- Email: [email protected]
5-HT3 receptor-active drugs alter development of spinal serotonergic innervation: lack of effect of other serotonergic agents
Brain Research, 571 (1992) 293-297 © 1992 Elsevier Science Publishers B.V. All rights reserved. 0006-8993/921505.00
293
BRES 17407
5-HT 3
receptor-active drugs alter development of spinal serotonergic innervation lack of effect of other serotonergic agents James Bell III, Xini Zhang and Patricia M. Whitaker-Azmitia
Department of Psychiatry and Behavioral Sciences, State University of New York, Stony Brook, NY 11794 (U.S,A.) (Accepted 17 September 1991)
Key words: Serotonin; Neurotransmitter and development; Serotonin3 receptor; Spinal cord; Nociception; Ontogeny
Our work has focused on identifying the type of serotonin receptor through which serotonin acts as a developmental signal in the central nervous system. Previously, we have found that the regulation of development of ascending serotonergic neurons is through the balance of two serotonin receptors. One, the 5-HTxa receptor, releases a growth factor from astroglial cells. The other receptor is related to a releaseregulating autoreceptor and can be stimulated indirectly by serotonin releasers such as fenfluramine. In the present study, we examined the receptors which regulate development of the descending neurons by treating pregnant rats with selective serotonergic drugs, from gestation day 12 until birth. Pups were subsequently tested for alterations in development by nociceptive testing (tail-flick latency) and by determining the binding of 3H-paroxetine, an indicator of serotonin terminal density, in spinal cord. Our results show that agents stimulating the 5-HTla receptor (8-OH-DPAT) or the 5-HTlb receptor (TFMPP) or substances which release serotonin (fenfluramine) had no effect on the development of spinal serotonergic pathways. However, agents acting on the 5-HT3 receptor did - the agonist phenylbiguanide (PG) increased latency on tail-flick testing (postnatal days 10 and 30), while the antagonist, MDL 72222, decreased latency (postnatal days 10 and 18). Interestingly, both the agonist and the antagonist significantlyincreased 3H-paroxetine binding on postnatal day 18. Our results are discussed in terms of a possible mechanism by which 5-HT3 receptors may influence development. INTRODUCTION
terminal density, the binding of 3H-paroxetine, a 5-HT uptake inhibitor. O u r findings show that the development of this system can be altered by prenatal treatment
Serotonin has been established to play a role in regulating the development of the central nervous system, prior to assuming its role as a neurotransmitter in the mature brain 9'21. The function of serotonin is assumed
with agents acting on the 5-HT 3 receptor.
to take place through the action of specific serotonin receptors, some of which may be the same or similar to those found in the mature brain.
Subjects
In previous studies, we have found that the developm e n t of ascending serotonergic pathways is regulated through the balance of two high-affinity serotonin receptors 19, one of which is related to the 5-HTla receptor and releases growth factors 22, whilst the other is related to a receptor which regulates the release of serotonin 2°. These results were supported by a variety of investigations, including tissue culture and whole animal studies involving both neurochemical and neurobehavioral measures 16' 21
In the current project, we are for the first time examining the type of serotonin receptors which regulate the development of descending serotonergic pathways. The functional integrity of this system was investigated using a behavioral test (tail-flick) and a measure of serotonin
MATERIALS AND METHODS Primiparous Sprague-Dawley rats (Taconic Farms) were maintained in individual cages on a 12-h light/dark cycle with free access to food and water, within a humidity- and temperature-controlled colony room. Maternal dams O-F/treatment group) were given the following treatments: a 5-HT3 agonist, phenylbiguanide (PG) (5 mg/kg), a 5-HTIa agonist, 8-OH-DPAT (1 mg/kg), a 5-HTlb agonist, trifluoromethylphenylpiperazine (TFMPP) (1 mg/kg), the general 5-HT releaser, fenfluramine (5 mg/kg), a 5-HT3 antagonist MDL-72222 (5 mg/kg) and dimethyl sulfoxide (DMSO) or saline as a control. All drugs were administered to the dam s.c., in the nape of the neck, daily, starting on gestation day 12 and continuing to the day of birth. Neonates from each group were sacrificed on postnatal days (D) D18 and D30 for neurochemical analysis by 3H-paroxetine binding, or assessed for nociceptive alteration using the tail-flick latency test on D10, D18 and D30. Dams were weighed throughout pregnancy. Pups were weighed on D1, D5, D10, D18 and D30.
