Effect of GM1 ganglioside on neonatally neurotoxin induced degeneration of serotonin neurons in the rat brain

Developmental Brain Research, 16 (1984) 171-180

171

Elsevier BRD 50108

Effect of GM 1 Ganglioside on Neonatally Neurotoxin Induced Degeneration of Serotonin Neurons in the Rat Brain G. JONSSON 1, A. GORIO 2, H. HALLMAN 1, D. JANIGRO 2, H. KOJIMA l and R. ZANONI 2

1Department of Histology, Karolinska lnstitutet, S-104 01 Stockholm (Sweden) and 2Fidia Research Laboratories, Department of Cytopharmacology, 350 31 Abano Terme (Italy) (Accepted May 29th, 1984)

Key words: serotonin neurons - - CNS-development - - 5,7-dihydroxytryptamine - - degeneration - - GM 1ganglioside - - regeneration

The effect of exogenous GM l ganglioside on the 5,7-dihydroxytryptamine (5,7-HT; a selective serotonin neurotoxin) induced alteration of the postnatal development of central 5-hydroxytryptamine (5-HT; serotonin) neurons has been investigated using neurochemical and immunocytochemical techniques. Neonatal 5,7-HT (50 mg/kg s.c.) treatment is known to lead to a marked and a permanent degeneration of distant 5-HT nerve terminal projections (e.g. in cerebral cortex, hippocampus and spinal cord), while projections close to the 5-HT perikarya in the mesencephalon and pons-medulla increase their nerve density. These regional alterations are reflected by decreases and increases, respectively, of endogenous 5-HT, [3H]5-HT uptake in vitro and number of 5-HT nerve terminals demonstrated by immunocytochemistry. Treatment of newborn rats with GM l (4 × 30 mg/kg s.c.; 24 h interval) had no significant effect on the postnatal development of 5-HT neurons. GM 1 administration had furthermore no effect on the 5,7-HT induced alteration of the regional 5-HT levels and [3H]5-HT uptake in the cerebral cortex acutely, indicating that GM~ did not significantly interfere with the primary neurodegenerative actions of 5,7-HT. At the age of 1 month a clear counteracting effect of GM~ was observed, in particular of the 5,7-HT induced 5-HT denervations. The 5-HT levels in the frontal and occipital cortex were reduced to 25 and 20% of control after 5,7-HT alone, while these values were 70 and 40%, respectively, after 5,7-HT and GM 1 treatment. A similar antagonizing effect of GM 1 was found in the frontal cortex when measuring [3H]5-HT uptake. GM 1 treatment also caused a minor reduction of the 5,7-HT induced increase of the 5-HT levels in striatum and mesencephalon. Quantitation of 5-HT nerve terminal density in sections processed for 5-HT immunocytochemistry using an automatic image analysis system showed markedly more nerve terminals in the frontal and occipital cortex after 5,7-HT + GM 1compared to 5,7-HT treatment alone. Minor counteracting effects of GM~ were noted in the hippocampus and spinal cord (thoracic-lumbar) as evaluated by chemical 5-HT assay, although substantial counteracting effects were observed locally in these areas by quantitative immunocytochemistry. The present data are compatible with the view that GM l ganglioside administration has a preventing action on degeneration processes secondary to the direct 5,7-HT neurotoxicity and/ or a growth stimulatory effect on central 5-HT neurons damaged by a selective chemical neurotoxin in the neonatal stage.

a d d i t i o n d a t a p o i n t i n g to a r e l a t i o n b e t w e e n d e v e l -

INTRODUCTION

o p m e n t a l alterations in g a n g l i o s i d e levels and certain Gangliosides

are

sialic acid

containing

glyco-

stages of the d e v e l o p m e n t o f the n e r v o u s system35, 36.

sphingolipids t h a t are a s s o c i a t e d with m o s t cell m e m -

A role for gangliosides d u r i n g the d e v e l o p m e n t is

branes, a l t h o u g h p a r t i c u l a r l y a b u n d a n t in n e u r o n a l

also s u g g e s t e d f r o m d a t a s h o w i n g that the p r e s e n c e

m e m b r a n e s 23. D u r i n g t h e last years t h e r e has b e e n an

of low c o n c e n t r a t i o n s of gangliosides ( 1 0 - 6 - 1 0 -8 M )

increasing i n t e r e s t in t h e i r possible f u n c t i o n a l role in

in culture m e d i a o f cortical n e u r o n s

various m e m b r a n e - b o u n d n e u r o n a l processes. It is of

brain26 as well as of P C 12 cellsS, 8 can s t i m u l a t e n e u -

r e l e v a n c e in this c o n t e x t to n o t e that t h e r e is an increasing b o d y o f e v i d e n c e indicating that e x o g e n o u s a d m i n i s t r a t i o n of g a n g l i o s i d e s can p r o m o t e r e g r o w t h processes and f u n c t i o n a l r e c o v e r y f o l l o w i n g a lesion b o t h in the C N S and P N S 1,3,7,9,19,42,43. T h e r e are in

rite o u t g r o w t h , s p r o u t i n g and r e g e n e r a t i o n . T h e inv o l v e m e n t of gangliosides in n e u r o n a l g r o w t h and m a t u r a t i o n p r o c e s s e s has also b e e n i m p l i c a t e d f r o m in vitro studies on dorsal r o o t ganglia and n e u r o b l a s t o m a cells37, 40. It has f u r t h e r m o r e b e e n r e p o r t e d that

f r o m chick

Correspondence: G. Jonsson, Department of Histology, Karolinska Institutet, P.O. Box 60 400, S-104 01 Stockholm 60 (Sweden).

