysia 7'9'16, suggesting that protein synthesis is somehow required to regulate the circadian rhythm. In addition, recent studies have demonstrated that phase changes in the circadian rhythm of hamster wheel-running activity were produced not only by the peripheral injection of cycloheximide and anisomycin 25 but also by injection of anisomycin into the SCN 6. These findings suggest that it would be of interest to know whether the overall rate of protein synthesis in the SCN varies with the time of day. Previous studies
using radiolabeled amino acids as tracers have demonstrated possible circadian variations of protein synthesis in the S C N 15'27. However, other studies did not find a circadian rhythm of protein synthesis in the rat S C N 4'18. There is a large body of literature on the role of protein synthesis in the fields of learning, sleep, feeding and sexual behavior in mammals 1. In vivo experiments cannot exclude hormonal influences or neuronal influences from other brain areas, and 2DG uptake by the golden hamster SCN shows a more robust rhythm in the in vitro 22 situation than it does in vivo 21. Accordingly, the present study determined the overall rate of SCN protein synthesis using cultured SCN slices. MATERIALS AND METHODS Slice preparation Wistar rats, weighing 200-300 g were housed under a normal (lights on at 08.00 h) or reversed (lights on at 20.00 h) 12:12 light-dark cycle for at least 2 weeks prior to the study. To eliminate the direct effects of light on protein synthesis within the SCN slices, all rats were placed in darkness for 2 days prior to killing and then decapitated under dim red light. In this experiment, the circadian time (CT) refers to the clock time, with the lights-on time being arbitrarily designated as CT0.
Correspondence: S. Shibata, Dept. of Pharmacology, Faculty of Pharmaceutical Sciences, Kyushu University 62, Fukuoka 812, Japan.
252 Animals were killed on day 1 at 8 different times of the day (CT 0, 3, 6, 9, 12, 15, 18 and 21), and the brains were quickly removed from their skulls. Coronal hypothalamic slices (450 > m in thickness) were prepared through the SCN and the anterior hypothalamic area (AHA) using a tissue chopper. The slices were placed between two meshes and were continuously perifused with Krebs-Ringer solution warmed to 36°C and equilibrated with 95% O 2 / 5 % CO2, using an identical chamber reported previously 12. The composition of the control Krebs solution was (in mM): NaC1, 129; MgSO4, 1.3; NaHCO3, 22.4; KH2PO4, 1.2; KCI, 4.2; glucose, 10.0; CaCI> t.5; and HEPES, 10 (0.5% gentamycin was also added). This buffer was maintained at pH 7.3-7.4 throughout the experiment. After preincubation for 24 h, slices were transfered into the incubation chamber on day 2. The incubation medium contained 1 / , C i / m l of 2DG (2-deoxy-D-[3H]glucose, specific activity 30-60 Ci/mmol; Amersham) a n d / o r 0.1-0.2 > C i / m l of leucine(L-[14C]leucine, sp. act. 55 mCi/mmol; Amersham). Incubation was carried out for 45 min at 37°C using the previously reported chamber 12, which was equipped to recirculate 14 ml of buffer at 4.4 m l / m i n , with continuous bubbling of humidified 95% 0 2 / 5 % CO 2 through the buffer as it entered the chamber. To initiate incubation, the slices held between two meshes were removed from the preincubation chamber and drained briefly. Then 1 ml of the isotope buffer was pipetted over the slices, which were then placed in the incubation chamber within 10 s of removal from the first chamber. In this manner, a square wave pulse of radioactivity was delivered to the slices. The slices were then removed from the chamber, rinsed with 20 ml of warm gassed preincubation buffer, and returned to the original preincubation chamber for 30 min. The procedure for monitoring protein synthesis in rat brain slices using leucine incorporation has been reported previously J4,2s, and we used this method with some modifications. For the [14C]leucine incorporation study, slices were incubated and labeled with [14C]leucine in the same way, but before terminating incubation the soluble radioactive leucine was 'chased' with a 30-rain wash-out using 1 mM unlabeled leucine. This reduced the level of soluble radioactive leucine to < 10% of that in slices which were incubated with [14C]leucine and not chased. This procedure preferentially removed the soluble [14C]leucine, which confirmed the finding of Vornov and Coyle 2s. In preliminary experiments, the incorporation of leucine into protein was linear for up to 60 rain. After the wash-out, slices were placed on dry ice to stop metabolism for the biochemical measurement of radio-isotopes. In some experiments, slices were fixed immediately in ice-cold 4% paraformaldehyde (for 24 h) before autoradiography. It has been reported that over 80% of the radio-isotope that remains in the tissue after this treatment is incorporated in protein 14.
