Pharmacologic specificity of antidepressive activity by monoaminergic neural transplants

Psychopharmacology (1995) 118:10-18

© Springer-Verlag 1995

Darin D. Dougherty - Caryl E. Sortwell Jacqueline Sagen

Pharmacologic specificity of antidepressive activity by monoaminergic neural transplants

Received: 18 May 1994 / Final version: 10 August 1994

Abstract Previous studies in our laboratory have demonstrated the ability of monoaminergic transplants in the rat frontal cortex to produce antidepressive activity in both the learned helplessness model and the forced swimming test, as well as to increase monoamine levels in the implanted frontal cortex. These findings implicate increased cortical levels of norepinephrine (NE) and serotonin (5-HT) in the antidepressive activity of monoaminergic transplants. The goal of the present study was to characterize the pharmacologic mechanisms involved in the monoaminergic graft-induced antidepressive activity. Immobility scores in the forced swimming test (FST) were assessed after transplantation of 5-HT-containing pineal gland tissue, NE-containing adrenal medullary tissue, a combination of both tissues, or sciatic nerve (control) into the rat frontal cortex and compared to non-transplanted and chronic imipramine-treated rats. Monoaminergic transplants and imipramine treatment significantly reduced immobility scores in the FST in contrast to control transplanted or untreated animals. All groups were assessed pharmacologically with the adrenergic antagonists phentolamine (~) and propranolol (fi), and serotonergic antagonists metergoline (5-HTt/5-HT2) and pirenperone (5-HT2). Serotonergic antagonists, particularly the 5HT2 antagonist, blocked the reduction in FST immobility induced by the pineal implants. Adrenergic antagonists not only blocked FST immobility reductions in adrenal medullary grafted animals, but overcompensated for the adrenal transplants, producing a large increase in immobility. The FST reduction induced by pineal and adrenal cografts was blocked by all four monoaminergic antagonists. FST immobility scores in control transplanted and non-transplanted animals were not altered by any of the antagonists. The immobility reduction produced by chronic imipramine D. D. Dougherty " C. E. Sortwell - J. Sagen ( ~ ) Department of Anatomy and Cell Biology, University of Illinois at Chicago, 808 S. Wood Street, Chicago, IL 60612, USA

treatment was blocked significantly only by propranolol. These results indicate that adrenal medullary and pineal transplants produce sustained antidepressive activity via local interaction with a-and fl-adrenergic receptors or 5HT2 receptors, respectively, and may be mediated by mechanisms distinct from antidepressant drugs. Key words Pineal. Adrenal medulla Forced swimming test - Norepinephrine • Serotonin Depression

Introduction

Although many major depressive episodes subside following a brief course of psychotherapy or pharmacologic intervention, follow-up studies suggest that a large proportion of these patients are chronically depressed or have recurrent depressive episodes (Kupfer 1991). Furthermore, currently available antidepressant therapies are limited by potentially serious side effects, poor patient compliance, and intentional overdose (Tollefson 1991). A novel approach towards restoring imbalanced functioning in the CNS is the use of neural transplantation. It has been shown that it is possible to alter behavior by transplanting pharmacologically relevant tissues into appropriate CNS sites (cf. Dunnett and Richards 1990). These neural grafts can serve as a long-term and readily available source of neuroactive substances. While classical theories imply that antidepressants act via increasing synaptic availability of monoamines, particularly norepinephrine (NE) and serotonin (5HT), the delay in clinical antidepresent efficacy despite the nearly immediate increase in monoamine availability suggests that longer term equilibration in the monoarninergic systems is required to produce therapeutic benefit (Caldecott-Hazzard et al. 1991). The search for chronic effects of antidepressant drugs

11

has revealed that the sensitivity and density of monoaminergic receptors, particularly/3-adrenergic receptors, is altered with chronic antidepressant administration (Banerjee et al. 1977; Sulser et al. 1978). In addition, changes in serotonergic receptor subtypes, as well as pre- and post-synaptic ~-adrenergic receptors, have been reported (Svenson and Usdin 1978; Charney et al. 1981; Walmsley et al. 1987; Cowen 1990). Since the clinical potency of antidepressant drugs is not necessarily correlated with their receptor effects, it is unclear whether these changes are primarily responsible for the therapeutic antidepressant action, or merely circumstantial. Nevertheless, although the precise pathology of depression and the mechanism of action of antidepressants is still not understood, it is likely that effective antidepressant therapies increase the efficiency of the monoaminergic systems at functionally important sites in the brain (Siever and Davis 1985; Bryant and Brown 1986; Syvalahti 1987; Nair and Sharma 1989; Caldecott-Hazard et al. 1991). A widely used animal model, the forced swimming test (FST), can reliably screen for potentially effective antidepressant therapies (Porsolt et al. 1977; Borsini and Meli 1988). This test is most commonly used by pharmaceutical companies, as it is particularly sensitive in screening clinically eftbctive antidepressants, demonstrating significant correlation between clinical potency and potency in reducing behavioral immobility during forced swimming (Willner 1985). An additional advantage is that, in ,contrast to other animal measures of antidepressant activity, the FST allows for multiple testing of the same animal with a variety of pharmacologic agents. However, the precise role of monoamines and their receptor interactions in reducing behavioral immobility in the FST is unclear, due to a variety of seemingly contradictory findings using a wide range of pharmacologic agents. For example, while most studies suggest that antidepressants which increase synaptic NE reduce immobility, some reports find that fl-adrenergic agonists are inactive, either ~-adrenergic agonists or antagonists may reduce immobility, and the efficacy of antidepressants in reducing immobility may or may not be blocked by e- or/~-adrenergic antagonists (Porsolt et al. 1979; Borsini et al. 1981; Kitada et al. 1983; Pulvirenti and Samanin 1986; Finnegan et al. 1987; Borsini and Meli 1988; Cervo et al. 1990). Similarly, the role of 5-HT in FST immobility is uncertain, as some serotonergic uptake blockers and agonists can reduce immobility while others have no effect, and serotonergic antagonists have variable effects on antidepressants (Porsolt et al. 1979; Borsini et al. 1981; Luttinger et al. 1985; Borsini and Meli 1988; Wieland and Lucki 1990; Cervo et al. 1991). These contradictory findings may be due to the simultaneous manipulation of several active CNS sites depending on the route of administration. We have previously demonstrated in our laboratory that the transplantation of monoamine-containing