172 exogenous gangliosides can enhance synapse formation in nerve-muscle cultures 29. Finally, a significant role of gangliosides in neuronal differentiation and synaptogenesis is also suggested from studies on brains afflicted with ganglioside storage disease, where the development of meganeurites with abberant neurite extensions have been observed 34. In view of this, it was therefore considered of interest to investigate whether or not exogenous administration of gangliosides can affect the final outcome of a selective neurotoxin induced damage of transmitter-identified neuron systems in the developing rat brain. In the present study we report on the effect of the monosialoganglioside GM1 on the 5,7-dihydroxytryptamine (5,7-HT; a serotonin neurotoxin) induced alteration of the postnatal development of central 5-hydroxytryptamine (5-HT; serotonin) neurons as monitored by employing neurochemical and immunocytochemical techniques. Parts of the present study have been communicated in preliminary form 17. MATERIALS AND METHODS Newborn albino rats (Sprague-Dawley) were injected with 5,7-HT (50 mg free base/kg s.c.; 5,7-HT creatinine sulphate, Regis) within 6 - 8 h after birth. The rat pups received 4 injections of GM 1 (30 mg/kg s.c., Fidia), one injection daily during postnatal days 1 to 4, except in the acute experiments where only two injections were given; one on postnatal day 1 and one on day 2. The first GM 1 injection was always given 2-3 h after the 5,7-HT administration. Both 5,7-HT and GM~, were dissolved in physiological saline and injected in a volume of 0.05 ml. Controls received equal injections of the solvent. The litters were housed in air-conditioned rooms with controlled temperature and dark/light schedule (10/14 h). Each litter contained 8-10 pups, who were separated from their mothers at the age of about 3 weeks. In the neurochemical studies the rats were killed by decapitation using a guillotine. The brain and spinal cord (lower segments of the thoracic and upper segments of the lumbar cord) were rapidly removed and placed in ice-cold 0.9% NaCI, a few min pending further dissection. The regional dissection of the brain was performed as reported previously15~16. An approximately equal number of male and female rats were studied in each group of most experiments.

5-H T assay The sampled tissue specimens were immediately frozen on dry ice and stored at -80 °C wrapped in aluminium foil. 5-HT was determined employing liquid chromatography with electrochemical detection (LCEC)33. The 5-HT values were corrected for recovery (80-90%) and if not otherwise stated expressed as ng/g wet weight of the tissue. [3H]5-HT uptake in vitro Standardized slices were prepared by punching out discs (1.5 mm in diam.) from approximately 0.5 mm thick transverse slices from the frontal cortex of adult rats. In the 2-day old rats (acute experiment) the discs were prepared from a tangential slice including both frontal and parietal cortex (denoted cerebral cortex). The discs were preincubated for 5 rain at +37 °C in Krebs-Ringer bicarbonate buffer (pH 7.4) containing 0.2 mg/ml ascorbic acid (as anti-oxidant) and 10~tM of the MAO inhibitor pargyline HCI (Sigma) 24. Thereafter 10 ~1 [3H]5-HT (5-[G-3H]-HT creatinine sulphate; specific activity 10-13 Ci/mmol; Radiochemical Center) was added to the incubation medium to give a final concentration of 0.05 ~M and the incubation was continued for another 10 min. After termination of the incubation, the discs were briefly rinsed, solubilized (Protosol, NEN) and toluene phosphor added. Radioactivity taken up and retained in the discs was determined by liquid scintillation spectrometry (counting efficiency 21%). [3H]5-HT uptake was expressed as nCi/disc or slice and corrected for 'extraneuronal' uptake by subtracting uptake values obtained after identical incubations with 5 #M of the 5-HT uptake blocker chlorimipramine HCI (Ciba-Geigy) in the incubation medium 14~38.The radioactivity values thus obtained represent [3H]5-HT uptake in 5-HT nerve terminals. In one set of experiments the whole CNS (brain + spinal cord) was dissected out and homogenized in 0.3 M sucrose for the preparation of a synaptosomal fraction according to Jonsson et al. is. Aliquots of the synaptosomes were taken for endogenous 5-HT assay and [3H]5-HT uptake in vitro as described above. The uptake values were corrected for 'extraneuronal uptake'. Immunocytochemistry The rats were pretreated with pargyline (100

173 mg/kg i.p., 2 - 4 h), anesthetized with pentobarbital (30 mg/kg i.p.) and perfused through the left ventricle of the heart with 50 ml Ca2+-free Tyrode's solution containing heparin. Thereafter an equal volume of ice-cold (+4 °C) Tyrode's buffer was perfused, immediately followed by 400 ml of ice-cold (+4 °C) 4% paraformaldehyde in 0.1 M S6rensen's buffer (pH 7.4). The brain and spinal cord were dissected out and post-fixed in the same fixative for 2 h at +4 °C after which they were rinsed over-night at +4 °C in S6rensen's buffer containing 5% sucrose. The CNS tissue was frozen and cut on a cryostat at -25 °C. The sections (10/~m) were mounted on glass slides and processed for immunocytochemistry according to the indirect immunofluorescence technique of Coons 4 as previously described 41. The sections were first rinsed in phosphate buffered saline (PBS) and then incubated for 20 h at +4 °C with a rabbit antiserum to 5HT (Miles-Seravac) diluted (1:400) with PBS and containing 0.1% Triton X-100. After a 30 min rinse in PBS, the sections were incubated for 30 min at room temperature with fluorescein isothiocyanate (FITC)-labeled sheep anti-rabbit immunoglobulins diluted 1:20 with PBS. The sections were again rinsed in PBS and mounted in glycerin-PBS (3:1). A Zeiss or a Leitz fluorescence microscope equipped with epi- and transillumination was used for examination and microphotography.

