The effect of adrenalectomy and corticosterone on homotypic collateral sprouting of serotonergic fibers in hippocampus

Neuroscience Leiters, 54 (1985) 111-116 Elsevier Scientific Publishers Ireland Ltd.

III

NSL 03105

THE EFFECT OF ADRENALECTOMY AND CORTICOSTERONE ON HOMOTYPIC COLLATERAL SPROUTING OF SEROTONERGIC FIBERS IN HIPPOCAMPUS

FENG C. ZHOU I ,2. * and EFRAIN C. AZMITlA 2 I Department of Biology, 1009 Main Building, New York University, Washing/on Square, New York, NY 10003, and 2Department of Anatomy, New York College oj Podiatric Medicine, New York, NY 10035 (U.S.A,)

(Received June 29th, 1984; Revised version received December l lth, 1984; Accepted December 12th, 1984)

Key words: corticosterone - homotypic sprouting - serotonin - hippocampus - raphe - adrenalectomy - 5,7-dihydroxytryptamine

Lesioning of serot onergic (5-HT) fibers using 5,7-dihydroxytryptamine (5,7-DI-IT) in one of the two median raphe-hippocampal pathways induced homotypic sprouting from the other. We have utilized this model with the horseradish peroxidase (HRP) tracing technique to test the effects of an adrenal corticosteroid-corticosterone on homotypic sprouting in female adult rats. Removal of the adrenal glands does not interfere with the 5,7-DHT destruction of 5-HT fibers observed at 3 days. However, the animals without circulating adrenal steroids do not show the increase of the HRP-Iabeled cells normally seen 21 days after the 5,7-DHT lesions. Furthermore, s.c. implantation of corticosterone pellets was able to completely restore homotypic sprouting of 5-HT fibers in the hippocampus.

The neurons projecting to the dorsal hippocampus via the cingulum bundle and fornix-fimbria consist largely (90%) of two homologous groups of serotonin (5-HT)-producing neurons in the median raphe nucleus (MRN) [20]. Approximately 200 cells project through the cingulum bundle while another 100 cells project through the fornix-fimbria. This raphe-hippocampal system provides a model to observe homotypic collateral sprouting (in which the sprouting fibers carry the same transmitter as the lesioned ones, for details see refs. 2 and 21). Previous studies have shown that after 5,7-dihydroxytryptamine (5,7-DHT) lesions in the cingulum bundle area, there is a decrease in the activity of the 5-HT synthesis enzyme, tryptophan hydroxylase [5], in 5-HT fiber density and in [3H]proline and horseradish peroxidase (HRP) transport [2, 21] between the raphe and the hippocampus. Hippocampal-related dysfunctions have also been reported [2, 13]. However, structural, biochemical and functional restoration is observed in this system in the long-term cingulum-lesioned animals [2, 5, 21]. • Author for correspondence. 0304-3940/85/$ 03.30 © 1984 Elsevier Scientific Publishers Ireland Ltd.