tissue directly into the frontal neocortex can reduce immobility in the FST (Sortwell and Sagen 1993). The tissue sources used included adrenal medullary tissue, pineal gland tissue, or a combination of both. The adrenal medulla is a rich source of NE-producing chromaffin cells, particularly following removal of the steroid-secreting adrenal cortex (Wurtman and AxeIrod 1966; Pohorecky and Wurtman 1971). The pineal gland contains exceptionally high levels of 5-HT (Quay and Halevy 1962; Bertler et al. 1964; Wurtman et al. 1968; Cooper et al. 1986). Similar transplants have also been found to prevent the development of learned helplessness in earlier studies (Sagen et at. 1990a). In addition to :-educing behavioral deficits, these transplants have been demonstrated to increase monoamine levels in implanted frontal cortex (Sortwell and Sagen 1993). While these findings implicate increased cortical levels of NE and/or 5-HT in the antidepressant activity of monoaminergic transplants, determination of their receptor interactions may lead to an increased understanding of antidepressant mechanisms. The goal of this study was to characterize the pharmacologic mechanisms involved in the production of antidepressant activity by monoaminergic neural transplants in the frontal cortex. Preliminary findings of this study have been reported previously (Dougherty et al. 1991).

Materials and methods Behavioral testing The use of animals in these studies was in accordance with the guidelines of the Society for Neuroscience and approved by the Animal Care Committee at the University of Illinois at Chicago. Female Sprague-Dawley rats (Sasco, Wis.) weighing 150-200 g were used for these studies. Within 1 week prior to surgery, the animals underwent forced swim baseline pretesting (conditioning) and testing according to a modification of the protocol described previously (Porsolt et al. 1977; Sortwell and Sagen 1993). The rats were placed in a clear plastic cylinder filled with water at room temperature (25°C) at a depth of 20cm. During the pretesting session, the animals were placed in the cylinder for 15 min to acclimate them to the swimming conditions. Twenty-four hours later, the animals were again placed in the cylinder of water, this time for a 5-rain testing session. During this session the animals were observed for the readily recognizable immobile posture, a characteristic position consisting of an arched back and a lack of movement except to maintain the head above water while floating in the water-filled cylinder. The time spent in this immobile posture was recorded. Since increased locomotor activity can result in false positives in the FST, animals were placed in a locomotor activity chamber (69 x 63 x 49.5 cm) with floor grid markings (12 x 12 cm). The number of grid crossings made by each rat was recorded over a 5-rain period, similar to that described by others (Porsolt et al. 1979). Locomotor activity was assessed 24 h after FST testing, before and 6 weeks after transplantation. Since direct neocortical effects of NE or 5-HT on immobility have not previously been published, as a prelude to our monoaminergic transplant studies, it was necessary to confirm that the chosen neocorticat region was an appropriate site for local monoaminergic activity. In order to accomplish this preliminary study, a group of rats were equipped with guide cannulae aimed at the t¥ontal

12 neocortex, a site shown by other laboratories to be involved in antidepressant activity as assessed by the learned helplessness test (Petty and Sherman 1980). Cannulae were implanted 1 week prior to FST conditioning and testing (sterotaxic coordinates: A 3.0 mm, L 2.0 mm, H - 3.0 mm, incisor bar - 2.5 mm from Bregma), and anchored to the skull with dental acrylic. One day following FST conditioning, animals received NE (10 gg), 5-HT (10 l.tg) or vehicle (0.02% ascorbic acid in saline) in 1.0 gl delivered via a microinjector inserted through the guide cannula. FST testing was conducted 5 min following the microinjections.