and generation of the binary picture, the computer calculated the area covered by fluorescent 5-HT nerves in percent of the field of view (measuring field) after editing. The size of the non-edited measuring field (square) was at a primary magnification of x 16,400 x 400/~m, while 255 x 255ktm at x 25. Assuming that the fluorescence morphology of the 5HT nerve terminals is relatively uniform, the percentage values calculated in this way can be considered as a measure of the relative 5-HT nerve density per unit area in the section. RESULTS The rat pups tolerated very well the 5,7-HT and GM 1 treatments and no notable increase in mortality due to treatment was observed. All experimental groups showed approximately normal body and brain weight gain and no observable alterations in gross behaviour of the animals could be observed. Neurochemical observations In acute experiments, treatment of newborn rats with 5,7-HT (20 h) led to a marked (-70 to -90%) reduction of the endogenous 5-HT levels in cerebral cortex and spinal cord, whereas only moderate reductions (-20 to -30%) were observed in mesence-

[ ] 5,7-HT ~A5.7-HT*GM1 [ ] GM112o5 - H T nerve density measurements In order to obtain more objective information on the 5-HT nerve terminal density in the immunocytochemically stained sections for 5-HT, these were analyzed by an interactive image analysis system (IBAS, Kontron/Zeiss) 10. The fluorescence microscopical visualization of 5-HT neurons was made by using epiillumination and a primary magnification of x 16 or x 25 (Neofluar, Zeiss). An image intensified highresolution black and white TV-camera (Siemens K5b M21005) was attached to the fluorescence microscope. The IBAS instrument was programmed to adjust for uneven background illumination and to enhance contrast. Thereafter (if needed) non-specific background fluorescent structures and artefactual areas in the sections (e.g. cracks) were excluded from the field of view. Finally, a gray level threshold was chosen to select for fluorescent nerves excluding virtually all background. After gray level definition,

A. ....

S.

4+

....

+ 80-

d ~6o

60-

! 20

CCx.

Mes

!

+

F

P-m.

S~+c

~40-

I

20-

Cex.

Fig. 1. Acute effects of 5,7-HT (50 mg/kg s.c., 20 h) and/or GM1 (2 × 30 mg/kg s.c., 14 h interval) treatment of newborn rats on the regional 5-HT levels (A) and in vitro uptake of [3H]5-HT (0.05/~M, 10 min) in slices from the cerebral cortex. The first GM1dose was injected 2 h after 5,7-HT and the animals killed 4 h after the second dose. Each bar represents the mean + S.E.M. of 7-10 determinations, expressed as % of saline injected control. A: 5-HT control values; CCx. (cerebral cortex), 60 + 5.2; Mes. (mesencephalon), 136 + 19; P-m. (pons-medulla), 180 + 13; Sp.c. (spinal cord), 192 + 21 ng 5HT/g; B: [3H]5-HTuptake in control, 10.3 + 0.97 nCi/slice.

174 phalon and pons-medulla (Fig. 1A). A substantial reduction of [3H]5-HT uptake in cerebral cortex was also found after 5,7-HT, although less pronounced than that of endogenous 5-HT (Fig. 1B). The reason for this discrepancy is somewhat unclear, but might be related to a certain degree of 5-HT depletion produced by 5,7-HT in the acute phase not being associated with 5-HT nerve terminal degenerationlM3, as well as to methodological difficulties in obtaining a correct 'blank value' for [3H]5-HT uptake in immature cortex. Part of the [3H]5-HT uptake recorded is probably 'extra-neuronal'. GM 1 (x 2; 14 h interval) administration did not significantly modify the 5,7HT induced alterations; neither the regional 5-HT levels nor the [3H]5-HT uptake in cerebral cortex. GM 1 treatment alone was furthermore unable to alter the 5-HT parameters recorded as compared to control (Fig. 1). A similar pattern of 5,7-HT induced 5-HT depletions was found when performing the analysis at the age of 1 week, except that a moderate 5-HT increase (+30%) was now observed in ponsmedulla (Fig. 2). Also at this developmental stage, GM 1 (x 4) was not found to have any significant effect on the changes of 5-HT levels caused by 5,7-HT in any of the regions analyzed. Treatment with GM~ alone was again ineffective in altering the regional 5HT levels as compared to saline injected control. The regional 5-HT analysis in 1-month-old rats disclosed that the relative 5,7-HT induced reduction of 5-HT was approximately the same in the cortical re-

15o-

0[

[ ] 5,7-HT [ ] 5,7-HT~GM~ [ ] GM 1

1 week.

Fr.Cx.

Occ.Cx.

Str.

Mes.

P-m.

Sp.c.

Fig. 2. Effect of GM 1 on the 5,7-HT induced alteration of the regional 5-HT levels in the brain of 1-week-old rats. 5,7-HT (50 mg/kg s.c.) was administered on the day of birth and thereafter GM 1 (4 x 30 mg/kg s.c., 24 h interval) was injected daily. The first dose of G M 1 was administered 2 h after 5,7-HT. Each bar represents the mean + S.E.M. (n = 4), expressed as a percentage of the respective region obtained from age-matched salineinjected controls. Fr.Cx., frontal cortex; Occ.Cx., occipital cortex; Str., striatum; Mes., mesencephalon; P-re., pons-medulla; Sp.c., spinal cord (thoracic-lumbar region).