autoradiogram. The optical densities of the standards were also determined and a standard curve was obtained. Using this standard curve, the measured tissue optical densities were converted to tissue radioactivity expressed as n C i / g tissue/45 rain.
Drugs The drugs used in this study were tetrodotoxin (TTX, Sankyo Inc. Japan), cycloheximide (Sigma Chemicals), and anisomycin (Sigma Chemicals). All drugs were dissolved in distilled water before use. Cycloheximide and anisomycin were added to the Krebs-Ringer solution from 30 min prior to the addition of [14C]leucine and throughout exposure to the radiolabeled amino acid to block protein synthesis. It is reported that treatment with cycloheximide blocks protein synthesis but does not affect the accumulation of precursors t4. Tetrodotoxin was also added to the slices from 30 min prior to the addition of [14C]leucine and [3H]2-deoxyglucose and throughout exposure to the radiolabeled amino acid. TTX treatment abolised firing by SCN neurons within 1 min (unpublished observation). Therefore, pretreatment with TTX for 30 min was regarded as sufficient to block all SCN neuronal activity.
Statistical analysis All data were analyzed with A N O V A and significant differences between groups were determined using Duncan's test. RESULTS
The present experiment demonstrated a robust circadian rhythm of leucine uptake as well as 2DG uptake
_= E 10 o
Biochemical analysis The SCN was punched out from the slices which were placed on dry ice according to method reported prviously m, and the remaining hypothalamic regions in the slices were designated as the anterior hypothalamic area (AHA). The SCN and A H A containing [3H]2DG and [14C]leucine were homogenized in 1 ml of phosphate buffer containing 0.5% perchloric acid, and 450/,1 of the homogenate was used to determine the total protein content. The radioactivity in another 450/,1 of homogenate was determined with a liquid scintilation counter after sufficient solubilization and normalization for the protein content of the sample. Averages for groups of animals are expressed as the mean _+S.E. in d p m / m g total protein/45 min.
Autoradiography After fixation with paraformaldehyde for 24 h, tissue slices containing [14C]leucine were frozen at - 8 0 ° C for autoradiography. A previous study showed that over 80% of the 14C that remains in the tissue after this treatment is incorporated into protein 14. Frozen tissue slices were cut into 2 0 / , m sections on a cryostat mounted with uniform pressure in a casette with 14C-labeled standards, and exposed to X-ray film for 7 days. After exposure, the same sections were stained with Cresyl violet for morphological identification. The optical density of each region was measured with an image analyzer (MCID, Muromachi, Japan) that permits quantitative analysis of the
E= -o v
(7)
17)
(7)
2'1
0'
3.
(10)
(7)
(7)
17)
(10)
(7)
(7)
12 Time
1's
1'8
21 '
0'
5
01: . . 6.
9 Circadian
•
•
2-deoxyglucoae uptake
©
,~
leucine incorporation
Fig. 1. Circadian rhythm of [14C]leucine uptake and [3H]2-deoxyglucose uptake by the rat hypothalamic slices. After a 45-min exposure to [lac]leucine and [3H]2-deoxyglucose simultaneously, the SCN was punched out from the slices and the remaining hypothalamic regions were designated as the AHA. Then the radioactivity in this nucleus was measured and the protein content was determined. Each point shows the mean _+S.E.M. of normalized radioactivity (dpm per mg of protein). The numbers in parentheses are the number of animals. The slices measured at each CT (0, 3, 6, 9, 12, 15, 18 and 21) on day 2 were prepared at the corresponding CT (0, 3, 6, 9, 12, 15, 18 and 21) on day 1. SCN, suprachiasmatic nucleus; AHA, anterior hypothalamic area. Circadian changes of leucine incorporation into protein and 2-deoxyglucose uptake were significant ( P < 0.01) for the SCN, but they were not significant for the A H A ( P > 0.05) (ANOVA).