Monoaminergic transplants Transplant tissue included pineal gland, adrenal medulla, a combination of both pineal and adrenal medulla, and a sciatic nerve group to serve as a non-monoaminergic control. In addition, the sciatic nerve transplants could serve as an additional control for trophic factor effects of the transplants, since Schwann cells produce several trophic factors (Heumann et al. 1987). Adrenal glands were obtained from donor rats of the same strain (female Sprague-Dawley, 150-200 g) and the adrenal medulla was carefully dissected from cortical tissue and cut into small pieces (< 0.5 mm 3) while submerged in ice-cold Hank's buffer. Whole pineal glands were excised fiom adult female donors. Sciatic nerve tissue was also dissected from adult female donors, the sheath removed, and the nerve cut into 0.5-mm pieces. For implantation, tissue was then loaded into a cannula and stereotaxicalty implanted into the frontal neocortex of an anesthetized animal (Nembutal 30 mg/kg, IP; stereotaxic coordinates: A 3.0 ram, L 2.0 mm, H -1.0 mm, incisor bar - 2 . 5 mm, from Bregma). These coordinates result in placement of the graft into the right frontal neocortex, an area that has proven effective in prior studies in our laboratory for producing behavioral effects (Sagen et al. 1990a; Sortwell and Sagen 1993). In order to keep the volume of the transplants constant, animals received tissue from approximately two-thirds of an adrenal medulla (n = 20 animals), a whole pineal gland (n = 19 animals), or one half of each of these amounts of monoaminergic tissues for co-grafts (n = 20 animals). Controls received equal volumes of sciatic nerve tissue (n = 17 animals). Animals were allowed to recover under a heat lamp following implantation. In addition to the four transplantation groups, two other groups of animals received no transplant o1: surgical procedures and were given either imipramine daily (HC1, 15 mg/kg, IP, n = 15 animals) for 6 weeks or no antidepressant treatment (n = 20 animals). In imipramine-treated animals, daily injections began on the same day on which the transplant groups received surgery, following pretesting. Daily imipramine treatment continued through antagonist testing (see below). Imipramine was prepared fresh daily, and was administered in the morning (9 a.m.). All training and testing sessions took place in the early afternoon, approximately 5 h following imipramine injection.

Suffern, N.Y.; 10.0 mg/kg, SC], an ~-adrenergic antagonist. The dose of phentolamine was chosen since it has produced attenuation of adrenal medullary transplant effects in previous studies in our laboratory (Sagen et al. 1990b), while the propranolol dose chosen was that reported by other laboratories to reduce the effects of antidepressants on immobility in the FST (Borsini et al. I981, 1985; Finnegan et al. 1987; Mancinelli et al. 1991). Doses for the serotonergic antagonists were chosen based on previous reports demonstrating maximal 5-HT antagonism without confounding agonist activity (Colpaert et al. 1982; Colpaert and Janssen 1983). All drugs were dissolved in saline and injected 30 min prior to 5-rain swim testing. The systemic route of administration was chosen to allow for repeated testing of the same transplanted animals with the various antagonists, in order to avoid potential damage to the transplanted cells or surrounding host perenchyma that may result from repeated intracerebral microinjection. All animals received the same drug on a given testing day, separated by an interval of 48 h between drug testing sessions to allow for complete drug elimination. Thus, each animal was tested in the FST a total of 6 times (before transplantation, after transplantation, and in the presence of each of the four antagonists). Drugs were given in the following order: metergoline, pirenperone, propranolol, and phentolamine, and FST scores were rated by the same blind observer. Since the interval between testing sessions was short, a pretesting session was only used once posttransplantation.

Morphological and statistical analysis Following the conclusion of pharmacologic testing, the animals were killed and the transplants were evaluated immunocytochemicallyto verify the viability of monoaminergic cells. Animals were deeply anesthetized (50 mg/kg Nembutal IP), and perfused intracardiatly with saline, followed by 4% paraformaldehyde in 0.1 M phosphate buffer. Tissue containing the transplants was dissected and placed in a 20% sucrose solution overnight for cryoprotection. Twenty-micron sections were cut on a cryostat, incubated in phosphatebuffered saline containing 1% normal goat serum, and placed in primary antibody solution (tyrosine hydroxylase antibody, Incstar, diluted 1:500; 5-HT antibody, lncstar, diluted 1:500; or control pre-immune serum) overnight at 4°C. Following several washes, a fluorescein- or rhodamine-linked secondary antibody (diluted 1: 100) was applied for 1 h at room temperature. The sections were washed, coverslipped in Fluoromount, and observed in a Zeiss Axiophot microscope. Statistical comparisons between treatment groups were done using ANOVA and Duncan's multiple range test (Tallarida and Murray 1987). Repeated measures ANOVA was used for comparisons within treatment groups.

Results Testing protocol Within 3-5 days after obtaining baseline immobility scores, the animals were transplanted with either monoaminergic or control tissues as described above. Six weeks after transplantation surgery, the rats again underwent forced swimming conditioning and testing in the same manner as before the surgery. To assess the pharmacologic specificity of the transplants, immobility scores were then recorded after pretreatment with one of the following monoaminergic antagonists: metergoline (kindly supplied by Famitalia, Milan, Italy; 0.16 mg/kg, SC), a 5-HT~/5-HT~ antagonist; pirenperone (Research Biochemicals International,Natick, Mass.; 0.16 mg/kg, SC), a 5-HT2 antagonist; d,l-propranolol (HC1, Research Biochemicals International, Natick, Mass.; 5.0 mg/kg, SC), a fl-adrenergic antagonist; and phentolamine [mesylate (Regitine), kindly supplied by Ciba-Geigy,