[] 53-HT

[ ] 5,7-HT+GM1

o Fr.Cx.

Occ.Cx.

Hipp.

Str.

Mes.

P-re.

Sp.c.

Fig. 3. Effect of 5,7-HT and/or GM 1 treatment in the neonatal stage on the regional 5-HT levels in the brain of 1-month-old rats. GM 1 and 5,7-HT were administered as in Fig. 2. Each bar represents the mean + S.E.M. (n = 6-7), expressed as % of saline injected control. Control values: Fr.Cx. (frontal cortex), 159 + 4.6; Occ.Cx. (occipital cortex), 87 + 3.3; Hipp. (hippocampus), 130 + 14; Str. (striatum), 137 + 7.5; Mes. (mesencephalon), 352 + 16; P-m. (pons-medulla), 284 _+ 9.4; Sp.c. (thoraci-lumbar spinal cord), 236 + 15 ng 5-HT/g. Statistical differences between 5,7-HT and 5,7-HT + G M I groups: * 0.05 > P > 0.01; *** P < 0.001 (Student's t-test).

gions and spinal cord as in one week old rats (Fig. 3). In striatum and mesencephalon it was observed that the 5-HT levels were now moderately increased after 5,7-HT, and markedly so in the pons-medulla region. In contrast to the situation in 2-day- and 1-week-old rats, it was found, however, that the GM 1 treatment led to a significant counteraction of the 5,7-HT induced decrease of 5-HT in both neocortical areas, most pronounced in the frontal cortex (Fig. 3). Minor counteracting effects of GM1 were also noted in hippocampus and spinal cord. A tendency towards a reversal of the 5,7-HT induced increase of 5-HT levels in striatum (P < 0.05) and mesencephalon (P < 0.1) was also noted, while GM1 did not appear to have any significant effect on the marked 5-HT increase in pons-medulla. Similar to the situation in 2day- and 1-week-old rats neonatal GM~ treatment alone did not produce any significant effect on the regional 5-HT levels. When recording the effect of GM l on the 5,7-HT induced decrease of the 5-HT levels in the frontal and occipital cortex at various developmental stages no effects of GM1 were thus noted during the first postnatal week and the first signs of a counteracting action were observed in 2.5-week-old animals, most clear-cut in the frontal cortex (Fig. 4). No effects of GM1 treatment alone were found at any age analyzed.

175 ,oo

Fronta/cx.

~%

-

0

loo]~ Occipital Cx.

-

1

%

2

4W

0

1

2

4w

Fig. 4. Effect of 5,7-HT and/or G M 1 treatment in the neonatal stage on the endogenous 5-HT levels in frontal and occipital cortex of 2-day-, 1-week-, 2.5- and 4-week-old rats. G M 1 and 5,7-HT was administered as in Fig. 2, except for the 2-day-old rats that received only two injections of GM 1 (14 h interval) and were killed 4 h after the last GM 1 injection. Each point represents the mean + S.E.M. of 4-7 determinations, expressed as % of age-matched saline treated control.

Analysis of [3H]5-HT uptake in vitro in slices (discs) from frontal cortex of 2-month-old rats showed that neonatal 5,7-HT administration caused a marked reduction of [3H]5-HT uptake. Additional GM 1 treatment led to a significant counteraction of this effect, while GM 1 treatment alone had no effect on [3H]5-HT uptake (Fig. 5). Measurement of endogenous 5-HT and [3H]5-HT uptake in the whole CNS after neonatal 5,7-HT and/or GM 1 treatment in 2month-old rats showed no significant effects on these parameters in any of the groups (Table I). Minor and similar reductions of both 5-HT parameters were noted after 5,7-HT and 5,7-HT + GM 1 treatment.

very marked reduction of the density of demonstrable fluorescent 5-HT nerve terminals in frontal and occipital cortex. Very marked reductions of 5-HT nerve density were also observed in the hippocampus and lumbar spinal cord (Fig. 6), whereas a very moderate decrease only was noted in the cervical spinal cord. It was also noted that in the ventrally located areas of the cerebral cortex (e.g. piriform cortex and basal posterior amygdaloid nucleus), the effects of 5,7-HT were minor to nil. The neonatal 5,7-HT treatment did not have any clear-cut effect on the number and the fluorescence morphology of the 5-HT perikarya in the mesencephalon and pons-medulla, although the general impression was a slight reduction of the number of cell-bodies in the 5-HT perikaryal groups of the latter region. The density of the 5-HT 120 ¸ Frontal Cx. 100 % =k

~

!ill

60

~ 40

!ill

20

iii

I m r n u n o c y t o c h e m i c a l observations

5,74dT5,7-H T GM. ÷ GM1

T h i s a n a l y s i s was p e r f o r m e d in 3 - 4 - w e e k - o l d r a t s t r e a t e d w i t h 5 , 7 - H T a n d / o r G M 1 in t h e n e o n a t a l s t a g e as d e s c r i b e d in Fig. 2. T h r e e r a t s in e a c h e x p e r imental group was investigated. While GM t treatment alone did not appear to cause any clearly detectable alteration of the fluorescence morphology of t h e 5 - H T n e u r o n s in a n y o f t h e a r e a s a n d r e g i o n s a n a lyzed, it w a s o b s e r v e d t h a t 5 , 7 - H T t r e a t m e n t l e d t o a

C

Fig. 5. Effect of 5,7-HT and/or GM 1 treatment in the neonatal stage on the in vitro uptake of [3H]5-HT (0.05/tM, 10 min) in slices from the frontal cortex of 2-month-old rats. 5,7-HT and GM 1 was administered as in Fig. 2. Each bar represents the mean + S.E.M. 8-12 determinations, expressed as % of saline injected control. Control value: 14.6 + 1.56 nCi/slice. Statistical difference between 5,7-HT and 5,7-HT + GM 1 groups: * 0.05 > P > 0.01 (Student's t-test).