Drug preparation MDL-72222 was dissolved in 100% DMSO, all other drugs were dissolved in physiological saline. Each drug solution was prepared in 40 ml volumes which were aliquoted into individual 1 ml samples, frozen the first day of the study and thawed 30 rain prior to
Correspondence: P.M. Whitaker-Azmitia, Department of Psychiatry and Behavioral Sciences, State University of New York, Stony Brook, NY 11794, U.S.A.
294 use. Dams were given injections at a volume of 0.1 ml/100 g body weight per day. Research Biochemicals Inc. supplied all drugs used in prenatal treatment.
60 ]
.
.
.
.
.
, ..........
t 50 ,
PG
i
The tail-flick latency test a7 was used to document analgesic response. Animals were tested between 09.00 and 14.00 on each of the designated postnatal days, D10, D18 and D30. Prior to testing each pup was weighed and placed with its littermates in a holding cage. During testing, the torso of each animal was placed on a section of gauze which was previously rubbed on the maternal dam and hand-held over the apparatus. During the pilot study, pups held in this way exhibited attenuated gross body movements in a manner similar to that described previously ~7. Upon testing, each animal received 3 latency trials with a 30 s intertrial recovery interval. Prior to each trial, the rat pup's tail was adjusted to cover the aperture of the heat source. The trial was started only when the pup displayed momentary quiescence. A trial consisted of the time elapsed from the introduction of a noxious stimulus (heat produced by a beam of light focused on the distal end of the tail) to the reflexive lateral movement of the tail away from the source. Tail movement was not considered an actual flick if it occurred concurrently with limb or body movements. This criteria prevented the inclusion of data generated by early triggering of the timer due to monoclonic tics and random motion. The mean of 3 trials was taken for the tail-flick latency (TFL) of each animal. Results were analyzed using the Student's t-test.
3H-Paroxetine binding Tissue preparation. Animals were sacrificed on postnatal days 18 and 30 by decapitation and whole spinal cords were removed on ice. Spinal cords were individually homogenized in 50 v/w ice-cold binding buffer (50 mM Tris-HCl, 120 mM NaCl, 5 mM KCI and 1% ascorbic acid in dH20, pH 7.4) with 10 strokes of a loosely fitting glass homogenizer and teflon pestle. The homogenate was centrifuged for 10 rain at 30 000 x g and the pellet resuspended to a final protein concentration of approximately 0.1 mg/ml. A 100/4 sample of the solution was retained for later protein determination using Lowry analysis and the remainder of the tissue frozen at -70 °C until assay. Incubations for 3H-paroxetine binding were performed in triplicate for each homogenate using 6 concentrations (in nM, 0.05, 0.1, 0.5, 1.0, 1.7, 2.0) of [3H]-paroxetine (New England Nuclear: specific activity = 20.5 Ci/mmole). Each set of triplicates contained 500 ktl of one radioligand concentration and 100 ktl of either 50 mM Tris buffer or an equal volume of 10 ~M fluoxetine (Eli-Lily) to define nonspecific binding occurring in the presence of 3H-paroxetine, which was less than 25% at the lowest ligand concentration. The assay was started by the addition of 400 ffl of tissue homogenate and incubated for 1 h at room temperature (22°C). The incubation was terminated by rapid vacuum-filtration of the solution through a PHD model cell harvester (Cambridge Tech Inc.) using Whatman GF/C filterstrips pretreated with 0.05% polyethylenimine. Filters were washed immediately, using 3 rinse cycles of ice-cold 50 mM Tris buffer lasting 4 s each. Radioactive retention was determined by scintillation counting of dried filter strips in a Beckman counter. Bmax values were calculated using linear regression analysis through a modified commercial statistics software package (SYSTAT, Inc.). Individual plots had significances greater than P < 0.05. Scatchard graphs were generated and compared.
"/i
b Js; ~H
3 0 ~-
h
25t
(D
20
T -
i
/ ' Y / /
J
10
'
SALINE
0 /
]
5 T O J~ - - - - - t ~
5
10
.....
+
....
~
15
~ I - - - - - ~
. . . .