253 by the SCN, while the AHA showed no circadian rhythm when incubated simultaneously with [3H]2DG and [laC]leucine (Fig. 1). In this experiment, we simul-
taneously examined glucose and leucine uptake. The peak time of 2DG uptake in the SCN was around CT6-CT9, and thus the present results coincided well
Fig. 2. Autoradiographic analysis of [ 14C]leucine incorporation into protein by the slices of rat suprachiasmatic nucleus. Slices were prepared and incubated with [14C]ieucine as described in Materials and Methods and [~4Clleucine incorporation was determined by autoradiography, b and f, control slices exposed to [14C]leucine for 45 min. b, CT9; f, CT0. d and h, slices treated with cycloheximide (1 /.LM) from 30 rain prior to the addition of radiolabeled chemicals, and also throughout the exposure to tracers for 45 min as in b and f. d, CT9; h, CT0. a, c, e, g, Cresyl violet stain. Bar = 500 ~tm. A high rate of [14C]leucine incorporation can be seen as heavily labeled areas in the SCN at CT0. At CT9, the SCN shows little [14C]leucine incorporation.
254 TABLE I
Effects of protein synthesis inhibitors and tetrodotoxin on the rhythm of leucine incorporation and 2-deoxyglucose uptake in the rat suprachiasmatic nucleus Treatment
Circadian time (CT)
Control
Cycloheximide (0.1 ixM)
Leucine incorporation
CT0 CT9
18.2-+2.3 11.9 _+1.2 '~
11.8_+1.2 * 9.5 _+ 1.1
2-Deoxyglucose uptake
CT0 CT9
7.2 -+ 1.1 12.8 + 1.0 ~#
7.5 _+3.0 11.1 + 1.7
Anisomycin (1 IzM) 4.4:t:1.1 ** 4.3 + 1.0 * * 5.8 -+ 7.0 10.5 ___1.8
TTX (1 txM) 18.2+2.0 12.0 _+ 1.2 5.9 + 1.2 6.4 + 6.0 * *
[14C]leucine incorporation and [3H]2-deoxyglucose uptake was calculated as normalized radioactivity (dpm per mg of protein) after a 45-min exposure to [14C]leucine and [3H]2-deoxyglucose simultaneously. Cyloheximide, anisomycin and T T X were added to the slices for 30 min prior to the addition of radiolabeled chemicals and throughout the exposure to tracers. After SCN was punched out, the radioactivity in this nucleus was measured and the protein content was determined by Bio-Rad protein assay. T h e data are given as d p m / m g p r o t e i n / 4 5 min (mean -+ S.E. from 6 to 9 animals). * P < 0.05, ** P < 0.01 vs. control group (Duncan's multiple-range test) and # P < 0.05, ~ ' P < 0.01 vs. CT0 group (Duncan's multiple-range test).
