Temporospatial characteristics of light-induced Fos immunoreactivity in suprachiasmatic nuclei are not modified in Syrian hamsters treated neonatally with monosodium glutamate

Brain Research 808 Ž1998. 250–261

Research report

Temporospatial characteristics of light-induced Fos immunoreactivity in suprachiasmatic nuclei are not modified in Syrian hamsters treated neonatally with monosodium glutamate I. Chambille

)

Laboratoire de Physiologie Sensorielle, Institut National de la Recherche Agronomique (INRA), CRJ-78352, Jouy en Josas cedex, France Accepted 11 August 1998

Abstract Neonatal treatment of rodents by intraperitoneal injections of monosodium glutamate ŽMSG. destroys many retinal ganglion cells whose neurons belong to the circadian system; howertheless, adults always synchronize their locomotor activity rhythm ŽLAR. to the lightrdark cycle. Recent studies have shown that light-induced phase shifts of LAR are associated with the c-fos induction in suprachiasmatic nuclei ŽSCN. of nocturnal rodents. In this study, the circadian system was analyzed in treated and control hamsters maintained in constant darkness and exposed to light at circadian times ŽCTs. 13 and 18 during subjective night, 1 and 6 h after the onset of LAR. The period of the LAR and delay ŽCT13. and advance ŽCT18. phase shifts of LAR were not significantly different between MSG-treated and control hamsters. Temporospatial variations of Fos induction after light exposure were similar in both MSG-treated and control hamsters although the total number of Fos immunoreactive ŽFos-ir. nuclei in the SCN was always lower in treated hamsters. However, the decrease in Fos-ir was significant only for the caudal third of the SCN of treated hamsters, the part where retinal afferents are most dense. The effect of light exposure on Fos expression in SCN of MSG-treated and control hamsters was the same at CT13 and CT18: Ž1. Fos-ir nuclei were significantly more numerous at CT18 than at CT13 in the rostral SCN; Ž2. dorsal Fos-ir cells were observed in the SCN only at CT18; Ž3. a ventral subgroup expressed Fos protein in intermediate SCN only at CT13. This study demonstrates that MSG-treatment does not significantly modify the phase-shifting effects of light on either the LAR or the associated cellular activation. q 1998 Elsevier Science B.V. All rights reserved. Keywords: Circadian clock; Locomotor activity rhythm; Phase shifts; Fos protein; Circadian times; Subjective night

1. Introduction The suprachiasmatic nuclei ŽSCN. of the anterior hypothalamus of mammals control many endogenous physiological and behavioral rhythms w31x that persist with ‘tau’ Žt . periods of approximately 24 h in the absence of periodic environmental time cues. For many rhythms such as locomotor activity ŽLAR., NSC are the pacemaker whose ablation or complete destruction disrupts circadian rhythmicity w12,39,47x. This central oscillator is reset to the stable 24 h period of external lightrdarkness ŽLD. cycle by photic information derived from the retina. Photic inputs are conveyed to the SCN via a direct projection, the retinohypothalamic tractus ŽRHT. w23,34,38x, and via an indirect projection by the geniculohypothalamic pathway ŽGHT.. This projection originates from the intergeniculate )

Fax: q33-1-34-65-21-02; E-mail: [email protected]

leaflet ŽIGL., a subdivision of the lateral geniculate nucleus ŽLGN. w3,35x. The RHT w24x and GHT w13,19,25,41x both contribute—by complex interactions—to producing normal entrainment of the pacemaker to natural lightrdark cycles. Since Rea’s report w43x, many studies have demonstrated the involvement of the c-fos immediate early gene in the photic entrainment of the rodent circadian rhythms w2,15,28,48,49,51x. C-fos expression is rapidly and transiently stimulated in cells of the SCN during the subjective night of an animal kept in constant darkness ŽDD. but only when light is able to phase-shift LAR w7,44,53x. Photic induction of the gene and photic expression of the Fos protein in the SCN are circadian phase-dependent, as the maximal response is obtained for a stimulus delivered during mid to late subjective night w7,22,28,53x. In the retina, c-fos induction is also regulated by light w27,50,57x and the maximal expression of Fos protein in the ganglion

0006-8993r98r$ - see front matter q 1998 Elsevier Science B.V. All rights reserved. PII: S 0 0 0 6 - 8 9 9 3 Ž 9 8 . 0 0 8 3 1 - 2

I. Chambiller Brain Research 808 (1998) 250–261

cell layer ŽGCL. is also observed after light stimulation applied in the middle of the subjective night w5,7x. Fos induction provides a useful tool to study the effects of light on activation of the components of the circadian system in both the retina and SCN. In this study, it was therefore used to functionally investigate the photic entrainment of a hamster model with retinal degeneration due to monosodium glutamate ŽMSG. neurotoxic treatment. It has been previously demonstrated that glutamate, or its by-product MSG, given to neonate rodents, induced acute degeneration of the retina, optic nerve, visual pathways and some areas of the brain, such as the arcuate nuclei w8,30,37x. Physiological and behavioral abnormalities have been reported in these MSG-treated animals, such as obesity, sterility w14,29,54x, loss of visual placing response w26x and deficiency in discrimination learning in a maze experiment w42x. However, many of their circadian rhythms are still entrained by light w6,33,40,45x despite a 37% loss of the retinal GCs specifically responding to photic stimuli w5x and despite a reduction of retinal projections onto the targets of the circadian system, such as the SCN, IGL and ventral part of the lateral geniculate nuclei ŽvLGN. w6,40x. The aim of the present study was to gain information on the functioning of the circadian system in MSG-treated hamsters. This analysis investigated: Ž1. the autonomous course of the circadian clock, reflected by the LAR using the ‘tau’ Žt . period as the measurement criterion, Ž2. the clock resetting capacity, tested by the phase shifts induced by light stimuli at two critical times of the circadian cycle of animals maintained in DD and measured by the amplitude of phase delay or advance of LAR, Ž3. the photic activation of SCN measured by light-induced Fos immunoreactivity in the various populations of SCN neurons at the same CTs as those used for behavioral analyses. The results show that the period and amplitude of LAR phase shifts were not significantly altered by treatment. Fos immunoreactivity induced during subjective night was quantitatively but not qualitatively modified since MSGtreated hamsters kept the same temporospatial patterns of Fos immunoreactive cells ŽFos-ir. in SCN as those of normal hamsters. Part of these results were presented at the ‘troisieme ` colloque de la Societe ´ ´ des Neurosciences’ held in Bordeaux, France w4x.