TABLE I Effect of neonatal 5, 7-HT and~or GM I treatment on the in vitro uptake of [3H]5-HT and on the endogenous 5-HT in the whole CNS

5,7-HT and GM~ were administered as in Fig. 2 and the rats killed in the adult stage (2 months old). The whole CNS (brain + spinal cord) was dissected out, homogenized and samples taken for [3H]5-HT uptake in vitro (0.05/~M, 5 min) and chemical 5-HT assay. The [3H]5-HT uptake data are expressed as cpm x 102/sample, corrected for 'extraneuronal uptake', while 5-HT is expressed as pg/ sample. Mean + S.E.M. of 3-4 determinations.

[3H]5-HT uptake Endogenous 5-HT

Control

5, 7-HT

5, 7-HT + GM 1

GM t

29 + 2.5 929 + 111

26 + 2.5 727 + 83

25 + 2.1 765 + 60

31 + 2.7 955 + 75

I~

l,o

177 nerve terminal fields in the cell-body areas were in general increased after 5,7-HT, although marked variations were noted, from moderate to very marked increases. The latter was in particular observed in the pons-medulla region. Administration of GM 1 to 5,7-HT treated rats led to a clear counteraction of the 5,7-HT induced reduction of demonstrable 5-HT nerve terminals in both frontal and occipital cortex as well as in the hippocampus and lumbar spinal cord (Fig. 6). In cervical spinal cord no clearly detectable difference in 5-HT nerve terminal density was observed as compared to 5,7-HT treatment alone. This was also the case for the 5-HT nerve terminal fields in ventral cortical areas and in mesencephalon and pons-medulla. As to the 5-HT perikarya it was found that both the fluorescence morphology and number of cell-bodies in the various cell-groups did not differ to any notable degree from that of control. In order to obtain more objective and quantitative information on the effect of 5,7-HT and/or GMI on 5HT neurons, their nerve terminal density was determined in some regions by computer assisted image analysis. In frontal and occipital cortex (transverse sections) 4 measuring fields in the superficial layers were sampled in each section and 3 sections per animal were analyzed. The 12 measurements for each region were averaged to represent the 5-HT nerve terminal density value for the animal in question. Similar measurements were made in dorsal hippocampus, ventral cortical areas (piriform cortex and n. amygdaloideus basalis posterior) as well as in lumbar spinal cord (anterior horn in transverse sections). The values presented below is the mean from 3 animals in each group. In control rats it was found that the average nerve density values obtained by image analysis (computed as the area covered by fluorescent 5-HT nerve terminals in % of the whole measurement field) varied from region to region, being 2.9% in the frontal and 1.6% in the occipital cortex. The control nerve density value for dorsal hippocampus was 1.7% and for lumbar spinal cord (anterior horn) 9.5%, while for piriform cortex 7.7% and for the amygdaloid region 10.8%. GM~ treatment alone in the neonatal stage did not cause any significant alteration of the 5-HT nerve density in any of these regions as evaluated by image analysis. There was a very high correlation (r

= 0.996) between the 5-HT levels recorded by chemical assay (Fig. 3) in the frontal cortex, occipital cortex, hippocampus and spinal cord (corrected for 5HT concentration in the gray matter; 761 ng/g) and the nerve density values from these regions obtained by quantitative image analysis. This strongly suggests that the endogenous 5-HT levels measured in these regions under the present conditions are a reasonably reliable index of 5-HT nerve terminal density, which also has been found in previous studies~2. The neonatal 5,7-HT treatment produced a marked reduction of 5-HT nerve density in frontal ( - 7 4 % of control value), occipital (-66%), hippocampus ( - 8 4 % ) and lumbar spinal cord (-80%). No clear alterations were recorded in the piriform cortex ( - 8 % ) and the amygdaloid region analyzed (-7%). After combined 5,7-HT and GM1 treatment it was found that the computed 5-HT nerve density values for frontal and occipital cortex, hippocampus and spinal cord were markedly higher than those obtained after 5,7-HT treatment alone. In the frontal cortex the computed value, was thus 0.75% after 5,7-HT alone, while 1.6% after 5,7-HT + GM1, which is quantitatively very similar to the data obtained by chemical 5-HT assay. In occipital cortex the nerve density value increased from 0.55 to 1.75%, in hippocampus from 0.27 to 0.82% and in spinal cord from 1.9% after 5,7HT alone to 4.3% after 5,7-HT + GM 1. These counteracting effects of GM x are more pronounced than those found when measuring 5-HT levels in these regions (cf. Fig. 3). No clear-cut effects were recorded in the piriform cortex and the amygdaloid region. DISCUSSION The central 5-HT neurons are one of the earliest forming neuronal populations in the brain and the cell-body groups as well as the general outline of their main axonal pathways are formed before birth 22,30. The postnatal development of these neurons consists mainly of a proliferation and maturation of the nerve terminals in the target areas 22,25,3°. The present results confirm previously published results showing that neonatal 5,7-HT treatment produces a heterogenous alteration of the regional 5-HT levels, 5-HT uptake and 5-HT nerve density 13,14,32,38. The general feature of this 5,7-HT induced effect is a marked and permanent degeneration of distal 5-HT