20 25 P O S T N A T A L DAY
-4-
-
30
-
35
Fig. 1. Effect of prenatal treatment with phenylbiguanide (PG) on tail-flick latency of rat pups at various postnatal days. Data are given as mean + S.E.M. Number of animals in each group and the exact level of significance for each time-point are given in the resuits section. *P < 0.05.
saline controls (D10, 10 + 0.7 s; n = 33; D30, 28.12 + 3.8 s; n = 18) using a Student's t-test (Fig. 1). On postnatal D18 PG- (22.8 + 2.1 s; n = 33) and saline (17.8 + 2,2 s; n = 28) treated animals were not significantly different. Prenatal treatment with M D L 72222, the 5-HT3 antagonist, produced animals with statistically significant reductions in nociceptive response times on postnatal
D10 (7.3 _+ 0.7 s; n = 38; P < 0.001) and postnatal D18 (15.6 + 1.7; n = 36; P < 0.01) as compared to D M S O controls on D10 (12.25 + 1.3 s; n = 24) and D18 (24.7 +-_2.3 s; n = 24) (Fig. 2). On postnatal D30, M D L 72222 (39.1 + 4.6 s; n = 22) animals were not significantly different from their controls (31.7 + 7.4 s; n = 11). No significant differences in nociception were observed in the offspring of any other treatment group at any time-point. The data are given in Table I.
+~
45
MDLT
~° t
--~ 35 i ,% S0 z
_~ z
RESULTS
]
0
45+
Nociceptive testing
:
/
55
:~
25*
O /
104
5 o ~
5
-~-~--
10
~. . . . . .
15
~--------~
20
25
.....
~......
30
35
P O S T N A T A L DAY
Nociceptive testing Phenylbiguanide-treated
offspring showed
significant
e l e v a t i o n s o n D 1 0 (13.3 + 1.5 s; n = 26; P < 0.05) a n d D 3 0 (48.9 + 4.9 s; n = 22; P < 0.01) as c o m p a r e d
to
Fig. 2. Effect of prenatal treatment with MDL 72222 (MDL) tail-flick latency of rat pups at various postnatal days. Data given as mean +_ S.E.M. Number of animals in each group and exact level of significance for each time-point are given in the sults section. *P < 0.01,
on are the re-
295 TABLE I
POSTNATAL DAY30
Tail-flick latencies (s) in animals treated prenatally with serotonergic agents, at 10, 18 or 30 days postnatally
100~
Data are given as the mean + S.E.M. The number in parenthesis is the number of animals tested. There were no significant differences for any of the treatment groups at any time-point. SAL = saline; DPAT = 8-hydroxy-DPAT; FEN = fenfluramine; TFMPP = trifluoromethylphenylpiperazine.
80-
PD 10
PD 18
B mox 60fmoles/mg 40~
PD 30 20
SAL DPAT FEN TFMPP
10.0 11.6 9.8 11.6
+ + + +
0.7 2.3 0.9 2.8
(22) (11) (24) (31)
17.8 18.9 25.2 19.3
+ + + +
2.2 4.2 4.0 2.3
(28) (8) (17) (20)
28.1 19.7 26.2 31.7
+ + + +
3.8 4.5 3.2 1.8
(18) (5) (13) (14)
SAL
PG
DMSO
MDL
Fig. 4. Amount of 3H-paroxetine binding (Bmax) in whole spinal cord homogenates, from 30-day-old rat pups treated prenatally with 5-HT 3 receptor-active drugs (n = 5). There were no significant differences in either group.
[3H]Paroxetine binding Postnatal D18. Homogenate binding assays in D18 spinal cord indicated significantly greater specific binding in animals prenatally treated with phenylbiguanide (126.7 + 15.6 fmoles/mg, n = 7) than animals treated with saline (61.8 + 13.3, n = 7; P < 0.0086). Animals prenatally treated with M D L 72222 showed a significant increase (P < 0.0325) in binding of 3H-paroxetine compared to animals treated with D M S O (165 + 30, n = 9 vs 57.2 + 7, n = 6). These data are shown in Fig. 3. Postnatal D30. Binding of 3H-paroxetine to spinal cord was not significantly different in animals prenatally treated with PG (74.7 + 20 fmoles/mg, n = 5 vs 66.0 + 21, n = 5) or in animals treated prenatally with M D L 72222 (48 + 6, n = 7 vs 65 + 15, n = 5). These data are shown in Fig. 4. There were no significant differences in the weights of dams or offspring.