251

transferred with their mother into ventilated boxes under controlled lighting conditions consisting of a 12:12 LD cycle. Food and water were given ad libitum. 2.2. MSG-treatment Treatment was described in a previous paper w6x. Briefly, 5-day-old males received a first intraperitoneal injection Ži.p.. of 5 mgrg of body weight of MSG ŽMerck; Darmstadt; Germany. dissolved in 0.1 ml of saline, followed by a daily injection increasing from 5 to 8 mgrg from D6 to D 10 . Control males Žone per litter. received daily injections of saline from D5 to D 10 or were not injected. 2.3. Locomotor actiÕity and light stimulation Two- to 3-month-old control and MSG-treated males were individually housed in cages equipped with running wheels with one cage per box. Hamsters were maintained under the same lighting schedule as before for at least 2 weeks and were then allowed to free-run under constant darkness ŽDD. for another 2 weeks. Only hamsters generating a robust free-running LAR and a clear onset of the active phase were used. Two groups of control animals and two groups of MSG-treated hamsters were constituted for behavioral and Fos expression experiments. Light stimulation consisted of sequences of 30 flashes Ž200 ms each. at a frequency of six per min, as described and discussed previously w7x. Phase-shift effects of this stimulus were studied in the first group of control and MSG-treated hamsters at two circadian times ŽCTs. of the subjective night, i.e., at CT13 Ž1 h after the onset of LAR; by convention CT12 corresponds to the locomotor activity onset. and in the middle of the subjective night, at CT18. These two CTs were chosen because they are known to cause maximal phase delay and advance of LAR, and intense Fos-expression in the retina and SCN. The first light stimulation was delivered either at CT13 or CT18; 15 days later a second stimulation was applied in reverse order. Wheel-running activity was continuously monitored using the DATAQUEST III computerized data acquisition system ŽMini Mitter, Sunriver, OR, USA.. Counts of wheel revolutions were stored in 10 min bins. 2.4. Analysis of circadian locomotor actiÕity

2. Materials and methods 2.1. Animals All animal experiments were carried out in accordance with the European Communities Council Directive of November 24, 1986 Ž86r609rEEC.. The hamsters were born in the laboratory colony maintained under an LD cycle of 16:8 at room temperature of 23 " 28C. On the day of birth ŽD1 ., newborn males were

The free-running Žt . period and the magnitude of phase shifts elicited by light stimulation were analyzed with Tau software ŽMini Mitter, Sunriver, OR, USA. according to the analytical procedures of Daan and Pittendrigh w9x. The free-running period was defined by the linear regression of onsets of activity over a 10- to 20-day period of constant darkness prior to the light stimulation experiments. To measure the phase shift induced by each light stimulation, a line was fitted visually through the activity onsets ŽCT12.

I. Chambiller Brain Research 808 (1998) 250–261

252

Table 1 Periods of the locomotor activity rhythm and amplitudes of phase shifts induced by light during the subjective night at the circadian times ŽCTs. 13 or 18 in control and neonatally monosodium glutamate ŽMSG.-treated hamsters Hamsters

Periods Žh.

Phase shifts at CT13 Žmin.

Phase shifts at CT18 Žmin.

Control mean" 24.036"0.015 Ž16. y89"9 Ž8. q130"13 Ž12. S.E.M. Ž n. MSG-treated 24.10"0.008 Ž16. y81"8 Ž14. q118"12 Ž13. mean" S.E.M. There were no significant differences in periods, delay ŽCT13. and advance ŽCT18. of phase shifts between control and MSG-treated hamsters. ns Number of individuals.

for about 15 days before light stimulation and extrapolated to predict the onset of activity on the day following the flash. Likewise, a second line was fitted through the activity onsets for 7–10 days of stable rhythmicity following the light stimulation. The 2 to 4 days immediately after stimulation were generally not considered because of the existence of transient cycles. The amplitude of the phase shifts was measured on the day after stimulation by the horizontal difference between these two lines. Phase advances were plotted as ‘q’ and phase delays as ‘y’. 2.5. Retinal projections onto SCN and Fos immunohistochemistry procedure 2.5.1. Retinal projections Hamsters under light anesthesia with pentobarbital ŽSanofi, Libourne, France; 12 mgr100 g body weight, i.p.. received a 6-ml injection of the b subunit of cholera toxin, Ctb ŽSigma, France. in the posterior chamber of one eye. Two days later, animals were deeply anesthetized and perfused transcardially with 200 ml of warm saline solution containing 1% sodium nitrite, followed by 300 ml of 4% paraformaldehyde in 0.1 M phosphate buffer, pH s 7.4 ŽPB.. Brains were removed immediately and post-fixed for 4–6 h in the same fixative at 48C before being cryoprotected successively in 10–30% sucrose PB solutions. Serial coronal sections were cut at y208C Ž60 mm. and incubated in goat Ctb antiserum ŽList biological, Interchim, France; 1:8000. for 24 h at room temperature, washed several times before incubation overnight at 48C with