with previous in vivo data ~9. The maximal rate of leucine incorporation occurred at around CT22-CT0, so leucine incorporation preceded that of 2DG by 6-9 h. The leucine incorporation rate by the SCN was unaffected by coincubation with TTX (1 t~M), while 2DG uptake was reduced by this treatment (Table I). In contrast, cycloheximide (0.1/~M) and anisomycin (1 ~M) significantly reduced leucine incorporation by the SCN, while 2DG uptake was unaffected by these compounds. The reduction of leucine uptake by cycloheximide and anisomycin was more marked at CT0 than at CT9. Autoradiographic analysis of the leucine uptake rate was performed. Autoradiographs of the SCN and AHA obtained from slices incubated in [~4C]leucine-containing solution at CT9 and CT0 are presented in Fig. 2. The SCN appears as a pair of dark regions above the optic chiasm at CT0, but it is not evident at CT9. In contrast, the incorporation of leucine by the AHA was very low. Autoradiographs of the SCN from slices incubated with 1 /~M cycloheximide are shown in Fig. 2. The incorporation of leucine by the SCN was high in a normal culture medium, but was reduced by incubation with cycloheximide at CT9 and CT0. When slices were incubated in cycloheximide to inhibit protein synT A B L E II
Effect of cycloheximide on the rhythm of [14C]leucine incorporation in the rat suprachiasmatic nucleus Circadian time (CT)
Control
Cycloheximide (1 Iz M)
CT0 CT9
238+ 12 1 3 9 + 3 0 **
67-+ 19 * 70+ 5 *
[14C]Leucine incorporation was calculated using the M C I D system for the control and cycioheximide-treated groups shown in the autoradiograph in Fig. 2. Optical densities were converted to n C i / g tissue using the m e a n + S.E. from 3 animals, * P < 0.01 vs. control group (Duncan's multiple-range test), * * P < 0.05 vs. CT0 group (Duncan's multiple-range test).
thesis, the optical density of the autoradiographs was reduced, indicating that the radioactivity detected by autoradiography was incorporated into protein. The data are summarized in Table II. In control groups, the incorporation rate was 238 ± 12 (n = 3) nCi/g tissue/ 45 min in the SCN at CT0 and 139 ± 30 (n = 3) at CT9. In cycloheximide-treated cultures, autoradiographs of the SCN showed no significant difference between CT9 and CT0. The incorporation rate of leucine by the SCN was 67 + 19 (n = 3) at CT0 and 70 ± 5.0 (n = 3) at CT9. DISCUSSION The present experiment demonstrated a robust circadian rhythm of protein synthesis in the rat SCN when it was cultured for 24 h. The maximum rate of leucine incorporation occurred around CT22-CT0, while the peak 2DG uptake was around CT6-CT9. Thus, incorporation of leucine preceded 2DG uptake by 6-9 h and the spontaneous discharge of SCN neurons 5'23. Recently, Shinohara et al. 24 reported that somatostatin (SST) and SST mRNA were present in the SCN and that SST and SST mRNA levels fluctuated significantly with the circadian cycle, so that the peak mRNA level preceded spontaneous neuronal discharge by 6-8 h. This result is consistent with our present findings and suggests that the circadian rhythm of the leucine incorporation rate reflects the circadian changes of synthesis of peptides like SST, since the phase of leucine incorporation coincides with the expression of SST mRNA. However, at present we do not know the precise mechanisms that cause the peak of leucine incorporation to precede the peaks of spontaneous discharge or 2DG uptake. The rate of leucine incorporation was high at CT0, but decreased rapidly at CT3 and CT6. The shape of the phase-response curve (PRC) to the