POSTNATAL DAY18
200,
150 B mox 100 fmoles/rng 50
0 SAL
PG
DMSO
MDL
Fig. 3. Amount of 3H-paroxetine binding (Bmax) in whole spinal cord homogenates, from 18-day-old rat pups treated prenatally with 5-HT 3 receptor-active drugs. Data are given as mean + S.E.M. *P < 0.03, **P < 0.008. Numbers of animals are given in the results section.
DISCUSSION
In the adult central nervous system, serotonin (5-HT) has long been known to function as a major neurotransmitter, but in the fetal CNS serotonin is believed to also function as a developmental signal or a nerve growth factor 9. Our laboratory is interested in the role of 5-HT in the regulation of neuronal development. In the past we have examined the effects of 5-hydroxytryptamine on ascending serotonergic tracts and found serotonin to serve an essential function as a signal for directing neuronal growth and receptor formation in the maturing nervous system. This was demonstrated through studies utilizing both cell cultures 19'2° and animal models 16'21, which indicated the involvement of the 5-HTla receptor and a release-regulating autoreceptor 2°. The 5-HTla receptor releases a trophic substance from astrocytes, S-10022. The mechanism by which the release-regulating receptor is involved is less clear. In this study we examined the changes in both nociceptive response and spinal cord terminal density during development after prenatal exposure to several 5-HT active drugs. There were no significant changes after prenatal treatment with 5-HTla or 5-HTlb receptor agonists or the serotonin releaser, fenfluramine. This is in contrast to our findings in receptor regulation of development of ascending serotonergic pathways. The 5-HT 3 receptor drugs, however, did have significant effects. Animals with chronic prenatal exposure to 5-HT 3 receptor-active drugs showed alterations in both behavior and terminal density as indicated by tail-flick analgesia measures and 3H-paroxetine binding. Behavioral observations revealed increases in analgesia for animals treated with PG while M D L produced opposite effects by decreasing analgesia. Both drugs, however, increased the apparent number of serotonin terminals, as assessed by
296 3H-paroxetine binding. The 5-HT 3 receptor has been located in several areas of the rat brain with the greatest density of sites occurring in the entorhinal cortex, area postrema, amygdala and hippocampus s. The location of 5-HT 3 receptors in these areas can help to explain the anxiolytic, antiemetic and other therapeutic properties of 5-HT 3 receptor-active drugs. 5-HT 3 binding sites have also been located on synaptosomal m e m b r a n e s from the spinal cord 5. More specifically, these receptors are located presynaptically on primary sensory fibres in the dorsal horn 7. 5-HTla receptors appear to be located similarly3. In tests of nociception, serotonin applied intrathecally acts as an analgesic, increasing tail-flick latency 6. This effect is reversed by 5-HT 3 antagonists. H o w e v e r , 5-HTta receptor-agonists decrease tail-flick latency 15. This suggests that these 2 types of serotonin receptors actually mediate opposite effects on nociception. A n understanding of the different signal transduction mechanisms for these receptors can then perhaps be used to explain why these receptors not only have different effects on pain threshold, but also on regulation of spinal cord d e v e l o p m e n t (i.e. the 5-HT 3 receptors have an effect, while the 5-HTta receptors apparently do not). 5-HT 3 receptors function independently of the G-protein complex mechanism of both the 5-HTla and 5-HT 2 receptors and a p p e a r instead to be ligand-gated ion channels 4 which conduct sodium, potassium and, to some extent, calcium 23. This may be important in understanding our results, since ions, like G - p r o t e i n coupled receptors, may play a role in directing d e v e l o p m e n t 18. This is particularly true for calcium w-t2. A l t h o u g h the 5-HT 3
REFERENCES 1 Cohan, C.S., Haydon, P.G. and Kater, S.B., Single channel activity differs in growing and nongrowing growth cones of isolated identified neurons of helisoma, J. Neurosci. Res., 13 (1985) 285-300. 2 Costall, B,, Naylor, R.J. and Tyers, M.B., The psychopharmacology of 5-HT3 receptors, Pharmacol. Ther., 47 (1990) 181202. 3 Daval, G., Verge, D., Basbaum, A.I., Bourgoin, S. and Hamon, M., Autoradiographic evidence of serotonin~ binding sites on primary afferent fibres in the dorsal horn of the rat spinal cord, Neurosci. Lett., 83 (1987) 71-76. 4 Derkach, V., Suprenant, A. and North, R.A., 5-HT3 receptors are membrane ion channels, Nature, 339 (1989) 706-709. 5 Glaum, S.R. and Anderson, E.G., Identification of 5-HT 3 binding sites in rat spinal cord synaptosomal membranes, Eur. J. Pharmacol., 156 (1988) 287-290. 6 Glaum, S.R., Proudfit, H.K. and Anderson, E.G., 5-HT 3 receptors modulate spinal nociceptive reflexes, Brain Research, 510 (1990) 12-18. 7 Hamon, M., Gallissot, M.C., Menard, E, Gozlan, H., Bourgoin, S. and Verge, D., 5-HT3 receptors are on capsaicin-sensitive fibres in the rat spinal cord, Eur. J. Pharmacol., 164