biotinylated donkey anti-goat antiserum ŽJackson Immunoresearch, USA; 1:2000.. Sections were then incubated with streptavidin–horseradish peroxidase Žstreptavidin–HRP, Boehringer, Mannheim, France; 1:1000.. HRP products were visualized with 0.03% 3,3X-diaminobenzidine tetrahydrochloride ŽDAB.. 2.5.2. Fos-immunoreactiÕity Immunohistochemistry of Fos expression was studied in control and MSG-treated hamsters of the second group 1 h after the light stimulation at CT13 or CT18. The animals were anesthetized under dim red illumination and perfused as above. Free floating sections Ž60 mm. were collected in phosphate buffered saline pH s 7.4, PBS ŽUnipath, Basingstoke, England., incubated for 30 min at room temperature in 0.1% hydrogen peroxide, carefully washed, then incubated in a blocking solution of 10% normal goat serum in PBS with 0.3% Triton X-100 ŽNGPBST.. They were then incubated at 48C for 48–72 h in the anti-c-fos primary antibody diluted to 1:20,000 in 1% NGPBST ŽOncogene Science; NY, USA., rinsed three times in 1% NGPBST and incubated in biotinylated goat anti-rabbit IgG diluted 1:500 ŽBiosys, France. in 1% NGPBST, overnight. After several rinses in PBS, sections were incubated for 1 h in an ABC reagent Vectastain kit ŽVector, Biosys, France. then for 5 to 10 min in a chromogen solution of 0.02% DAB, 0.3% nickel ammonium sulfate in PBS and 0.035% hydrogen peroxide. Sections were then mounted on gelatin, dehydrated and coverslipped in Depex. Specificity of staining was controlled by omitting the primary antibody; these controls produced no immunoreactivity. 2.6. Data collection and analysis Mapping of Fos-immunoreactive ŽFos-ir. nuclei in the SCN region was performed using an image analysis system ŽHisto 2000 from Biocom, France. linked to a microscope equipped with an X–Y stage recording device. Three categories of Fos-ir nuclei were determined as previously reported w7x, but only uniformly dark-stained nuclei and pycnotic dark-grey nuclei, well above the basal background staining of surrounding tissue, were mapped and used for quantitative results. For each brain, all serial sections including the SCN were analyzed. After counter-

Table 2 Mean volume of the two suprachiasmatic nuclei ŽSCN., number of cells and total Fos-immunoreactivity induced in SCN after light stimulation delivered at circadian times ŽCTs. 13 and 18 in control and MSG-treated hamsters Total volume Žmm3 . mean " S.E.M. Control hamsters MSG-treated hamsters a

vs. e , P - 0.01;

b

Total number of cells mean " S.E.M.

a

b

0.1422 " 0.007 0.1119 " 0.007 e

vs. f , P - 0.03;

c,g

vs.

d,h

, P s 0.055;

22,956 " 715 18,348 " 418 f c,d

vs.

g,h

, P s 0.055.

Total number of Fos-ir nuclei mean " S.E.M. CT13

CT18 c

1918 " 237 1603 " 127 g

2425 " 253 d 1928 " 171h

I. Chambiller Brain Research 808 (1998) 250–261

253

Fig. 1. Representative photomicrographs of coronal sections Ž60 mm. illustrating the overlap between retinal afferent terminals and Fos-immunoreactivity ŽFos-ir. at similar rostro-caudal levels of suprachiasmatic nuclei ŽSCN.: ŽA. – ŽH. s control hamsters; ŽI. – ŽP. s MSG-treated hamsters. Retinal projections into SCN ŽA–D and I–L. were identified after intraocular injection of a Cholera toxin b-subunit. Fos-ir was induced by a light stimulation delivered at CT18. ŽA.,ŽE.,ŽI.,ŽM. s rostral level of SCN; ŽB.,ŽC.,ŽF.,ŽG.,ŽJ.,ŽK.,ŽN.,ŽO. s two intermediate levels; ŽD.,ŽH.,ŽL.,ŽP.: caudal SCN. Scale bar: 200 mm.

254

I. Chambiller Brain Research 808 (1998) 250–261

staining with neutral red, the SCN in each section were delineated and their area was measured. The sum of all measures was used to determine the total number of Fos-ir nuclei included in the total volume of the SCN, this volume being the total area of SCN multiplied by the thickness of the section. The total number of cells in the SCN was also estimated for control and MSG-treated hamster brains. After counterstaining of sections, all cells located inside SCN bound-

aries, whether or not they expressed Fos-ir, were counted in alternate sections of four brains of each group. The number of sampled cells was then expressed for each animal taking into account the volume of its SCN. 2.7. Statistical analyses Student’s t test was used for comparison between control and MSG-treated hamsters of Ž1. mean period of LAR,

Fig. 2. Rostro-caudal distribution of Fos-immunoreactive nuclei counted in all SCN serial coronal sections Ž60 mm. from control and MSG-treated hamsters receiving a light-stimulation at CTs 13 and 18.

I. Chambiller Brain Research 808 (1998) 250–261

255

Ž2. amplitude of phase-shifts induced by light at CT13 and CT18, Ž3. total volume of SCN and Ž4. total number of cells in SCN. The total number of Fos-ir nuclei included in the entire SCN were used for statistical analyses. Two-way analysis of variance ŽANOVA, generalized linear model, Systat for Windows, version 5, Evanston, IL 1992. was performed to test the effects of MSG treatment and the CTs and the interaction on the total expression of the Fos protein in the SCN. Fos-ir nuclei counted in rostral Žfirst four sections; 240 mm., intermediate Žfollowing four sections; 240 mm. and posterior Žlast two to four sections, 120 to 240 mm. SCN of control and MSG-treated hamsters were also compared using two-way ANOVA.