255 direct microinjection of anisomycin into the SCN was very similar in shape to the PRCs obtained when protein synthesis inhibitors were applied to Aplysia and Neurospora. The 'breakpoint' (the transition from phase delay to phase advance) of the PRCs occur at around CT4 in the hamster 6'25. This breakpoint (CT4) may be related to the rapid decrease of protein synthesis between CT0 and CT6. Treatment with cycloheximide and anisomycin strongly inhibited the circadian changes of leucine incorporation, but did not affect 2DG uptake. This result strongly suggests that circadian changes of leucine incorporation by the SCN may reflect its clock function or some related function. It is not clear whether the phase shifts induced by the microinjections of protein synthesis inhibitors into the SCN are due to a general inhibition of protein synthesis or to the inhibition of specific protein(s) that may be related to SCN oscillatory functions. The effects of cycloheximide and anisomycin on the incorporation of leucine were dependent upon the circadian time (CT), since the reduction of leucine incorporation by these chemicals was more evident at CT0 than at CT9. In addition, the phase change of hamster activity induced by inhibitors of protein synthesis depends upon the time of injection6'25. These findings suggest that circadian changes of leucine incorporation may be tied to a component of the SCN circadian clock mechanism rather than the result of some non-specific events. Whether leucine incorporation is a direct component of the pacemaker mechanism or just another 'hand of clock' remains to be elucidated. TTX treatment inhibited 2DG uptake, but did not affect leucine incorporation. Recent studies have suggested that TTX blocks the appearance of the overt rhythm, vasopressin release 2, and locomotor activity2°, but does not change the circadian clock mechanism itself. Therefore, the present result is consistent with these above-mentioned reports. The present findings taken together with previous reports have suggested that circadian changes of 2DG uptake or firing activity may be related to the overt rhythm rather than the clock mechanism. In previous in vivo experiments, despite the observed robust circadian rhythms of electrical activity and energy metabolism 5A3,19,23, the overall protein synthesis in the SCN appeared to remain constant 4A8. Therefore, the data presented here disagree with these previous in vivo studies. The reason for the difference is uncertain, but there are three possible explanations: (i) the in vivo data are incorrect; (ii) the in vitro situation unmasks a rhythm not evident in vivo; and (iii) the in vitro situation induces a rhythm not present in vivo. It is not possible to distinguish between these
possibilities with the information currently available. Regarding possibility (i); although there were no differences at the two time points sampled in vivo, it remains possible that the use of other sampling times may yet disclose a rhythm like that demonstrated in the present results. Possibility (ii) may have arisen due to differences of experimental conditions. We previously reported that 2DG uptake by the golden hamster SCN appeared to show a more robust rhythm in vitro 22 than in vivo 21. Therefore, culture for 24 h may unmask a rhythm not evident in vivo. At present, the possibility that option (iii) is correct remains unknown. In summary, we elucidated a clear circadian change of in vitro leucine incorporation by the SCN suggesting that our method may be useful for studying the circadian clock function of the SCN. REFERENCES 1 Deutsch, J.A., Physiological Basis of Memory, 2nd edn., Academic Press, New York, 1983. 2 Earnest, D.J., Digiorgio, S.M. and Sladek, C.D., Effects of tetrodotoxin on the circadian pacemaker mechanism in suprachiasmatic explants in vitro, Brain Res. Bull., 26 (1991) 677-680. 3 Feldman, J.F., Lengthening the period of a biological clock in Euglena by cycloheximide, an inhibitor of protein synthesis., Proc. Natl. Acad. Sci. USA, 57 (1967) 1080-1087. 4 GIotzbach, S.F., Randall, T.L., Radeke, C.M. and Heller, H.C., Absence of a circadian rhythm of protein synthesis in the rat suprachiasmatic nucleus, Neurosci. Left., 76 (1987) 113-118. 5 Inouye, S.I. and Kawamura, H., Characteristics of a circadian pacemaker in the suprachiasmatic nucleus, J. Comp. Physiol., 146 (1982) 153-160. 