112

Steroids have been shown to affect both 5-HT biosynthesis and neuronal plasticity. Glucocorticoids stimulate 5-HT synthesis in the adult brain [1, 3, 7, 18] and increase tryptophan hydroxylase activity in the midbrain during development [17]. These steroids have also been shown to enhance axonal regeneration in the severed olfactory bulb [10], stimulate the regeneration of peripheral nerve transplanted in the cerebrum [12], facilitate the growth of ganglionic cells in newborn rats [6] and increase the population of small, intensely fluorescent cells in sympathetic ganglia [9]. In contrast, glucocorticoids inhibit the outgrowth of neuronal processes in cultured sympathetic neurons [19], and retard cholinergic axon sprouting in the rat dentate gyrus [8, 15] and vasopressin fiber sprouting in the paraventricular nucleus [14]. The present work was designed to study the effect of glucocorticoids on collateral sprouting of 5-HT fibers in the hippocampus. 5,7-Dihydroxytryptamine lesion. Female Sprague-Dawley rats (200-250 g) were anesthetized with chloropent (Fort Dodge, IA; i.p., 1 ml/l00 g body wt.), The neurotoxin, 5,7-DHT, was used to lesion the 5-HT fibers. A norepinephrine uptake blocker, desipramine (i.p., 1 mg/IOO g body wt.), was injected 30 min before the 5,7-DHT injection to protect the norepinephrine fibers in the same bundle. The 5,7-DHT injection (5 p.g in 400 ml of 0.025070 ascorbic saline) was made over 10 min using a micro-pipette (60-100 p.m tip) into the right cingulum bundle (angle, 15° lateral from vertical line; coordinates (mm): P 1.0, L 1.3, V 2.4 to bregma). The 5,7-DHT microinjections were made lateral to the cingulum bundle (the 5-HT fiber bundle is located in the medial portion of cingulum bundle) to avoid mechanical damage to this tract. The spread of 5,7-DHT covered the cingulum bundle at the lesion site but did not extend below the corpus callosum or cross the midline. The dose, the coordinates, the spread of toxin, the specificity and the time course of the 5,7-DHT lesion in the raphe-hippocampus system have been previously studied and discussed [2, 4, 5, 20, 21]. Adrenalectomies. Bilateral adrenalectomies (ADX) or sham operations were performed immediately following the 5,7-DHT or sham lesion of 5-HT fibers in the cingulum bundle while the animals were under deep anesthesia. A longitudinal incision was made bilaterally in the skin and the adrenal glands were carefully removed with fine surgical scissors. The animals were sutured and injected once daily with 106 units of penicillin potassium i.m. for two consecutive days and once again a week later. Implantation of corticosterone. Corticosterone pellets were s.c. implanted in the ADX rats [14]. Pellets were made 100% pure by melting corticosterone (Sigma, 81. Louis, MO) in a vial over a low gas flame. The liquid steroid was then poured into a 6 mm diameter hole in a paraffin block. After the pellets solidified, they were removed from the paraffin and trimmed to the correct weight (approximately 100 mg). The pellets were implanted in the nape of the neck of the animals immediately following ADX. Pellets were replaced every 7 days to ensure stable circulating levels of corticosterone. With this method of replacement therapy, a normal circulating corticosterone level (21 ± 2 p.g/ml serum) was achieved, and the large variability in plasma hormone levels associated with corticosterone injections was avoided [14].

113

Female rats were separated into 5 groups: ascorbic saline-sham injected control (n=7), ADX control (n==3), 5,7-DHT injection in the cingulum bundle (n==12), 5,7 -DHT injection in the cingulum bundle of ADX rats (n == 9), and 5,7-DHT lesion in the cingulum bundle of ADX rats with implantation of corticosterone pellets (n =5). These animals under the same anesthetic procedure were then microinjected with 100 nl of 10010 HRP (Sigma, type VI, 20 nl/min) into the ipsilateral dorsal hippocampus 3 or 21 days post-lesion (angle: 90 to skull surface; coordinates (mm): A 4.5, L 1.5, V 3.7 to the lambda) of various groups. All animals were intracardially perfused 24 h after HRP injection with 2.5010 glutaraldehyde-O.l % MgS04 in 0.1 M phosphate buffer, pH 7.2. The brains were removed, sectioned and processed for HRP activity in the brainstem [20, 21]. The number of labeled cells in the MRN were counted using a bright-field microscope in all serial sections through the brainstem of each animal. Abercrombie's formula was used to correct for counting the same neuron twice in adjacent sections [20, 21]. Within 2 h of HRP injection, the majority of the dorsal hippocampus was filled [20, 21]. Within 24 h, retrogradely transported HRP from this injection labeled the neurons in the superior mammillary body, locus ceruleus, interfascicular portion of dorsal raphe nucleus and the median raphe nucleus of the brainstem [20, 21]. In normal animals 295 ± 48 neurons were labeled in the MRN. Prior ADX had no significant effect on the number of HRP-labeled neurons in the MRN (286 ± 13) (Figs. I and 2). The number of neurons in the MRN projecting to the dorsal hippocampus decreased by approximately 56% 3 days after 5,7-DHT lesioning of the cingulum bundle of normal and ADX rats as compared with the sham-injected control (Figs. I and 2). After 21 days post-lesion, in normal rats the number of HRPlabeled cells in the MRN returned to control levels. In contrast, after 21 days postlesion, the number of labeled neurons in the MRN of 5,7-DHT-ADX rats remained significantly reduced (64010 lower than sham-injected control). The failure of labeled cells in ADX animals to increase their number was reversed by s.c. implantation of corticosterone pellets following the ADX procedure (288 ± 53 neurons labeled) (Figs. 1 and 2). Previous evidence indicates that the increase in the number of HRP labeled cells is due to collateral sprouting in the fornix-fimbria rather than regeneration in the cingulum bundle [2, 4, 21]. Furthermore, these labeled neurons exhibited a significant increase in their size when compared to normal controls or to 3 days post-lesion animals. The change in size of the labeled neurons implies a cellular response to the expansion of their terminals. However, the homotypic collateral sprouting of hippocampal5-HT axons seen at 21 days post-lesion is suppressed by ADX. A subsequent replenishment of corticosterone by s.c. implantation of corticosterone pellets after ADX restored the number of labeled cells (Figs. 1 and 2). Thus, glucocorticoids may be necessary for induced 5-HT fiber re-growth in the adult brain, as has been suggested for the neonatal 5-HT system [17]. These anatomical results demonstrate that cor0