receptor is associated predominantly with other ions in the mature brain, its ion conductance characteristics have never been examined in the immature brain or in spinal cord. It is possible that calcium is indeed involved in the effects of 5-HT 3 drugs in development, since the hybrid mouse neuroblastoma-glioma cell line, NG-108-15, derived partly from a spinal tumor, expresses 5-HT 3 receptors linked to a calcium channel 13. Moreover. in another model commonly used to study serotonin as a developmental signal, the snail helisoma, an as yet unidentified serotonin receptor has been shown to regulate a calcium channel I . We are currently expanding our findings into 2 further studies. Firstly, we are performing immunochemistry for serotonin neurons in the spinal cord, to determine whether the changes produced by the agonist and antagonist could be regionally different, since in the whole spinal cord h o m o g e n a t e they apparently had similar effects on the n u m b e r of serotonin terminals. Secondly, we are interested in determining what neurochemical changes may have occurred which resulted in the changes in tail-flick latency, since these changes could have been due to differences in the amount of serotonin innervation, or in the sensitivity of a serotonin receptor. In conclusion, this study shows that the serotonin receptors which influence the d e v e l o p m e n t of descending serotonergic fibres are distinct from those which influence the d e v e l o p m e n t of ascending fibres.
Acknowledgements. The authors wish to thank Joshua Sherman and Christine Stark for their assistance. The work described in this paper was supported by a grant from The National Institute for Child Health and Human Development, to P.M.W.-A.
(1989) 315-322. 8 Kilpatrick, G.J., Jones, B.J. and Tyers, M.B., Identification and distribution of 5-HT3 receptors in the rat brain using radioligand binding, Nature, 330 (1987) 746-748. 9 Lauder, J.M., Hormonal and humoral influences on brain development, Psychoneuroendocrinology, 8 (1981) 121-155. 10 Mattson, M.P., Murain, M. and Guthrie, P.B., Localized calcium influx orients axon formation in embryonic hippocampal pyramidal neurons, Dev. Brain Res., 52 (1990) 201-209. 11 McCobb, D.P., Best, P.M. and Beam, K.G., Development alters the expression of calcium currents in chick limb motoneurons, Neuron, 2 (1989) 1633-1643. 12 McMahon, D., Chemical messengers in development: a hypothesis, Science, 185 (1974) 1012-1021. 13 Reiser, G., Donie, F. and Binmoller, EJ., Serotonin regulates cytosolic Ca2+ activity and membrane potential in a neuronal and in a gtial cell line via 5-HT 3 and 5-HT2 receptors by different mechanisms, J. Cell Sci., 93 (1989) 545-555. 14 Richardson, B.P. and Engel, G., The pharmacology and function of 5-HT 3 receptors, Trends Neurosci., 99 (1986) 424-428. 15 Richardson, B.P., Serotonin and nociception, Ann. N. Y. Acad. Sci., 600 (1990) 511-520. 16 Shemer, A.V., Azmitia, E.C. and Whitaker-Azmitia, P.M., Dose-related effects of prenatal 5-methoxytryptamine (5-MT)
297
17
18 19 20
on development of serotonin terminal density and behavior, Dev. Brain Res., 59 (1991) 59-64. Spear, L.P., Enters, E.K., Aswad, M.A. and Louzan, M., Drug and environmentally induced manipulations of the opiate and serotonergic systems alter nociception in neonatal rat pups, Behay. Neural Biol., 44 (1985) 1-22. Whitaker-Azmitia, P.M., Role of neurotransmitter receptors in development: the basis for developmental pharmacology, Pharmacol. Rev., in press. Whitaker-Azmitia, P.M. and Azmitia, E.C., Autoregulation of fetal serotonergic neuronal development: role of high-affinity serotonin receptors, Neurosci. Lett., 67 (1986) 307-311. Whitaker-Azmitia, P.M. and Azmitia, E.C., Tissue culture of
MDMA toxicity. In S.J. Peroutka (Ed.), M D M A Ecstasy or Human Toxin, Kuweiler Press Inc., 1988, pp. 201-211. 21 Whitaker-Azmitia, P.M., Shemer, A.V., Caruso, J., Molino, L., and Azmitia, E.C., Role of high-affinity serotonin receptors in development, Ann. N. Y. Acad. Sci., 600 (1990) 315-330. 22 Whitaker-Azmitia, P.M., Murphy, R. and Azmitia, E.C., S-100 protein is released from astroglial cells by stimulation of 5-HTI~ receptors and regulates development of serotonin neurons, Brain Research, 528 (1990) 155-159. 23 Yakel, J.L., Shao, X.M. and Jackson, M.B., The selectivity of the channel coupled to the 5-HT3 receptor, Brain Research, 533 (1990) 46-52.