T h e direct m i c r o i n j e c t i o n of either 5 - H T or N E c o n firmed o u r choice of the f r o n t a l n e o c o r t e x as a n active site for m o n o a m i n e s in the F S T . I m m o b i l i t y scores were 54.6 + 4.3 (n = 5) a n d 58.4 + 7.4 (n = 5) for N E i n j e c t e d a n d 5 - H T - i n j e c t e d a n i m a l s , c o m p a r e d to 127.2 __ 7.9 (n = 5) for the v e h i c l e - i n j e c t e d g r o u p . B o t h 5 - H T a n d N E s i g n i f i c a n t l y r e d u c e d i m m o b i l i t y scores c o m p a r e d to vehicle [ F ( d f 3,12) = 38.67; P < 0.01]. As d e s c r i b e d b y earlier studies (Sortwell a n d S a g e n 1993), all three m o n o a m i n e r g i c t r a n s p l a n t s p r o d u c e d significant r e d u c t i o n s i n i m m o b i l i t y 6 weeks p o s t - t r a n s p l a n t a t i o n [Fig. 1; o v e r a l l F (df 5,18) = 3.9 t; p r e - v e r s u s p o s t - P < 0.05 for p i n e a l t r a n s p l a n t s ] (Fig. t A ) a n d

13 PINEAL TRANSPLANTS

A

PRE

B

ADRENAL

PRE

C

POST

MET

PIR

MEDULLARY

POST

MET

PIR

PROP

PHEN

TRANSPLANTS

PROP

PIIEN

ADRENAL/PINEAL CO-TRANSPLANTS

PRE

POST

MET

PIR

PROP PHEN

Fig. 1 A-C The effect of adrenergic and serotonergic antagonists on FST immobility reductions induced by monoaminergic transplants in the frontal neocortex. A FST immobility scores from pineal implanted rats (n = 19). B FST immobility scores from adrenal medullary implanted rats (n = 20). C FST immobility scores from adrenal and pineal cografted rats (n = 20). FST scores are shown pre-transplantation (PRE), 6 weeks post-transplantation (POST), and 30 rain following systemic administration of metergoline (MET), pirenperone (PIR), propranolol (PROP), or phentolamine (PHEN). Values represent the mean _+ SEM for each group. + P < 0.05, + + P < 0.01 compared to pre-transplant scores; *P < 0.05, **P < 0.01 compared to post-transplant scores

adrenal-pineal co-transplants [overall F (df 5,19)= 3.09; Fig. 1C] and P < 0.01 for adrenal transplants [overall F (df 5,19)= 14.74; Fig. 1B] compared to pre-transplant immobility scores. Locomotor activity was not significantly altered in any of the groups (number of crossings = 75.8 _+ 12.1, pretransplant; 63.0 _+ 10.8, pineal transplant group; 66.8 + 10.7, adrenal transplant group; 75.2 + 14.8, co-transplant group; 67.8 + 11.2, sciatic nerve transplant group;

82.7 _+ 14.2, non-transplant group; 70.4 __ 13.8, imipramine group; P > 0.05). Both serotonergic antagonists, metergoline and pirenpcrone, reversed the reduction in immobility by pineal transplants (Fig. 1A; P < 0.05 compared to post-transplant immobility scores). In addition, both of these agents reversed immobility scores in pineal-transplanted animals to pre-transplant values (P > 0.05 compared to pre-transplant scores). Pirenperone appeared to be slightly more effective than metergoline, although this difference was not significant (P > 0.05). The e-adrenergic antagonist phentolamine did not significantly alter immobility scores in pineal-transplanted animals, while the /%adrenergic antagonist propranolol produced a slight, but statistically insignificant reversal in these animals (P > 0.05). In contrast to pineal-transplanted animals, the 5-HT antagonists did not significantly alter the reduction in immobility produced by adrenal medullary transplants (Fig. 1B; P > 0.05 compared to post-transplant scores). However, both adrenergic agents significantly reversed the reduction in immobility in these animals (P < 0.01 compared to post-transplant scores). Interestingly, both propranotol and phentolamine not only blocked the reduced immobility in adrenal medullary-transplanted animals, but resulted in even more pronounced immobility than that found prior to transplantation (P < 0.05 compared to pre-transplant scores). In animals receiving co-transplants of both pineal and adrenal medullary tissue, the reduction in immobility by the transplants was reversed by both serotonergic and adrenergic antagonists (Fig. 1C, P < 0.05 for all four agents compared to post-transplant scores). In addition, immobility scores following pretreatment with any of these agents were not significantly different than scores obtained prior to transplantation (P>0.05 compared to pre-transplant scores). As a non-monoaminergic transplant that may also produce trophic factor effects, some animals received sciatic nerve transplants. As reported in previous studies (Sortwell and Sagen 1993), these transplants appeared to produce a slight, but statistically insignificant, reduction in immobility scores [Fig. 2A; overall F (df 5,16)= 0.69; P > 0.05 compared to pre-transplant scores]. All three monoaminergic transplants were significantly more effective in reducing immobility scores (P < 0.05 compared to sciatic nerve transplants). None of the antagonists altered either the pre- or posttransplant immobility scores in animals with sciatic nerve transplants (P > 0.05). To control for possible antagonist effects on immobility behavior in the absence of neural transplants, unimplanted animals were tested according to the same schedule. None of the agents at the doses tested significantly altered immobility scores in these animals [-Fig. 2B; overall F (df 5,19)= 0.22; P > 0.05]. Finally, to compare monoaminergic transplant eftects with a traditional

14

A

SCIATIC NERVE TRANSPLANT

o

PRE

B

POST

MET

PIR

PROP PHEN

NO TRANSPLANT

ency to reverse the imipramine-induced reduction in immobility, only propranolol produced a statistically significant reversal (P < 0.05 compared to imipramine alone scores). This finding may be due to the high variability in the effectiveness of imipramine in reducing immobility. Immunocytochemical analysis of the transplants revealed robust survival of adrenal medullary, pineal gland and cografted transplants at least 2 months following transplantation. The pineal gland grafts contained densely packed 5-HT positive cells (Fig. 3A) that did not stain for tyrosine hydroxylase. Adrenal medullary grafts displayed distinct tyrosine hydroxylase staining (Fig. 3B), although the implant was often encapsulated by a collagen matrix, masking detection of individual chromaffin cells in some portions of the graft.