178 nerve terminal projections (e.g. in the cerebral cortex, hippocampus, and spinal cord), while the treatment leads to an increased 5-HT nerve density in areas close to the 5-HT cell-body groups in the mesencephalon and pons-medulla. Previous fluorescence histochemical studies and the present immunocytochemical analysis indicated that the majority of the 5-HT perikarya remain after 5,7-HT 14. The rather dramatic effect of the neurotoxin on the development of 5-HT neurons is most likely mainly related to a 'pruning effect '39, where prevention of the development of distally projecting axonal branches leads to a proliferative growth response in intact collateral branches close to the cell-body regions (see ref. 12). This explanatory model implies that the central 5-HT neurons have a relatively strictly regulated developmental program for the expression of a certain number of axonal arbours during the postnatal development, which also is supported by the fact that 5,7HT treatment does not significantly alter 5-HT levels and [3H]5-HT uptake in whole CNS (Table I). Both the biochemical and immunocytochemical results obtained in the present study clearly demonstrate that exogenous administration of GM 1 ganglioside has a counteracting effect on the 5,7-HT induced alteration of the development of the 5-HT neuron systems. This counteracting effect appears to be most evident in regions where 5,7-HT induces permanent 5-HT denervations, although tendencies for a reversal of the increases of 5-HT levels caused by 5,7-HT were noted in the striatum and mesencephalon. Although the chemical 5-HT assay indicated rather small counteracting effects of GM 1 in certain regions (e.g. hippocampus and spinal cord), the quantitative immunocytochemical data disclosed a rather marked reversal of the 5,7-HT induced denervation locally. The discrepancy between the neurochemically recorded 5-HT levels and the 5-HT nerve density values obtained by image analysis, is most likely related to the fact that 5-HT levels were measured in relatively large pieces of brain tissue, leading to a 'dilution effect' masking even fairly large differences present locally. It thus appears as if GMI has a normalizing effect on the over-all changes (both denervation and hyperinnervation) of the 5-HT nerve terminal distribution that a neonatal 5,7-HT treatment produces. Since the 'pruning effect' hypothesis implies that the hyperinnervation following the neuro-

toxin administration is a consequence of the degenerative actions on certain axonal branches, it is conceivable to assume that the effects of GM t observed in the present study is primarily related to a counteracting effect of the degenerative actions induced by 5,7-HT. As to how GM1 produces the partial abolishment of the 5,7-HT induced 5-HT nerve terminal degeneration, the present data do not provide any definite answer to this question. An obvious point in this context is whether GM1 interferes with the primary neurodegenerative (cytotoxic) action of 5,7-HT or not. Although this possibility cannot be ruled out completely due to methodological limitations, it seems less likely. The reason for this statement is two-fold; firstly, GM 1 was administered at a time-point when the irreversible neurotoxic effects of the relatively fast-acting neurotoxin 5,7-HT are more or less completedlk13; secondly, the combined 5,7-HT + GMI treatment had approximately the same effect on the 5-HT parameters analyzed when measured in the acute phase (up to 1 week) as 5,7-HT treatment alone. It would thus seem as if the primary neurodegenerative actions of 5,7-HT are not altered to any detectable degree by the GM l treatment schedule used. Consequently, the observed effects of GM 1 rather have to be related to effects on regrowth and/ or other processes. Considering the 5,7-HT lesion model used where the chemical neurotoxin acts by being almost exclusively accumulated in 5-HT neurons and there undergoes autocatalytic oxidation reactions leading to an irreversible damage and degeneration (see refs. 2, 11), it is known from previous studies that 5,7-HT systemically administered acts preferentially on 5HT nerve terminals, while 5-HT axons and cell-bodies are clearly less susceptiblelk However, there are data indicating that the nerve terminal degeneration induced by 5,7-HT is accompanied by a certain degree of retrograde degeneration (unpublished data), although it is at present unclear if there is any significant degree of 5-HT cell-body death after 5,7-HT. The minor reduction of 5-HT levels and [3H]5-HT uptake in whole CNS after 5,7-HT (Table I) might reflect a limited cell-body death. It is also well known from many studies that developing neurons in general exhibit a more intense axon reaction and retrograde degenerative changes after an axonal lesion

179 than that of mature neurons subjected to an identical lesion (see ref. 11). It is therefore possible that the effect of GM 1 observed in the present study is due to a prevention or reduction of secondary retrograde degenerative processes following the primary 5-HT nerve terminal degeneration induced by 5,7-HT. Such an effect could very well explain the present resuits. This is also consistent with previously reported data indicating a cell-body sparing effect of GM 1 on dopamine neurons after an axonal transection of nigrostriatal dopamine neurons 1. This would also be in line with the present immunocytochemical observations that there was an impression of a counteracting or sparing effect of GM1 on the moderate reduction of demonstrable 5-HT perikarya induced by 5,7HT in pons-medulla. However, if the small reduction of 5-HT levels and uptake in whole CNS after 5,7-HT reflects a certain degree of cell-body death, it was noted that GM1 did not seem to counteract this effect arguing against a cell-body sparing effect of GM 1. On the other hand, it has to be kept in mind that measurement of whole CNS levels of 5-HT and 5-HT uptake is probably a relatively insensitive way of demonstrating a limited 5-HT cell-body death. Further studies conducting cell-body counts are, therefore, needed to substantiate this preliminary observation of cell-body sparing effect of GM 1. It should also be mentioned that exogenous gangliosides have been shown to improve experimental diabetic neuropathy at a stage when insulin proved to be ineffective27, 28, suggesting a more general protective action of these substances against neuronal damage. In view of the previously reported sprouting or reREFERENCES 1 Agnati, L. F., Fuxe, K., Calza, L., Benefenati, F., Cavicchioli, L., Toffano, G. and Goldstein, M., Gangliosides increase the survival of lesioned nigral dopamine neurons and favour the recovery of dopaminergic synaptic function in striatum of rats by collateral sprouting, Acta physiol. scand., 119 (1983) 347-363. 2 Baumgarten, H. G. and Bjrrklund, A., Neurotoxic indoleamines and monoamine neurons, Ann. Rev. Pharmacol. Toxicol., 16 (1976) 101-111. 3 Ceccarelli, B., Aporti, F. and Finesso, M., Effects of brain gangliosides on functional recovery in experimental regeneration and reinnervation. In G. Proceilati, B. Ceccarelli and G. Tettamanti (Eds.), Advances in Experimental Medicine and Biology, Vol. 71, Plenum Press, New York, pp. 275-293. 4 Coons, A. H., Fluorescent antibody methods. In J. F. Canielli (Ed.), General Cytochemical Methods, Vol. L Aca-