In rostral SCN of control hamsters, Ctb- and Fos-immunoreactivities were confined to a round region above the optic chiasma ŽFig. 1A, E.. In the middle part of SCN, labeling extended to the dorsal and latero-external boundaries of SCN ŽFig. 1B–C, F–G.; a dense network of retinal fibers and a high concentration of Fos-ir nuclei invaded the ventro-dorso-lateral region of caudal SCN ŽFig. 1D, H.. As shown in Fig. 1 ŽI–L., the ventro-dorsal and rostro-caudal aspects of the spatial distribution of retinal projections in MSG-treated hamsters did not appear to be markedly altered. In parallel, the pattern of Fos-immunoreactivity did not vary between the two groups ŽFig. 1E–H, M–P.. A good overlap between retinal afferents and Fos-immunoreactivity induced at CT18 was therefore observed in all hamsters throughout the SCN rostro-caudal axis.

3. Results

3.3. Fos-expression in the SCN

3.1. Effects of neonatal MSG treatment on the period of the LAR and light-induced phase shifts The t period of the LAR was calculated in 16 control and 16 MSG-treated hamsters free-running in DD ŽTable 1.. In control hamsters, t was 24.036 " 0.015 h; in MSGtreated hamsters, t was 24.10 " 0.008 h ŽTable 1.. MSGtreatment therefore did not significantly modify the period of the LAR of MSG-treated hamsters. Likewise, when light stimulation was delivered at CT13, the mean delay of the active phase of the LAR was 1.484 h for control hamsters and 1.351 h for MSG-treated hamsters ŽTable 1.; when light was applied at CT18, the advance phase shifts were 2.174 and 1.978 h in control and MSG-treated hamsters respectively. In the two series, animals exhibited phase shifts in the same direction for the same CT which was opposite for CT13 and CT18 and whose amplitudes were always larger at CT18 than at CT13. The amplitude of phase shift in MSG-treated hamsters was 8 min shorter at CT13 and 12 min shorter at CT18. Differences between control and MSG-treated hamsters were never significant ŽTable 1.. 3.2. MSG treatment and SCN 3.2.1. SCN and retinal afferents MSG treatment significantly reduced the mean volume of the SCN by 21% Ž P - 0.01, Table 2.; these changes occurred in three dimensions.

3.3.1. QuantitatiÕe analysis In control hamsters, the mean number of Fos-ir nuclei was higher after light stimulation delivered at CT18 Ž n s 4. than after light stimulation at CT13 Ž n s 4; Table 2.. Similar results were observed in MSG-treated hamsters, Ž n s 5 at CT13 and n s 4 at CT18. but, for each CT, the mean number of Fos-ir nuclei was lower than that of the control hamster group ŽTable 2.. Differences in Fos expression between CT13 and CT18 were 21% in control hamsters and 17% in MSG-treated hamsters. Differences of Fos-immunoreactivity between the two groups of animals were 16% and 20% for light stimulations delivered at CT13 and CT18 respectively. No significant differences in total Fos-immunoreactivities were detected between control and MSG-treated hamsters or between CTs for control or MSG-treated hamsters, since the limit of significance of comparisons was always equal to 0.055 ŽANOVA analysis.. In contrast, we observed a significant 20% reduction of the total number of cells in the SCN of MSG-treated hamsters ŽTable 2.. In control hamsters, the distribution of Fos-ir nuclei varied all along the SCN rostro-caudal axis and differed according to the CT of light stimulation. As shown in Fig. 2A for a light stimulation at CT18, the rostro-caudal progression of Fos-ir nuclei showed a sinusoidal wave characterized by non negligible values in the rostral SCN, very high means in the caudal third and a weak Fos expression in some sections of the intermediate SCN. In contrast, at CT13 Fos-immunoreactivity increased progres-

Table 3 Fos-immunoreactivity induced by a flash at CT13 or CT18 in rostral, intermediate and caudal parts of SCN in control and MSG-treated hamsters Rostral SCN CT 13 Control hamsters mean " S.E.M. MSG-treated hamsters mean " S.E.M.

65 " 8

Intermediate SCN CT 18

a

54 " 13 a

164 " 32

CT 13 b

107 " 17 b

176 " 59

Caudal SCN CT 18

c

231 " 34 c

For each part of SCN, values with different superscripts differ significantly Ž P - 0.05..

158 " 32

CT 13 c

180 " 32 c

CT 18 d

292 " 51

356 " 67 d

216 " 23 e

243 " 32 e

256 I. Chambiller Brain Research 808 (1998) 250–261 Fig. 3. Shematic illustration of strong and weak levels of Fos induction in SCN of control hamsters exposed to light stimulations delivered at CTs 13 and 18. Hamsters showing the smallest and highest responses at each CT were chosen to illustrate the great individual variabilites.

I. Chambiller Brain Research 808 (1998) 250–261 257

Fig. 4. Spatio-temporal variations of photic induction of Fos protein observed in SCN of control and MSG-treated hamsters submitted to light stimulations at CT13 or CT18.

258

I. Chambiller Brain Research 808 (1998) 250–261

sively from a low level in the first sections of SCN to maximal values at the beginning of the caudal third; these high levels always remained lower than those reached for light stimulation at CT18 ŽFig. 2A.. In MSG-treated hamsters, the rostro-caudal distribution of the mean number of Fos-ir nuclei per section showed similar patterns to those observed in control hamsters in response to the same light stimulation ŽFig. 2B.. On the basis of these similar patterns of Fos-immunoreactivity in control and MSG-treated hamsters, we divided the SCN of each animal into three rostro-caudal regions ranging over 240 mm for the first two parts, but sometimes less for the terminal part. Results reported in Table 3 show that: Ž1. at CT13 and CT18, significantly more Fos-ir nuclei were detected in the caudal SCN of control hamsters than in that of MSG-treated animals Ž P - 0.05., Ž2. the mean numbers of nuclei expressing Fos protein in the rostral third of the SCN in control and MSG-treated hamsters were significantly higher for light stimulation delivered at CT18 than at CT13 Ž P - 0.001., Ž3. in the two groups of animals, the Fos immunoreactivity was slightly higher at CT13 in the intermediate part of the SCN than at CT18, but the differences were not significant.