Discussion PRE

C

POST

MET

PIR

PROP PHEN

IMIPRAMINE

PRE

POST

MET

PIR

PROP PHEN

Fig. 2 A-C The effect of adrenergic and serotonergic antagonists on FST immobility following: A Sciatic nerve transplantation (n = 17), B No transplant (n = 20), and C Chronic treatment with imipramine (n = 15). Testing conditions included baseline or pretransplantation scores (PRE), scores 6 weeks later or 6 weeks following transplantation (POST),and 30 min following systemic administration of metergoline (MET), pirenperone (PIR), propranolol (PROP), or phentolamine (PHEN). Values represent the mean 4_:SEM for each group. + P < 0.05, + + P < 0.01 compared to pre-transplant scores; *P < 0.05, **P < 0.01 compared to posttransplant scores

antidepressant agent, some non-transplanted animals received chronic imipramine treatment. This treatment resulted in reduced immobility scores [Fig. 2C; overall F (df 5,14) = 3.01; P > 0.05 compared to pre-injection score@ When compared with the transplant groups, imipramine treatment produced reductions in immobility similar to all three monoaminergic transplant groups (P > 0.05), and was significantly more effective than control sciatic nerve transplants (P < 0.05). While all of the monoaminergic antagonists showed a tend-

The results of these studies demonstrate that local increases in NE or 5-HT in the frontal neocortex of rats by either direct microinjection, or from transplanted monoaminergic cells, can reduce immobility in the FST. These results support previous findings in our laboratory which have suggested that monoaminergic transplants can act as antidepressants in rodent models by reducing behavioral deficits in both the learned helplessness (Sagen et al. 1990a) and forced swimming tests (Sortwell and Sagen t993). Furthermore, the effectiveness of these monoaminergic transplants in reducing immobility was comparable to that obtained by chronic daily injections of the tricyclic antidepressant imipramine for a 6-week period. The potential advantage of a transplant approach for antidepressant therapy is the ability to provide a continually renewable and sustainable local level of monoamines without the need for repeated pharmacotherapies which are often confounded by serious side effects, poor patient compliance, and intentional overdose. However, before novel therapeutic approaches can be considered, a full understanding of the mechanisms of antidepressant activity, as well as potential limitations, is necessary. The goal of this study was to evaluate the pharmacologic mechanisms of monoaminergic transplants using a reliable animal model for screening antidepressant activity, the FST. Since previous findings in our laboratory have demonstrated increased cortical levels of either 5-HT or NE in regions near pineal or adrenal medullary transplants, respectively (Sortwell and Sagen 1993), and a wealth of studies in many other laboratories have implicated a role for these monoamines in depression, it is logical to suspect that the antidepressant effects provided by the monoaminergic transplants are via release of 5-HT or NE from the transplanted cells. In order to determine this

15 Fig. 3 APineal gland transplant in the frontal cortex 8 weeks after transplantation. The cells in the pineal graft are stained with a 5-HT antibody and a fluorescein-linked secondary antibody. B Adrenal medullary transplant in the frontal cortex 8 weeks after transplantation. The cells in the adrenal graft are stained with a tyrosine hydroxylase antibody and a rhodamine-linked secondary antibody

possible mechanism in the present study, selective serotonergic and adrenergic antagonists were used. Results of this study strongly implicate serotonergic mechanisms for the pineal transplants, adrenergic mechanisms for the adrenal medullary transplants, and a combination of serotonergic and adrenergic mechanisms for the co-transplants. The findings that both ~-adrenergic and ]~-adrenergic antagonists can block reduced immobility in adrenal medullary-implanted animals, but have no effect on pineal- or control-implanted animals suggest that adrenal medullary transplants produce their effects by interactions with both e- and /~-adrenergic receptors. These findings are interesting, as both e-adrenergic and ~-adrenergic mechanisms have been implicated in depression and antidepressant therapies. For example, the down-regulation in/~-receptor sensitivity is a well known consequence of numerous antidepressant therapies (Sulser 1983). In addition, CNS e2-adrenergic receptors, which may be pre-or post-synaptic, are reduced, while ~:q-adrenergic receptors show increased sensitivity with chronic administration of antidepressants (Svenson and Usdin 1978; Charney et al. 1981; Plaznik et al. 1984; Caldecott-Hazard et al. 1991). Agents acting directly at these receptors do not produce clear effects either clinically or in animal models, which may be partially due to low penetration of the blood-brain barrier or low activity at CNS receptor sites. Nevertheless, some studies have reported reduced immobility in the FST with either B-adrenergic (Finnegan et al. 1987) or ~-adrenergic (Porsolt 1979; Hasco& et al. 1991) agonists, and antagonists of either of these