growth stimulatory effects of gangliosides on neurons in culture 6, on peripheral autonomic and cholinergic neurons3,7, 9, as well as on central cholinergic and catecholamine neurons1,20,42,43, it is possible that the present data are also related to a facilitatory action of GM 1 on regrowth processes. Whether such an effect of GM 1 would be direct or indirect is at present unknown, although it has been demonstrated that gangliosides can enter the brain after exogenous administration21,31. In conclusion, the present study has demonstrated that GM1 ganglioside treatment has a counteracting effect on 5,7-HT induced alteration of the postnatal development of central 5-HT neurons. This effect was partial, although quite extensive in certain areas locally and most pronounced in regions where 5,7HT produces permanent degenerations. The results are compatible with the view that GM1 ganglioside treatment may have a degeneration preventing and/ or a growth stimuiatory effect on central 5-HT neurons damaged by 5,7-HT neurotoxin in the neonatal stage. ACKNOWLEDGEMENTS The present study has been supported by grants from the Swedish MRC (04X-2295), Karolinska Institutet, the 'Expressen' Prenatal Research Foundation, Bergvall, Jeansson and Samariten Foundations. The skilful technical assistance of B. Kaller, E. Lindqvist and B. Wiehager is gratefully acknowledged.

demic Press, New York, pp. 399-422. 5 Ferrari, G., Fabris, M. and Gorio, A., Gangliosides enhance neurite outgrowth in PC 12 cells, Develop. Brain Res., 8 (1983) 215-221. 6 Gorio, A., Sprouting and regeneration of peripheral nerve. In J. Zagoren and S. Fedoroff (Eds.), Advances in Cellular Neurobiology, AcademicPress, New York, in press. 7 Gorio, A., Carmignoto, G., Facc, L. and Finesso, M., Motor nerve sprouting induced by ganglioside treatment. Possible implications for gangliosides on neuronal growth, Brain Res., 197 (1980) 236-241. 8 Gorio, A., Carmignoto, G. and Ferrari, G., Axon sprouting stimulated by gangliosides: a new model for elongation and sprouting. In M. M. Rapport and A. Gorio (Eds.), Gangliosides in Neurological and Neuromuscular Function, Development and Repair, Raven Press, New York, 1981,

pp. 177-195. 9 Gorio, A., Marini, P. and Zanoni, R., Muscle reinnervation - - III. Motoneuron sprouting capacity, enhancement

180 by exogenous gangliosides, Neuroscience, 8 (1983) 417-429. 10 Henschen, A. and Olson, L., Hexachloropheneqnduced degeneration of adrenergic nerves: application of quantitative image analysis to Falck-Hillarp fluorescence histochemistry, Acta neuropathol., 59 (1983) 109-114. 11 Jonsson, G., Chemical lesioning techniques: monoamine neurotoxins. In A. Bj6rklund and T. H6kfelt (Eds.), Handbook of Chemical Neuroanatomy, Vol. l, Methods in Chemical Neuroanatomy, Elsevier, Amsterdam, 1983, pp. 463-507. 12 Jonsson, G. and Hallman, H., Response of central monoamine neurons following an early neurotoxic lesion, Biblt. Anat., 23 (1982) 76-92. 13 Jonsson, G. and Hallman, H., Effects of substance P on the 5,7-dihydroxytryptamine induced alteration of the postnatal development of central serotonin neurons, Med. Biol.. 61 (1983) 105-112. 14 Jonsson, G., Hallman, H., Pollare, T. and Sachs, Ch., Developmental plasticity of central serotonin neurons following 5,7-dihydroxytryptamine treatment, Ann. N.Y. Acad. Sci., 305 (1978) 328-345. 15 Jonsson, G., Hallman, H. and SundstrOm, E., Effects of the noradrenaline neurotoxin DSP4 on the postnatal development of central noradrenaline neurons in the rat, Neuroscience, 7 (1982) 2895-2907. 16 Jonsson, G. and Sachs, Ch., Regional changes in 3H-noradrenaline uptake, catecholamines and catecholamine synthetic and catabolic enzymes in rat brain following neonatal 6-hydroxydopamine treatment, Med. Biol., 54 (1976) 286-297. 17 Jonssom G., Kojima, H. and Gorio, A., GM 1 ganglioside has a counteracting effect on neurotoxin induced alteration of the postnatal development of central serotonin (5-HT) neurons, Neurosci. Lett., Suppl., 14 (1983) S185. 18 Jonsson, G., Pycock, Ch,, Fuxe, K. and Sachs, Ch., Changes in the development of central noradrenaline neurons following neonatal administration of 6-hydroxydopamine, J. Neurochem., 22 (1974) 621-626. 19 Karpiak, S. E., Ganglioside treatment improves recovery of alternation behaviour after unilateral entorhinal cortex lesion, Exp. Neurol., 81 (1983) 330-339. 20 Kojima, H., Gorio, A., Janigro, D. and Jonsson, G., GM 1 ganglioside enhances regrowth of noradrenaline nerve terminals in rat cerebral cortex lesioned by the neurotoxin 6hydroxydopamine, Neuroscience, in press. 21 Lang, W., Pharmacological studies with 3H-labeled exogenous gangliosides injected intramuscularly into rats. In M. M. Rapport and A. Gorio (Eds.), Gangliosides in Neurological and Neuromuscular Function, Development and Repair, Raven Press, New York, pp. 241-251. 22 Lauder, J., Wallace, J. A., Krebs, H., Petrusz, P. and McCarthy, K., In vivo and in vitro development of serotonergic neurons, Brain Res. Bull., 9 (1982) 605-625. 23 Ledeen, R., Ganglioside structures and distribution: are they localized at the nerve ending? J. Supramol. Struct., 8 (1978) 1-17. 24 Lidbrink, P. and Jonsson, G., Noradrenaline nerve terminals in the cerebral cortex - - effects on noradrenaline uptake and storage following axonal lesion with 6-hydroxydopamine, J. Neurochem., 22 (1974) 617-626. 25 Lidov, H. G. W. and Molliver, M. E., An immunohistochemical study of serotonin neuron development in the rat: ascending pathways and terminal fields, Brain Res. Bull., 9 (1982) 389-430. 26 Louis, J. C., Gorio, A., Massarelli, R., Harth, S. and Dreyfus, H., Effect of gangliosides on the development of the