293
BRES 17407
5-HT 3
receptor-active drugs alter development of spinal serotonergic innervation lack of effect of other serotonergic agents James Bell III, Xini Zhang and Patricia M. Whitaker-Azmitia
Department of Psychiatry and Behavioral Sciences, State University of New York, Stony Brook, NY 11794 (U.S,A.) (Accepted 17 September 1991)
Key words: Serotonin; Neurotransmitter and development; Serotonin3 receptor; Spinal cord; Nociception; Ontogeny
Our work has focused on identifying the type of serotonin receptor through which serotonin acts as a developmental signal in the central nervous system. Previously, we have found that the regulation of development of ascending serotonergic neurons is through the balance of two serotonin receptors. One, the 5-HTxa receptor, releases a growth factor from astroglial cells. The other receptor is related to a releaseregulating autoreceptor and can be stimulated indirectly by serotonin releasers such as fenfluramine. In the present study, we examined the receptors which regulate development of the descending neurons by treating pregnant rats with selective serotonergic drugs, from gestation day 12 until birth. Pups were subsequently tested for alterations in development by nociceptive testing (tail-flick latency) and by determining the binding of 3H-paroxetine, an indicator of serotonin terminal density, in spinal cord. Our results show that agents stimulating the 5-HTla receptor (8-OH-DPAT) or the 5-HTlb receptor (TFMPP) or substances which release serotonin (fenfluramine) had no effect on the development of spinal serotonergic pathways. However, agents acting on the 5-HT3 receptor did - the agonist phenylbiguanide (PG) increased latency on tail-flick testing (postnatal days 10 and 30), while the antagonist, MDL 72222, decreased latency (postnatal days 10 and 18). Interestingly, both the agonist and the antagonist significantlyincreased 3H-paroxetine binding on postnatal day 18. Our results are discussed in terms of a possible mechanism by which 5-HT3 receptors may influence development. INTRODUCTION
terminal density, the binding of 3H-paroxetine, a 5-HT uptake inhibitor. O u r findings show that the development of this system can be altered by prenatal treatment
Serotonin has been established to play a role in regulating the development of the central nervous system, prior to assuming its role as a neurotransmitter in the mature brain 9'21. The function of serotonin is assumed
with agents acting on the 5-HT 3 receptor.
to take place through the action of specific serotonin receptors, some of which may be the same or similar to those found in the mature brain.