receptors can block the anti-immobility effects of some antidepressants (Borsini et al. 1981, 1985; Kitada et al. 1983; Cervo et al. 1990; Mancinelli et al. 1990). Similarly, our findings indicate that adrenergic receptor antagonists, particularly the ~-adrenergic antagonist propranolol, can reverse the reduced immobility resulting from chronic imipramine treatment. Thus, one potential mechanism of antidepressant activity of adrenal medullary transplants is via interaction with host c~- and ~-adrenergic receptors. Another possibility is that the activity of the transplants is modulated by adrenergic autoreceptors on the transplanted chromaffin cells. In particular, chromaffin cell ~-adrenergic receptors may play a role in positive feedback mechanisms, since ~-adrenergic agonist stimulate, while ~-adrenergic antagonists such as propranolol inhibit, catecholamine secretion (Greenberg and Zinder 1982). Similarly, while ~2-adrenergic agonists seem to be involved in negative feedback mechanisms, phentolamine paradoxically produces significant reduction in catecholamine secretion from chromaffin cells (Wakade et al. 1986). Thus, in addition to blocking host adrenergic receptors, e- and ~-adrenergic antagonists may also produce their effects by reducing catecholamine release from the transplanted chromaffin cells. This dual action (at host and graft receptors) may also account for the observed °°overshoot" in immobility following adrenergic antagonists in animals with adrenal medullary transplants. Rather than simply block the anti-immobility effects of the transplants and bring the immobility scores to pre-transplant control

t6 levels, the presence of either e- and/~-adrenergic antagonists exacerbated the immobility in animals with adrenal medullary transplants. One explanation for this result may be that a compensatory host cortical receptor down-regulation has occurred, and thus the remaining active receptors are more readily blocked by antagonists. The inability to detect a similar overshoot in imipramine-treated animals, where /~-receptor down-regulation has been well established, argues against this explanation. However, a combination of host receptor down-regulation with direct reduction of catecholamine release from the transplanted cells in the presence of adrenergic antagonists may exacerbate a synaptic deficiency in catecholamine neurotransmission. Alternatively, since grafted adrenal chromatfin cells produce other neuroactive substances in addition to catecholamines, notably neuropeptides (Ortega and Sagen, 1993), the possibility that blocking the adrenergic effects of the transplanted chromaffin cells unmasks effects of other neuroactive agents cannot be overlooked. In addition to adrenal medullary transplants, pineal transplants also reduced immobility in the FST. The anti-immobility effects of pineal transplants were not altered by adrenergic antagonists, but were blocked by serotonergic antagonists, The greater effectiveness of the relatively selective 5-HT2 antagonist pirenperone in producing this effect compared to the 5-HT1/5-HT2 antagonist metergoline suggests that pineal transplants may mediate their antidepressant effects via interaction with 5-HT2 receptors. In support for this possibility is the lack of effect of propranolol in pineal-transplanted animals, since propranolol has been reported to block 5-HT~ receptors (Gilman et al. 1990; Mancinelli et al. 1991). The potential role of 5-HT2 receptors is interesting, as one of the most consistent effects of antidepressant drugs is 5-HT2 receptor down-regulation (Cowen 1990). In addition, these findings suggest that pineal transplants reduce immobility via mechanisms distinct from other antidepressants, since 5-HT2 antagonists do not appear to block the anti-immobility effects of antidepressant drugs in the FST (Borsini et at. 1981; Cervo et al. 1991; and results from present studies). The possibility that the antidepressive mechanisms of the transplants are distinct from those of antidepressant drugs such as imipramine is supported by preliminary studies in our laboratory indicating a dissociation between monoaminergic receptor changes in the neocortex of animals with transplants and chronic antidepressant treatment (Sagen et al. 1994). However, these results must be interpreted with caution, as 5-HT2 antagonists, e.g. mianserin, have been shown to themselves have antidepressant activity (Blackshear and Sanders-Bush 1982). Since 5-HT2 antagonists did not themselves affect the FST in the present study, the test may not be sensitive to some types of clinically effective antidepressants. Finally, the results from the co-transplants of adrenal medullary and pineal tissue are somewhat surprising,

as the reduction in immobility could be reversed by both adrenergic and serotonergic antagonists. These results may be due to the lower level of both monoamines provided by the smaller amounts of each tissue type in the co-transplants, resulting in an inability for one to compensate during blockade of the other. Alternatively, it is possible that the presence of both tissues results in dependent co-activation of monoamine release. As an example, the in situ pineal gland normally receives sympathetic input from the superior cervical ganglion, which activates /~-adrenergic (and also ~l-adrenergic) pinealocyte receptors (Morgan et al. 1989; Santana et al. 1989), and it is likely that catecholamine-releasing chromaffin cells in the co-transplants can stimulate release of 5-HT from the transplanted pinealocytes. It is possible that repeated testing in the FST could result in a trend towards increasing immobility scores, and that this effect accounts in part for the apparent effects of propranolol and phentolamine in adrenal medullary and adrenal/pineal co-grafted animals. However, this explanation seems unlikely, as this trend is not observed in non-transplanted or control transplanted animals. Another possible confounding variable is the potential interference in FST scores by alterations in locomotor activity. Using a locomotor test, no apparent changes in locomotor activity were observed in any of the transplant groups. However, while these results suggest that changes in motor activity do not account for changes in the FST, nevertheless the possibility cannot be completely eliminated by this simple locomotor test. In addition, since the transplants were placed unilaterally, it is possible that FST effects obtained represent motor effects produced by hemispheric asymmetries. This explanation seems unlikely, since similar changes were not observed in control transplanted animals. Finally, it should be noted that the immobility scores obtained were lower than that reported by others using the FST. This observation is most likely due to differences in water depth, and is similar to results obtained previously (Sortwell et al. 1993). Similarly, others have noted that water depth may alter FST scores, since deeper water limits the animal's ability to stay afloat and reduces immobility (Borsini et al. 1986). In summary, results from these studies supports a role for both serotonergic and noradrenergic mechanisms in antidepressant activity as assessed by the FST. In addition, these monoamines can be continually provided to critical areas of the CNS such as the frontal neocortex via the transplantation of monoaminergic tissues such as adrenal medulla or pineal gland. These findings suggest that adrenal medullary and pineal transplants reduce behavioral immobility via interaction with c~-and/~-adrenergic receptors or 5-HT2 receptors, respectively. In conjunction with other studies in our laboratory, these results indicate that monoaminergic transplants may produce sustained