I

) I

• ,

~

•

I

.

,

,-

..... ,

_ 1

=

.

~I

~

\ j

I

"'Ul. .4-

-·- - . ...

· ~,

"-

.

..

..

•

..

• •

I

•

• •

·

,

~

•

.} ~

'I. ..

.

.

:

. • I· \

.MR-ri '

I"

I

·f

,J

•

·

t

.

'.

,

• I

~ .

I

.

.

- ". -

"

~

.

..

•f

" ~i

. • .. .

-......

,<

0"

'"

,

•

,

, - ..-

.>

jr .

-l

..

MR N

! 0

\

a

.~.

c

d

e

I

,

10 Q um

Fig. I. Bright-field pho tomicrographs show labeled cells in the MRN after HRP injection in the do rsal hippocampus. a : normal, b: 3 days post- lesion . c: 21 days post-lesion. d: ADX animal at 21 days post-lesion . e: ADX and corticosterone implan ted ani mal at 21 days post-lesion .

~

115

ticosterone has a permissive action on adult 5-HT plasticity, Previous biochemical work in our laboratory has shown that a variety of stressors which increase the 5-HT synthesis rate may also be acting through circulating corticosterone [1, 3]. It is interesting to postulate that these two observations may be related. Homotypic and heterotypic collateral sprouting have been suggested to have very different properties concerning time course, distance of growth and restoration of normal function [4, 11]. Our results suggest that these two mechanisms may likewise be differentially affected by circulating corticosterone. In the heterotypic collateral sprouting studies, adrenalectomy did not significantly affect the increase in the fiber plexus seen after unilateral entorhinal electrolytic lesion [15]. However, low (3.5 p,g/1 00 ml) and high (28 p,g/l 00 ml) levels of plasma corticosterone retarded the heterotypic sprouting in the hippocampus. In contrast, our results suggest that a minimal amount of corticosterone is necessary for homotypic collateral sprouting of 5-HT fibers after specific chemical lesioning. Thus, the level of circulating corticosterone may playa crucial role in favoring different types of neuronal regrowth. The involvement of adrenal corticosteroids in brain recovery may therefore be dependent on the type of damage produced and the nature of the recovery process. This work was supported by a Mount Sinai Fellowship to F.C.Z. and Grant BNS 79-06474 from NSF to E.C.A. The authors wish to extend their appreciation to Dr. Bruce McEwen and Dr. William Rostene of Rockefeller University for their technical advice and to Delia IIIas for typing this paper. 400

300 -

en -J -J LU

o

mrl-

r-r-: f r-- f -

u,

0200

LU

z

0 w

* * l!100

«

11-

a:

X

...J

-J

:2

~

«

a:

X

0

z

«

7

3

0

I-SHAM--l

0 ;z

0

6

3

«

1-3 DAYS-i

a:

-J

«

~

0::

0 ;z

*

t 0

«

6 f--

I-

21

CI)

0

U

+ X

I-

a: 0

o u

«

5 6 DAYS--I

DAYS POST LESION WITH 5.7-DHT

Fig. 2. Number of HRP-Iabeled neurons (mean ± S.D.) in the median raphe nucleus of sham, ADX or ADX with corticosterone implanted animals (ordinate) after 3 or 21 days post-lesion by 5.7-DHT (abscissa). Data were compared by ANOYA. *P< 0.001 with respect to the normals. The number of animals in each group was shown on the bottom of each bar.