Subjects
In previous studies, we have found that the developm e n t of ascending serotonergic pathways is regulated through the balance of two high-affinity serotonin receptors 19, one of which is related to the 5-HTla receptor and releases growth factors 22, whilst the other is related to a receptor which regulates the release of serotonin 2°. These results were supported by a variety of investigations, including tissue culture and whole animal studies involving both neurochemical and neurobehavioral measures 16' 21
In the current project, we are for the first time examining the type of serotonin receptors which regulate the development of descending serotonergic pathways. The functional integrity of this system was investigated using a behavioral test (tail-flick) and a measure of serotonin
MATERIALS AND METHODS Primiparous Sprague-Dawley rats (Taconic Farms) were maintained in individual cages on a 12-h light/dark cycle with free access to food and water, within a humidity- and temperature-controlled colony room. Maternal dams O-F/treatment group) were given the following treatments: a 5-HT3 agonist, phenylbiguanide (PG) (5 mg/kg), a 5-HTIa agonist, 8-OH-DPAT (1 mg/kg), a 5-HTlb agonist, trifluoromethylphenylpiperazine (TFMPP) (1 mg/kg), the general 5-HT releaser, fenfluramine (5 mg/kg), a 5-HT3 antagonist MDL-72222 (5 mg/kg) and dimethyl sulfoxide (DMSO) or saline as a control. All drugs were administered to the dam s.c., in the nape of the neck, daily, starting on gestation day 12 and continuing to the day of birth. Neonates from each group were sacrificed on postnatal days (D) D18 and D30 for neurochemical analysis by 3H-paroxetine binding, or assessed for nociceptive alteration using the tail-flick latency test on D10, D18 and D30. Dams were weighed throughout pregnancy. Pups were weighed on D1, D5, D10, D18 and D30.
Drug preparation MDL-72222 was dissolved in 100% DMSO, all other drugs were dissolved in physiological saline. Each drug solution was prepared in 40 ml volumes which were aliquoted into individual 1 ml samples, frozen the first day of the study and thawed 30 rain prior to
Correspondence: P.M. Whitaker-Azmitia, Department of Psychiatry and Behavioral Sciences, State University of New York, Stony Brook, NY 11794, U.S.A.
294 use. Dams were given injections at a volume of 0.1 ml/100 g body weight per day. Research Biochemicals Inc. supplied all drugs used in prenatal treatment.
60 ]
.
.
.
.
.
, ..........
t 50 ,
PG
i
The tail-flick latency test a7 was used to document analgesic response. Animals were tested between 09.00 and 14.00 on each of the designated postnatal days, D10, D18 and D30. Prior to testing each pup was weighed and placed with its littermates in a holding cage. During testing, the torso of each animal was placed on a section of gauze which was previously rubbed on the maternal dam and hand-held over the apparatus. During the pilot study, pups held in this way exhibited attenuated gross body movements in a manner similar to that described previously ~7. Upon testing, each animal received 3 latency trials with a 30 s intertrial recovery interval. Prior to each trial, the rat pup's tail was adjusted to cover the aperture of the heat source. The trial was started only when the pup displayed momentary quiescence. A trial consisted of the time elapsed from the introduction of a noxious stimulus (heat produced by a beam of light focused on the distal end of the tail) to the reflexive lateral movement of the tail away from the source. Tail movement was not considered an actual flick if it occurred concurrently with limb or body movements. This criteria prevented the inclusion of data generated by early triggering of the timer due to monoclonic tics and random motion. The mean of 3 trials was taken for the tail-flick latency (TFL) of each animal. Results were analyzed using the Student's t-test.
3H-Paroxetine binding Tissue preparation. Animals were sacrificed on postnatal days 18 and 30 by decapitation and whole spinal cords were removed on ice. Spinal cords were individually homogenized in 50 v/w ice-cold binding buffer (50 mM Tris-HCl, 120 mM NaCl, 5 mM KCI and 1% ascorbic acid in dH20, pH 7.4) with 10 strokes of a loosely fitting glass homogenizer and teflon pestle. The homogenate was centrifuged for 10 rain at 30 000 x g and the pellet resuspended to a final protein concentration of approximately 0.1 mg/ml. A 100/4 sample of the solution was retained for later protein determination using Lowry analysis and the remainder of the tissue frozen at -70 °C until assay. Incubations for 3H-paroxetine binding were performed in triplicate for each homogenate using 6 concentrations (in nM, 0.05, 0.1, 0.5, 1.0, 1.7, 2.0) of [3H]-paroxetine (New England Nuclear: specific activity = 20.5 Ci/mmole). Each set of triplicates contained 500 ktl of one radioligand concentration and 100 ktl of either 50 mM Tris buffer or an equal volume of 10 ~M fluoxetine (Eli-Lily) to define nonspecific binding occurring in the presence of 3H-paroxetine, which was less than 25% at the lowest ligand concentration. The assay was started by the addition of 400 ffl of tissue homogenate and incubated for 1 h at room temperature (22°C). The incubation was terminated by rapid vacuum-filtration of the solution through a PHD model cell harvester (Cambridge Tech Inc.) using Whatman GF/C filterstrips pretreated with 0.05% polyethylenimine. Filters were washed immediately, using 3 rinse cycles of ice-cold 50 mM Tris buffer lasting 4 s each. Radioactive retention was determined by scintillation counting of dried filter strips in a Beckman counter. Bmax values were calculated using linear regression analysis through a modified commercial statistics software package (SYSTAT, Inc.). Individual plots had significances greater than P < 0.05. Scatchard graphs were generated and compared.