17

antidepressant activity via local interactions with monoaminergic transmission in the frontal neocortex. AcknowledgementThis work was supported by NIMH MH47491. References Banerjee SP, Kung LS, Riggi SJ, Chanda SK (1977) Development of beta-adrenergic subsensitivity by antidepressants. Nature 288:455-456 Bertler A, Falck B, Owmann C (1964) Studies on the 5-hydroxytryptamine stores in the pineal gland of the rat. Acta Physiol Scand 63:1-18 Blackshear MA, Sanders-Bush E (1982) Serotonin receptor sensitivity after acute and chronic treatment with mianserin. J Pharmacol Exp Thor 221:303 308 Borsini F, Meli A (1988) Is the forced swimming test a suitable model for revealing antidepressant activity? Psycholpharmacology 94:147-160 Borsini F, Bendotti C, Velkov V, Rech R, Saminin R (1981) Immobility test: effects of 5-hydroxytryptaminergic drugs and role of catecholamines in the activity of some antidepressants. J Pharm Pharmacol 33:33-37 Borsini F, Nowakowska E, Pulvirenti L, Samanin R (1985) Repeated treatment with amitriptyline reduces immobility in the behavioral "despair" test in rats by activating dopaminergic and fladrenergic mechanisms. J Pharm Pharmacol 37:137 138 Borsini F, Bolterra G, Meli A (1986) Does the behavioral "despair" test measure despair. Physiol Behav 38:385 386 Bryant SG, Brown CS (1986) Current concepts in clinical therapeutics: major affective disorder, part 1. Clin Phannacol 5:304-318 Caldecott-Hazard S, Morgan DG, DeLeon-Jones F, Overstreet DH, Janowsky D (I991) Clinical and biochemical aspects of depressive disorders: II. Transmitter/receptor theories. Synapse 9:251301 Cervo L, Grignaschi G, Samanin R (1990) ~2-Adrenoceptor blockade prevents the effect of desipramine in the forced swimming test. Eur J Pharmacol 175:301-307 Cervo L, Grignaschi G, Rossi C, Samanin R (1991) Role of central serotonergic neurons in the effect of sertraline in rats in the forced swimming test. Eur J Pharmacol 196:217-222 Charney DS, Menkes DB, Heninger GR (1981) Receptor sensitivity and the mechanism of action of antidepressant treatment. Arch Gen Psychiatry 38:1160-1180 Colpaert FC, Janssen PAJ (1983) The head-twitch response to intraperitoneal injection of 5-hydroxytryptophan in the rat: antagonist effects of purported 5-hydroxytryptamine antagonists and of pirenperone, an LSD antagonist. Neuropharmacology 22:993-1000 Colpaert FC, Niemegeers JE, Janssen PAJ (1982) A drug discrimination analysis of lysergic acid diethylamide (LSD): In vivo agonist and antagonist effects of purported 5-hydroxytryptamine antagonists and of pirenperone, a LSD-antagonist. J Pharmacol Exp Thor 221:206 .214 Cooper JR, Bloom FE, Roth RH (1986) The biochemical basis of neuropharmacology. Oxford University Press, New York Cowen PJ (1990) A role for 5-HT in the action of antidepressant drugs. Pharmacol Thor 46:42-51 Dougherty DD, SortwelI CE, Sagen J (1991) Pharmacologic specificity of reduced immobility in a rodent depression model by monoaminergic neural transplants. Soc Neurosci Abstr 17:51 Dunnett SB, Richards SJ (eds) (1990) Progress in brain research, vol 82, neural transplantation: from molecular basis to clinical applications. Elsevier, New York Finnegan KT, Terwilliger MM, Berger PA, Hollister LE, Csernansky JG (1987) A comparison of the neurochemical and behavioral effects of clenbuterol and desipramine. Eur J Pharmacol 134:131-136