6 Inouye, S.I., Takahashi, J., Wollnik, F. and Turek, F.W., Inhibitor of protein synthesis phase shifts a circadian pacemaker in mammalian SCN, Am. J. Physiol., 255 (1988) R1055-R1058. 7 Jacklet, J.W., Neuronal circadian rhythms; phase-shifting by a protein synthesis inhibitor, Science, 198 (1977) 69-71. 8 Karakashian, M.W. and Schweiger, H.G., Evidence for a cycloheximide-sensitive component in the biological clock of Acetabularia, Exp. Cell Res., 98 (1976) 303-312. 9 Lotshaw, D.P. and Jacklet, J.W., Involvement of protein synthesis in circadian clock of Aplysia eye, Am Z Physiol., 250 (1986) R5-R1. 10 Murakami, N. and Takahashi, K., Cicadian rhythm of adenosine 3',5'-monophosphate content in suprachiasmatic nucleus (SCN) and ventromedial hypothalamus (VMH) in the rat, Brain Res., 276 (1983) 297-304. 11 Nakashima, H., Perlman, J. and Feldman, J.F., Cycleheximide-induced phase-shifting of circadian clock of Neurospora, Am. J. Physiol., R31-R35. 12 Newman, G.C., Hospod, F.E. and Patlak, C.S., Kinetic model of 2-deoxygluxose metabolism using brain slices, Z Cereb. Blood Flow Metab., 10 (1990) 510-526. 13 Newman, G.C. and Hospod, F.E., Rhythm of suprachiasmatic nucleus 2-deoxyglucose uptake in vitro, Brain Res., 381 (1986) 345-350. 14 Raley-Susman, K.M. and Lipton, P., In vitro ischemia and protein synthesis in the rat hippocampal slice: the role of calcium and NMDA receptor activation, Brain Res., 515 (1990) 27-35. 15 Roberts, M.H. and Moore, R.Y., Diurnal changes of protein synthesis in the suprachiasmatic nucleus of the rat, Soc. Neurosci. Abstr., 12 (1986) 211. 16 Rothman, B.S. and Strumwasser, F., Phase-shifting the circadian rhythm of neural activity in the isolated Aplysia eye with
256
17 18
19
20
21
22
puromycin and cycloheximide: electrophysical and biochemical studies, ,L Gen. Physiol., 68 (1976) 359-384. Rusak, B. and Bina, K.G., Neurotransmitters in the mammalian circadian system, Annu. Rez~'. Neurosci., 13 (1990) 387-401. Scammell, T.E., Schwartz, W.J. and Smith, C.B., No evidence for a circadian rhythm of protein sythesis in the rat suprachiasmatic nuclei, Brain Res., 494 (1989) 155-158. Schwartz, W.J., Davidsen, L.C. and Smith, C.B., In vivo metabolic activity of a putative circadian oscillator, the rat suprachiasmatic nucleus, J. Comp. Neuro/., 189 (1980) 157-167. Schwartz, W.J., Gross, R.A., Morton, M.T., The suprachiasmatic nuclei contain a tetrodotoxin-resistant circadian pacemaker, Proc. NatL Acad. Sci. USA, 84 (1987) 1694-1698. Schwartz, W.J., Different in vivo metabolic activities of suprachiasmatic nuclei of Turkish and golden hamsters, Am. J. Physiol., (Regulatory, integrative and comparative physiology), 28 (1990) R1083-R1085. Shibata, S. and Moore, R.Y., Electrical and metabolic activity of suprachiasmatic nucleus neurons in hamster hypothalamic slices., Brain Res. 438 (1988) 374-378.
23 Shibata, S., Oomura, Y., Kita, H. and Hattori, K., Circadian rhythmic changes of neuronal activity in the suprachiasmatic nucleus of the rat hypothalamic slice, Brain Res., 247 (1982) 154-158. 24 Shinohara, K., Isobe, Y., Takeuchi, J. and Inouye, S.i., Circadian rhythms of somatostatin-immunoreactivity in the suprachiasmatic nucleus of the rat, Neurosci. Lett., 127 (1991) 9-12. 25 Takahashi, J.S and Turek, F.W., Anisomycin, an inhibitor of protein synthesis, perturbs the phase of a mammalian circadian pacemaker, Brain Res., 405 (1987) 199-203. 26 Taylor, W.R., Dunlap, J.C. and Hastings, J.W., Inhibitors of protein synthesis on 80 S ribosomes phase shift the Gonyaulax clock, J. Exp. Biol., 97 (1982) 121-126. 27 Van den Pol, A.M., Amino acid incorporation in medial hypothalamic nuclei; circadian, perikarya, and neuropil variations, Am. ,L Physiol., 240 (1981) RI6-R22. 28 Vornov, J.J. and Coyle, J.T., Glutamate neurotoxicity and inhibition of protein synthesis in the hippocampal slice, J. Neurochem., 56 (1991) 996-1006.