3.3.2. Temporospatial distribution of Fos-ir nuclei in the SCN of control and MSG-treated hamsters The temporospatial distributions of Fos-ir nuclei in control hamster SCN are illustrated in Fig. 3. As previously described, rostro-caudal patterns of Fos immunoreactivity differed between the two CTs. In the rostral third Žthe first four sections. at CT13, only a few Fos-ir cells were present and they were scattered inside the SCN ŽFig. 3A.. At CT18, Fos-ir cells were very numerous and many of them were located along the dorsal SCN delineating a crescent-shaped zone extending into the surrounding hypothalamus ŽFig. 3B.. In the intermediate third of the SCN, at CT18 Žthe following four sections., the dorsal crescent of Fos-ir nuclei decreased in the two anterior sections whereas a ventrally located cluster of labeled nuclei appeared in the two posterior sections ŽFig. 3B.. These ventral Fos-ir nuclei were more numerous at CT13. They formed a denser cluster, visible more rostrally than at CT18 ŽFig. 3A.. In the caudal part Žthree to four sections according to brains. at CT18, Ž1. numerous Fos-ir nuclei reappeared near the dorsal boundary of SCN, Ž2. a ventral cluster increased in size and appeared to be as consistent and as large as the ventral group observed at CT13 ŽFig. 3A, B., Ž3. nuclei expressing Fos protein were found along the latero-external margin of the nucleus merging into the ventral group and dorsal cells ŽFig. 3B.. In these experiments, we observed a marked individual variability despite homogeneous intergroup variances. The temporospatial patterns of Fos immunoreactivity, specific for each CT, remained similar regardless of the intensity of the light response Žlow response, Fig. 3C, D; high re-

sponse, Fig. 3A, B.. For each CT, the reduction in the number of Fos-ir nuclei was particularly marked in the intermediate third of the SCN. Nevertheless, the anterior subpopulation of the ventral group, only detected at CT13 in this part, remained visible in animals expressing a weak Fos-immunoreactivity. As in control hamsters, temporospatial distributions of Fos-ir nuclei were similar between MSG-treated animals, showing an intense Žhigh level of Fos immunoreactivity. or moderate Žweak level of Fos immunoreactivity. response to light Žnot shown.. In Fig. 4, C and D represent the spatial distribution of light-induced Fos-immunoreactivities in MSG-treated hamster SCN at CT13 and CT18. In treated animals, Fos expression was observed with the same temporospatial characteristics as those described in control hamsters with even more marked differences between CTs. Thus, in hamsters treated with MSG, Ž1. the cluster of Fos-ir nuclei, belonging to the ventral group, appeared more rostrally in the intermediate part of the SCN when animals were stimulated at CT13 than at CT18, Ž2. many more Fos-ir nuclei were observed in the rostral third at CT18, and Ž3. the latero-dorsal group was always present in the caudal part of the SCN in animals stimulated at CT18.

4. Discussion This investigation provides additional evidence of a good preservation of the fundamental properties of the circadian system in hamsters neonatally treated with MSG. Our results are in good agreement with those of previous studies reporting Ž1. the persistence for several weeks or months under conditions of constant darkness, of a clear and stable free-running locomotor activity circadian rhythm in MSG-treated hamsters, Ž2. an entrainment by light and a good resynchronization of locomotor activity after a 6-h phase shift of the environmental lightrdark cycle w6,40x. Our results show that the circadian clock and its capacities for light-induced phase shifting were not affected by MSG-treatment since the period and amplitude of phase shift delay or advance of the circadian LAR did not differ between MSG-treated and control hamsters. Neonatal MSG-treatment did not prevent c-fos induction in the SCN. However, Fos expression was attenuated throughout the SCN and was significantly reduced in the caudal part of the nucleus where the density of retinal projections was maximal. In MSG-treated hamsters, these projections were observed throughout the SCN with a similar pattern to that observed in control hamsters. The volume of the SCN and its cell population were significantly reduced by 20%. Counterstaining with neutral red to delineate structures revealed all cell types in the SCN, so it is impossible to determine whether MSG-treatment destroyed only glia, nervous cells or both cell types. How-

I. Chambiller Brain Research 808 (1998) 250–261

ever, as an excess of glutamate is toxic for mammalian neurons, I think that many neurons in the SCN were probably killed by neonatal MSG-treatment as reported in the rat w33,52x. Other studies using specific stains for nervous tissue and electron microscopy will be necessary to further investigate the excitotoxicity of MSG-treatment in SCN Žfor example, a possible effect of MSG on cell size.. Cell death is not the explanation only for the reduction of the volume of the SCN, as MSG also exerts a toxic effect on retinal afferents, as previous studies in MSGtreated hamsters have demonstrated, Ž1. a 37% loss of the population of Fos-ir retinal ganglion cells w5x, Ž2. a decrease in the volume occupied by terminals of the retinohypothalamic tract in the SCN w6x. A maximum of 3500 ganglion cells express the Fos protein in the retinal GCL in control hamsters after light stimulation in the middle of the subjective night vs. 2200 in MSG-treated hamsters w5x. In control animals, these cells represent the maximal population conveying input to the three targets ŽSCN, IGL, vLGN. involved in the photic entrainment of circadian rhythms. In rodents, Fos protein is expressed in these three targets in response to light pulses applied during the subjective night in animals kept either in darkness ŽDD. w43,48x, in constant illumination ŽLL. or in a light phase of an LD cycle w16,17x. The decrease in Fos immunoreactivity observed in the IGL and vLGN of MSG-treated rats w18x and in the SCN of treated hamsters Žour study. could therefore be at least partially due to loss of photic information derived from retinal ganglion cells. As the decrease of Fos-ir cells was observed throughout the SCN in MSGtreated hamsters, we can hypothesize that Ž1. retinal afferents sensitive to treatment could project onto the entire nuclei and Ž2. Fos-ir cells sensitive to MSG could also be located throughout the SCN. In MSG-treated hamsters, the maximal number of nuclei expressing Fos protein in retinal GCL was the same at CT13 and CT18 w5x which suggests that the same fraction of Fos-ir GCs was excited at both CTs of their subjective night. However, the spatial patterns of Fos expression in the SCN were similar to those of control animals, i.e., clearly phase-dependent. These observations are in favor of a concept, now admitted in the literature, that photic induction of c-fos gene in SCN is gated by the circadian oscillator itself. But, as the number of nuclei expressing the Fos protein in the retinal GCL in control hamsters fluctuates in phase with circadian responses of SCN Fos immunoreactivity w7x, it may be suggested that the retinal oscillator w56x is also controlled by the SCN. The photic induction of Fos protein in the SCN of rodents is characterized by its phase-dependency. An anatomical specificity of Fos-ir cell distribution in SCN was reported for the first time by Rea w44x in free-running hamsters. In a previous study conducted in the same species, we also demonstrated differences in Fos expression of four cellular subgroups according to the CT of light stimulation, i.e., rostral group, cells forming the dorsal