Gilman AG, Rall TW, Nies AS, Taylor P (t990) Goodman and Gilman's The pharmacological basis of therapeutics. Pergamon, New York Greenberg A, Zinder O (1982) e- and fl-receptor control of catecholamine secretion from isolated adrenal medulla cells. Cell Tiss Res 226:655 665 Hasco~t M, Bourin M, Bradwejn J (1991) Behavioral models in mice. Implication of the alpha noradrenergic system. Prog Neuro/Psychopharmacol Biol Psychiary 15:825-840 Heumann R, Korsching S, Bandtlow C, Thoenen H (1987) Changes of nerve growth factor synthesis in non-neuronal cells in response to sciatic nerve transection. J Cell Biol 104:1623d631 Kitada Y, Miyauchi T, Kanazawa H, Nakamich H, Satoh S (1983) Involvement of c~-and/3 l"adrenergic mechanisms in the immobility-reducing action of desipramine in the forced swimming test. Neuropharmacology 22:1055-1060 Kupfer DJ (1991) Long-term treatment of depression. J Clin Psychiatary 52:28 34 Luttinger D, Freedman M, Hamel L, Ward SJ, Perrone M (1985) The effects of serotonin antagonists in a behavioral despair procedure in mice. Eur J Pharmacol 107:53-58 Mancinelli A, D'Aranno V, Stasi MA, Lecci A, Borsini F, Meli A (1991) Effect of enantiomers of propranolol on desipramineinduced anti-mobility in the forced swimming test in the rat. Pharmacol Res 23:4~50 Morgan PJ, Davidson GR, Lawson W (1989) Evidence for dual adrenergic receptor regulation of ovine pineal function. J Pineal Res 7:175-183 Nair NPV, Sharma M (1989) Neurochemical and receptor theories of depression. Psychiatr J Univ Ottawa 14:328 341 Ortega JA, Sagen J (1993) Pharmacological characterization of opioid peptide release from chromaffin cell transplants using a brain slice perfusion method. Exp Brain Res 95:381387 Petty F, Sherman AD (1980) Regional aspects of the prevention of learned helplessness by desipramine. Life Sci 26:1447-1452 Plaznik A, Danyssz W, Kostowski W (1984) Behavioral evidence for alpha-l-adrenoceptors up- and alpha-2-adreneoceptor down-regulation in the rat hippocampus after chronic imipramine treatment. Eur J Pharmacol 101:305-306 Pohorecky LA, Wurtman RJ (1971) Adrenocortical control of epinephrine synthesis. Pharmacol Rev 23:1-35 Porsolt RD, LePichon M, Jalfre M (1977) Depression: a new model sensitive to antidepressant treatments. Nature 266:730-732 Porsolt RD, Bertin A, Blavit N, Deniel M, Jalfre M (1979) Immobility induced by forced swimming in rats: effects of agents which modify central catecholamine and serotonin activity. Eur J Pharmacol 57:201 210 Pulvirenti L, Samanin R (1986) Antagonism by dopamine, but not noradrenaline receptor blockers of the anti-immobility activity of desipramine after different treatment schedules in the rat. Pharmacol Res Commun 18:73-80 Quay WB, HaIevy A (1962) Experimental modifications of the rat pineal content of serotonin and related indole amines. Physiol Zool 35:1-7 Sagen J, Sortwell CE, Pappas GD (1990a) Monoaminergic neural transplants prevent learned helplessness in a rat depression model. Biol Psychiatry 28:1037--1048 Sagen J, Wang H, Pappas GD (1990b) Adrenat medullary implants in the rat spinal cord reduce nociception in a chronic pain model. Pain 42: 69-79 Sagen J, Dougherty D., Sortwell CE, UnnerstaI1 JR (1994) Dissociation between antidepressant effects and monoaminergic receptor changes in rats with monoaminergic transplants in the frontal neocortex. 3rd Int Behav Neurosci Abstr (in press) Santana C, Guerrero JM, Reiter RJ, Menendez-Pelaez (1989) Role of postsynaptic c~-adrenergic receptors in the fl-adrenergic stimulation of melatonin production in the Syrian hamster pineal gland in organ cutture. J Pineal Res 7:13-22

18 Siever LJ, Davis KL (1985) Overview: toward a dysregulation hypothesis of depression. Am J Psychatry 142:1017-1031 Sortwell CE, Sagen J (1993) Induction of antidepressive activity by monoaminergic transplants in rat neocortex. Pharmacol Biochem Behav 46: 225-230 Sulser F (1983) Mode of action of antidepressant drugs. J Clin Psychiat 44:14-20 Sulser F, Vetulani J, MoNey PI (1978) Mode of action of antidepressant drugs. Biochem Pharmacol 27:257-261 Svenson TH, Usdin T (1978) Feedback inhibition of brain noradrenaline neurons by tricyclic antidepressants: alpha-receptor mediation. Science 202:1089-1091 Syvalahti E (1987) Monoaminergic mechanisms in affective disorders. Med Biol 65:89 96 Tallarida RJ, Murray RB (1987) Manual of pharmacologic calculations with computer programs, 2nd edn. Springer, New York Berlin Heidelberg Tollefson GD (1991) Antidepressant treatment and side effect considerations. J Clin Psychiatry 52:4-13

Wakade AR, Malhotra RK, Wakade TD, Dixon WR (1986) Simultaneous secretion of catecholamines from the adrenal medulla and of [3H]norepinephrine from sympathetic nerves from a single test preparation: Different effects of agents on the secretion. Neuroscience 18:877-888 Walmsley JK, Byerley WF, McCabe RT, McCabe RT, McConnell EJ, Dawson TM, Grosser BI (1987) Receptor alterations associated with serotonergic agents: an autoradiographic analysis. J Clin Psychiatry 48:19-25 Wieland S, Lucki I (1990) Antidepressant-like activity of 5-HT1A agonists measured with the forced swim test. Psychopharmacology 101:497-504 Willner P (1985) Depression: a psychobiological synthesis. Wiley, New York Wurtman RJ, Axetrod J (1966) Control of enzymatic synthesis of adrenaline in the adrenal medulla by adrenal corticalsteroids. J Biol Chem 241:2301-2305 Wurtman RJ, Axelrod J, Kelly DE (1968) The pineal. Academic Press, New York