neurons in cell cultures. In B. Haber, R. Perez-Polo, A. M. Giuffrida and G. Hashim, Nervous system Regeneration, Liss, New York, 1983. 27 Norido, F., Canella, R. and Gorio, A., Ganglioside treatment of neuropathy in diabetic mice, Muscle Nerve, 5 (1982) 107-110. 28 Norido, F., Canella, R., Zanoni, R. and Gorio, A., The development of diabetic neuropathy in the C57 BL/KS (db/db) mouse and its treatment with gangliosides, Exp. Neurol., 83 (1984) 221-232. 29 Obata, K., Oide, M. and Handa, S., Effects of glycolipids on in vitro development of neuromuscular junction, Nature (Lond.), 266 (1977) 369-371. 30 Olson, L. and Seiger, ,~., Early prenatal ontogeny of central monoamine neurons in the rat: fluorescence bistochemical observations, Z. Anat. Entwickl.-Gesch., 137 (1972) 301-316. 31 Orlando, P., Cocciante, G., Ippolito, G., Massari, P., Roberti, S. and Tettamanti, G., The fate of tritium labeled GM~ ganglioside injected in mice, Pharmacol. Res. Commun.. 11 (1979) 759-773. 32 Ponzio, F. and Jonsson, G., Effects of neonatal 5,7-dihydroxytryptamine treatment on the development of serotonin neurons and their transmitter metabolism, Develop. Neurosci., 1 (1978) 80-89. 33 Ponzio, F. and Jonsson, G., A rapid and simple method for the determination of picogram levels of serotonin in brain tissue using liquid chromatography with electrochemical detection, J. Neurochem. 32 (1978) 129-132. 34 Purpura, D. P., Pappas, G. D. and Baker, H. J., Meganeurites and other abberant processes of neurones in GM~-gangliosidosis. A Golgi study, Brain Res., 145 (1977) 13-26. 35 Rahmann, H., Probst, W. and Muhleisen, M., Gangliosides and synaptic transmission, Jap. J. exp. Med., 52 (1982) 275 - 286. 36 Rapport, M. M. and Gorio, A., Gangliosides in Neurological and Neuromuscular Function, Development and Repair, Raven Press, New York, 1981. 37 Roisen, F. J., Bartfeld, H., Nagel, R. and York, G., Ganglioside stimulation of axonal sprouting in vitro, Science, 214 (1981) 577-578. 38 Sachs, Ch. and Jonsson, G., 5,7-dihydroxytryptamine induced changes in the postnatal development of central 5hydroxytryptamine neurons, Med. Biol., 53 (1975) 156-164. 39 Schneider, G. E., Early lesions of superior colliculus: factors affecting the formation of abnormal retinal projections, Brain Behav. Evol., 8 (1973) 73-109. 40 Schwartz, M. and Spirman, N., Sprouting from chicken embryo dorsal root ganglia induced by nerve growth factor is specifically inhibited by affinity-purified antiganglioside antibodies, Proc. nat. Acad. Sci. U.S.A., 79 (1982) 6080-6083. 41 Steinbush, H. W. M., Verhofstad, A. A. J. and Joosten, H. W., Localization of serotonin in the central nervous system by immunohistochemistry: description of a specific and sensitive technique and some applications, Neuroscience, 3 (1978) 811-819. 42 Toffano, G., Savoini, G., Moroni, F., Lombardi, G., Calza, L. and Agnati, L. F., GM l ganglioside stimulates the regeneration of dopaminergic neurons in the central nervous system, Brain Res., 261 (1983) 163-166. 43 Wojcik, M., Ulas, J. and Oderfelt-Nowak, B., The stimulating effect of ganglioside injections on the recovery of choline acetyltransferase activities in the hippocampus of the rat after septal lesions, Neuroscience, 7 (1982) 495-500.