259

crescent, cells of ventral and latero-external groups of the caudal parts of the SCN w7x. This study demonstrates that the response of each of these four subgroups was similar in control and MSG-treated hamsters. Spatial patterns of Fos expression were maintained in MSG-treated hamsters although the number of Fos-ir cells was reduced throughout the SCN. If the spatial pattern of Fos activation of SCN cells reflects the state of the clock w7,44,55x, the clocks of MSG-treated and control hamsters are in the same state at the same CTs. It is therefore not surprising to observe the same resetting effect of light on the clocks of MSG-treated and control hamsters; and consequently, similar phase shifts of their LAR. The present results show that a light stimulus induced a slightly more Fos-ir nuclei in the entire SCN of MSGtreated and control hamsters at CT18 than at CT13; the differences of Fos expression between the two CTs were highly significant in the rostral part of the SCN. This particularity was already reported in our previous study as well as the specific presence, after light stimulus at CT18, of cells forming the dorsal crescent and latero-external group of the caudal part of the SCN w7x. In the rat SCN, Romijn et al. w46x suggested that at least three different cell groups, characterized by their neuropeptide content play a prominent role during light-induced phase shifts. One of these groups only synthesizes peptide histidine isoleucine ŽPHI.; the other two produce either gastrin releasing peptide ŽGRP. or vasoactive intestinal peptide ŽVIP. alone or in partial colocalization with PHI. As PHI and VIP are synthesized from the same precursor protein, they obviously coexist in the same SCN neuronal cell bodies w10,36x. According to Romijn et al., PHI neurons might be more prominently involved in the light-induced phase delay, i.e., after light stimulation at ZT14. In our study, a small Fos-ir subpopulation in the ventral group was also observed in control and MSG-treated hamsters, but only after light stimulation at CT13. It was located near the optic chiasma at the beginning of the intermediate third of the SCN. In Syrian hamsters as in rats, PHI-immunoreactive perikarya ŽPHI-ir., expressing Fos protein or not w20,32x are located in the ventral region of the intermediate part of the SCN where VIP-ir neurons are also described w1,10,11,21x. As the location of PHI-irrFos-ir neurons was not described in detail by Romijn et al. w46x, some of them might constitute the rostral subpopulation of the ventral group, only visible after light stimulation at CT13.

Acknowledgements This work was supported by an INRA institutional grant. I would like to thank Dr. M. Caillol for her helpful comments on the manuscript, Dr. A. Saul and K. Rerat for her critical reading and S. Venla for her technical assistance.