116

2 3 4

5 6

7

8 9

IO II

12 13

14 15 16 I7 18 19

20 21

Azrnitia, E.C., The serotonin-producing neurons of the midbrain median and dorsal raphe nuclei. In L. Iversen, S.D. Iversen and S.H. Snyder (Eds.), Handbook of Psychopharmacology, Vol. 9, Plenum, 1978, pp. 233-314. Azmitia, E.C., Buchan, A.M. and Williams, J .H., Structural and functional restoration by collateral sprouting of hippocampal 5-HT axons, Nature (Lond.), 274 (1978) 374-376. Azrnitia, E.C. and McEwen, B.S., Corticosterone regulation of tryptophan hydroxylase in midbrain of the rat, Science, 166 (1969) 1274-1276. Azmitia, E.C. and Zhou, F.C., Induced homotypic collateral sprouting of hippocampal serotonergic fibers. In G. Gilad (Ed.), Processes of Recovery from Neuronal Trauma, Elsevier, Amsterdam, in press. Clewans, C. and Azmitia, E.C., Tryptophan hydroxylase in the hippocampus and midbrain following unilateral injection of 5,7-dihydroxytryptamine, Brain Res., 307 (1984) 125-133. Costa, M., Eranko, O. and Eranko, L., Hydrocortisone-induced increase in the histochemically demonstrable catecholamine content of sympathetic neurons of the newborn rat, Brain Res., 67 (1974) 457-466. DeKloet, E.R., Kovacs, G.L., Szabo, G., Telegdy, G., Bohus, B. and Versteeg, D.H.G., Decreased serotonin turnover in the dorsal hippocampus of rat brain shortly after adrenalectomy: selective normalization after corticosterone substitution, Brain Res., 239 (1982) 659-663. DeKosky, S., Scheff, S.W. and Cotman, C.W., Elevated corticosterone levels: a possible cause of reduced axon sprouting in aged animals, Neuroendocrinology, 38 (1984) 33-38. Eranko, 0., Heath, J. and Eranko, L., Effect of hydrocortisone on the ultrastructure of the small, intensely fluorescent, granule-containing cells in cultures of sympathetic ganglia of newborn rats, Z. Zellforsch., 134 (1972) 297-310. Fertig, A., Kiernan, J .A. and Seyan, S.S.A.S., Enhancement of axonal regeneration in the brain of the rat by corticotrophin and triiodothyronine, Exp. Neurol., 33 (1971) 372-385. Gage, F.H., Bjorklund, A., Stenevi, U. and Dunnett, S.B., Functional correlates of compensatory collateral sprouting by arninergic and cholinergic afferents in the hippocampal formation, Brain Res., 268 (1983) 39-47. Knowles, J.F. and Berry, M., Effects of deoxycorticosterone acetate on regeneration ofaxons in the mammalian central nervous system, Exp. Neurol., 62 (1978) 1-15. McNaughton, N., Azmitia, E.C., Williams, J.H., Buchan, A. and Gray, LA., Septal elicitation of hippocampal theta rhythm after localized de-afferentation of serotonergic fibers, Brain Res., 200 (1980) 259-269. Meyer, J.S., Micco, Di J., Stephenson, B.S., Krey, L.C. and McEwen, B.S., Subcutaneous implantation method for chronic glucocorticoid replacement therapy, Physiol. Behav., 22 (1979) 867-870. Scheff, S.W. and DeKosky, S.T., Steroid suppression of axon sprouting in the hippocampal dentate gyrus of the adult rat: dose-response relationship, Exp, Neurol., 82 (1983) 183-191. Silverman, A.J. and Zimmerman, E.A., Adrenalectomy increases sprouting in a peptidergic neurosecretory system, Neuroscience, 7 (1982) 2705-2714. Sze, P. Y., Glucocorticoids as a regulatory factor for brain tryptophan hydroxylase during development, Develop. Neurosci., 3 (1980) 217-223. Telegdy, G. and Vermes, I., Effects of adrenocortical hormones on activity of the serotoninergic system in limbic structures in rats, Neuroendocrinology, 18 (I975) 16-26. Unsicker, K., Krisch, B., Otten, U. and Thoenen, H., Nerve growth factor-induced fiber outgrowth from isolated rat adrenal chromaffin cells: impairment by glucocorticoids, Proc. Natl. Acad. Sci. USA, 75 (1978) 3498-3502. Zhou, F.e. and Azmitia, E.C., Effects of 5,7-dihydroxytryptamine on HRP retrograde transport from hippocampus to midbrain raphe nuclei in the rat, Brain Res. Bull., 10 (1983) 445-451. Zhou, F.e. and Azmitia, E.C., Induced homotypic collateral sprouting of serotonergic fibers in hippocampus, Brain Res., 308 (1984) 53-62.