260

I. Chambiller Brain Research 808 (1998) 250–261

References w1x J. Aıoun, I. Chambille, J. Peytevin, L. Martinet, Neurons containing ¨ gastrin-releasing peptide and vasoactive intestinal polypeptide are involved in the reception of the photic signal in the suprachiasmatic nucleus of the Syrian hamster: an immunocytochemical ultrastructural study, Cell Tissue Res. 291 Ž1998. 239–253. w2x N. Aronin, S.M. Sagar, F.R. Sharp, W.J. Schwartz, Light regulates expression of a Fos-related protein in rat suprachiasmatic nuclei, Proc. Natl. Acad. Sci. USA 87 Ž1990. 5959–5962. w3x J.P. Card, R.Y. Moore, Organization of lateral geniculate hypothalamic connection in the rat, J. Comp. Neurol. 284 Ž1989. 135–147. w4x I. Chambille, Variations spatio-temporelles de l’expression photique de la proteine ´ Fos dans les noyaux suprachiasmatiques ŽNSC. du hamster Syrien: etude comparee ´ ´ chez l’animal normal et un modele ` aux retines regressees, ´ ´ ´ ´ troisieme ` colloque de la Societe ´ ´ des Neurosciences, Bordeaux, France, 1997, p. 95. w5x I. Chambille, Retinal ganglion cells expressing the Fos protein after light stimulation in the syrian hamster are relatively insensitive to neonatal treatment with monosodium glutamate ŽMSG., J. Comp. Neurol. 392 Ž1998. 458–467. w6x I. Chambille, J. Serviere, ` Neurotoxic effects of neonatal injections of monosodium L-Glutamate ŽL-MSG. on the retinal ganglion cell layer of the golden hamster: anatomical and functional consequences on the circadian system, J. Comp. Neurol. 338 Ž1993. 67–82. w7x I. Chambille, S. Doyle, J. Serviere, Photic induction and circadian expression of Fos-like protein. Immunohistochemical study in the retina and suprachiasmatic nuclei of hamster, Brain Res. 612 Ž1993. 138–150. w8x A.I. Cohen, An electron microscopic study of the modification by monosodium glutamate of the retinas of normal and ‘Rodless Mice’, Am. J. Anat. 120 Ž1967. 319–356. w9x S. Daan, C.S. Pittendrigh, A functional analysis of circadian pacemakers in nocturnal rodents: II. The variability of phase response curves, J. Comp. Physiol. 106 Ž1976. 253–266. w10x S. Daikoku, Y. Okamura, H. Kawano, Y. Tsuruo, M. Maegawa, T. Shibasaki, CRF-containing neurons of the rat hypothalamus, Cell Tissue Res. 240 Ž1985. 575–584. w11x S. Daikoku, S. Hisano, Y. Kagotani, Neuronal associations in the rat suprachiasmatic nucleus demonstrated by immunoelectron microscopy, J. Comp. Neurol. 325 Ž1992. 559–571. w12x F.C. Davis, R.A. Gorski, Unilateral lesions of the hamster suprachiasmatic nuclei: evidence for redundant control of circadian rhythm, J. Comp. Physiol. 154 Ž1984. 221–232. w13x M.J. De Vries, A.J. De Vries, Change in the pattern in circadian running-wheel activity after lesions of the intergeniculate leaflet and ventral lateral geniculate nucleus in the Syrian hamster, Biol. Rhythm Res. 26 Ž1995. 331–341. w14x R.S. Donham, K.M. Ogilvie, T.M. Kerner, M.H. Stetson, Daily rhythms of luteinizing hormone and follicle-stimulating hormone persist in female hamsters sterilized by neonatal administration of monosodium glutamate, Biol. Reprod. 43 Ž1990. 392–396. w15x D.J. Earnest, M. Isadarola, H.H. Yel, J.A. Olschowka, Photic regulation of c-fos expression in neural components governing the entrainment of circadian rhythms, Exp. Neurol. 109 Ž1990. 353–361. w16x K. Edelstein, S. Amir, Non-photic manipulations induce expression of Fos protein in the suprachiasmatic nucleus and intergeniculate leaflet in the rat, Brain Res. 690 Ž1995. 254–258. w17x K. Edelstein, S. Amir, Constant light induces persistent Fos expression in rat intergeniculate leaflet, Brain Res. 731 Ž1996. 221–225. w18x K. Edelstein, J.G. Pfaus, B. Rusak, S. Amir, Neonatal monosodium glutamate treatment prevents effects of constant light on circadian temperature rhythms of adult rats, Brain Res. 675 Ž1995. 135–142. w19x M.E. Harrington, B. Rusak, Lesions of the thalamic intergeniculate leaflet alter hamster circadian rhythms, J. Biol. Rhythms 1 Ž1986. 309–325. w20x M.H. Hasting, J.D. Best, F.J.P. Ebling, E.S. Maywood, S. McNulty,

w21x

w22x w23x

w24x

w25x

w26x

w27x w28x

w29x

w30x

w31x w32x

w33x

w34x w35x

w36x

w37x

w38x

w39x

w40x

I. Schurov, D. Selvage, P. Sloper, K.L. Smith, Entrainment of the circadian clock, in: R.M. Buijs, A. Kalsbeek, H.J. Romijn, C.M.A. Pennartz, M. Mirmiran ŽEds.., Progress in Brain Res., Vol. III, Elsevier, 1996, pp. 147–174. S.I.T. Inouye, S. Shibata, Neurochemical organization of circadian rhythm in the suprachiasmatic nucleus, Neurosci. Res. 20 Ž1994. 109–130. N. Ishida, Molecular biological approach to the circadian clock mechanism, Neurosci. Res. 23 Ž1995. 231–240. R.F. Johnson, L.P. Morin, R.Y. Moore, Retinohypothalamic projections in the hamster and rat demonstrated using cholera toxin, Brain Res. 462 Ž1988. 301–312. R.F. Johnson, R.Y. Moore, L.P. Morin, Loss of entrainment and anatomical plasticity after lesions of the hamster retinohypothalamic tract, Brain Res. 460 Ž1988. 297–313. R.F. Johnson, R.Y. Moore, L.P. Morin, Lateral geniculate lesions alter activity rhythms in the hamster, Brain Res. Bull. 22 Ž1989. 411–422. J.S. Kizer, C.B. Nemeroff, W.W. Youngblood, Neurotoxic amino acids and structurally related analogs, Pharmacol. Rev. 29 Ž1978. 301–318. J. Koistinaho, S.M. Sagar, Light-induced c-fos expression in amacrine cells in the rabbit retina, Mol. Brain Res. 24 Ž1995. 53–63. J.M. Kornhauser, D.E. Nelson, K.E. Mayo, J.S. Takahashi, Photic and circadian regulation of c-fos gene expression in the hamster suprachiasmatic nucleus, Neuron 5 Ž1990. 127–134. A. Lamperti, G. Blaha, The effects of neonatally-administered monosodium glutamate on the reproductive system of adult hamsters, Biol. Reprod. 14 Ž1976. 362–369. D.R. Lucas, J.P. Newhouse, The toxic effect of sodiuml L-glutamate on the inner layers of the retina, Arch. Ophthalmol. 58 Ž1957. 193–201. J.H. Meijer, W.J. Rietveld, Neurophysiology of the suprachiasmatic circadian pacemaker in rodents, Physiol. Rev. 69 Ž1989. 671–707. J.D. Mikkelsen, P.J. Larsen, G.G. Sorensen, D. Woldbye, T.G. Bolwig, M.H. Hastings, F.J.P. Ebling, A dual-immunocytochemical method to localize c-fos protein in specific neurons based on their content of neuropeptides and connectivity, Histochemistry 101 Ž1994. 245–251. S. Miyabo, I. Yamamura, E. Ooya, N. Aoyagi, Y. Horikawa, S. Hayashi, Effects of neonatal treatment with monosodium glutamate on circadian locomotor rhythm in the rat, Brain Res. 339 Ž1985. 201–208. R.Y. Moore, N.J. Lenn, A retinohypothalamic projection in the rat, J. Comp. Neurol. 146 Ž1972. 1–14. L.P. Morin, J. Blanchard, R.Y. Moore, Intergeniculate leaflet and suprachiasmatic nucleus organization and connections in the golden hamster, Vis. Neurosci. 8 Ž1992. 219–230. H. Okamura, S. Murakami, K. Uda, T. Sugano, Y. Takahashi, C. Yanaihara, N. Yanaihara, Y. Ibata, Coexistence of vasoactive intestinal peptide ŽVIP.-, peptide histidine isoleucine amide ŽPHI.- and gastrin releasing peptide ŽGRP.-like immunoreactivity in neurons of the rat suprachiasmatic nucleus, Biomed. Res. 7 Ž1986. 295–299. J.W. Olney, Glutamate-induced retinal degeneration in neonatal mice electron-microscopy of the acutely evolving lesion, J. Neuropathol. Exp. Neurol. 28 Ž1969. 455–474. G.E. Pickard, Bifurcating axons of retinal ganglion cells terminate in the hypothalamic suprachiasmatic nucleus and the intergeniculate leaflet of the thalamus, Neurosci. Lett. 55 Ž1985. 211–217. G.E. Pickard, F.W. Turek, Effects of partial destruction of the suprachiasmatic nuclei on two circadian pacemakers: wheel running activity and short-day induced testicular regression, J. Comp. Neurol. 156 Ž1985. 803–815. G.E. Pickard, F.W. Turek, A.A. Lamperti, A.J. Silverman, The effect of neonatally administered monosodium glutamate ŽMSG. on the development of retinofugal projections and the entrainment of circadian locomotor activity, Behav. Neural Biol. 34 Ž1982. 433–444.

I. Chambiller Brain Research 808 (1998) 250–261 w41x G.E. Pickard, M.R. Ralph, M. Menaker, The intergeniculate leaflet partially mediates effects of light on circadian rhythms, J. Biol. Rhythms 2 Ž1987. 35–56. w42x S.N. Pradhan, J.F. Lynch Jr., Behavioral changes in adult rats treated with monosodium glutamate in the neonatal stage, Arch. Int. Pharmacodyn. 197 Ž1972. 301–304. w43x M.A. Rea, Light increases Fos-related protein immunoreactivity in the rat suprachiasmatic nuclei, Brain Res. Bull. 23 Ž1989. 577–581. w44x M.A. Rea, Different populations of cells in the suprachiasmatic nuclei express c-fos in association with light-induced phase delays and advances of the free-running activity rhythm in hamsters, Brain Res. 579 Ž1992. 107–112. w45x W.J. Rietveld, J.H. Meijer, J. Ruis, P. Buys, The effect of neonatal administration of monosodium glutamate on the circadian control of food intake in the rat, J. Interdiscipl. Cycle Res. 17 Ž1986. 327–334. w46x H.J. Romijn, A.A. Sluiter, C.W. Pool, J. Wortel, R.M. Buijs, Differences in colocalization between Fos and PHI, GRP, VIP and VP in neurons of the rat suprachiasmatic nucleus after a light stimulus during the phase delay versus the phase advance period of the night, J. Comp. Neurol. 372 Ž1996. 1–8. w47x B. Rusak, I. Zucker, Neural regulation of circadian rhythms, Physiol. Rev. 59 Ž1979. 449–526. w48x B. Rusak, H.A. Robertson, W. Wisden, S.P. Hunt, Light pulses that shift rhythms induce gene expression in the suprachiasmatic nucleus, Science 248 Ž1990. 1237–1240. w49x B. Rusak, L. Mc Naughton, H.A. Robertson, S.P. Hunt, Circadian

w50x w51x

w52x

w53x

w54x

w55x

w56x w57x

261

variation in photic regulation of immediate-early gene mRNAs in rat supraschiasmatic nucleus cells, Mol. Brain Res. 14 Ž1992. 124–130. S.M. Sagar, F.R. Sharp, Light induces a Fos-light nuclear antigen in retinal neurons, Mol. Brain Res. 7 Ž1990. 17–21. W.J. Schwartz, N. Aronin, J. Takeuchi, M.R. Bennett, R.V. Peters, Towards a molecular biology of the suprachiasmatic nucleus: photic and temporal regulation of c-fos gene expression, Neurosciences 7 Ž1995. 53–60. Y.-M. Sun, H.-K. Hsu, S.-I. Lue, M.-T. Peng, Sex-specific impairment in sexual and ingestive behaviors of monosodium glutamatetreated rats, Physiol. Behav. 50 Ž1991. 873–880. E.L. Sutin, T.S. Kilduff, Circadian and light-induced expression of immediate early gene mRNAs in the rat suprachiasmatic nucleus, Mol. Brain Res. 15 Ž1992. 281–290. T.J. Tafelski, A.A. Lamperti, The effects of a single injection of monosodium glutamate on the reproductive neuroendocrine axis of the female hamster, Biol. Reprod. 17 Ž1977. 404–411. R. Teclemariam-Mesbah, P. Vuilliez, A. Van Rossum, P. Pevet, Time course of neuronal sensitivity to light in the circadian timing system of the golden hamster, Neurosci. Lett. 201 Ž1995. 5–8. G. Tosini, M. Menaker, Circadian rhythms in cultured mammalian retina, Science 272 Ž1996. 419–421. K. Yoshida, K. Kawamura, J. Imaki, Differential expression of c-fos mRNA in rat retinal cells: regulation by lightrdark cycle, Neuron 10 Ž1993. 1049–1054.