"/i
b Js; ~H
3 0 ~-
h
25t
(D
20
T -
i
/ ' Y / /
J
10
'
SALINE
0 /
]
5 T O J~ - - - - - t ~
5
10
.....
+
....
~
15
~ I - - - - - ~
. . . .
20 25 P O S T N A T A L DAY
-4-
-
30
-
35
Fig. 1. Effect of prenatal treatment with phenylbiguanide (PG) on tail-flick latency of rat pups at various postnatal days. Data are given as mean + S.E.M. Number of animals in each group and the exact level of significance for each time-point are given in the resuits section. *P < 0.05.
saline controls (D10, 10 + 0.7 s; n = 33; D30, 28.12 + 3.8 s; n = 18) using a Student's t-test (Fig. 1). On postnatal D18 PG- (22.8 + 2.1 s; n = 33) and saline (17.8 + 2,2 s; n = 28) treated animals were not significantly different. Prenatal treatment with M D L 72222, the 5-HT3 antagonist, produced animals with statistically significant reductions in nociceptive response times on postnatal
D10 (7.3 _+ 0.7 s; n = 38; P < 0.001) and postnatal D18 (15.6 + 1.7; n = 36; P < 0.01) as compared to D M S O controls on D10 (12.25 + 1.3 s; n = 24) and D18 (24.7 +-_2.3 s; n = 24) (Fig. 2). On postnatal D30, M D L 72222 (39.1 + 4.6 s; n = 22) animals were not significantly different from their controls (31.7 + 7.4 s; n = 11). No significant differences in nociception were observed in the offspring of any other treatment group at any time-point. The data are given in Table I.
+~
45
MDLT
~° t
--~ 35 i ,% S0 z
_~ z
RESULTS
]
0
45+
Nociceptive testing
:
/
55
:~
25*
O /
104
5 o ~
5
-~-~--
10
~. . . . . .
15
~--------~
20
25
.....
~......
30
35
P O S T N A T A L DAY
Nociceptive testing Phenylbiguanide-treated
offspring showed
significant
e l e v a t i o n s o n D 1 0 (13.3 + 1.5 s; n = 26; P < 0.05) a n d D 3 0 (48.9 + 4.9 s; n = 22; P < 0.01) as c o m p a r e d
to
Fig. 2. Effect of prenatal treatment with MDL 72222 (MDL) tail-flick latency of rat pups at various postnatal days. Data given as mean +_ S.E.M. Number of animals in each group and exact level of significance for each time-point are given in the sults section. *P < 0.01,
on are the re-
295 TABLE I
POSTNATAL DAY30
Tail-flick latencies (s) in animals treated prenatally with serotonergic agents, at 10, 18 or 30 days postnatally
100~
Data are given as the mean + S.E.M. The number in parenthesis is the number of animals tested. There were no significant differences for any of the treatment groups at any time-point. SAL = saline; DPAT = 8-hydroxy-DPAT; FEN = fenfluramine; TFMPP = trifluoromethylphenylpiperazine.
80-
PD 10
PD 18
B mox 60fmoles/mg 40~
PD 30 20
SAL DPAT FEN TFMPP
10.0 11.6 9.8 11.6
+ + + +
0.7 2.3 0.9 2.8
(22) (11) (24) (31)
17.8 18.9 25.2 19.3
+ + + +
2.2 4.2 4.0 2.3
(28) (8) (17) (20)
28.1 19.7 26.2 31.7
+ + + +
3.8 4.5 3.2 1.8
(18) (5) (13) (14)
SAL
PG
DMSO
MDL
Fig. 4. Amount of 3H-paroxetine binding (Bmax) in whole spinal cord homogenates, from 30-day-old rat pups treated prenatally with 5-HT 3 receptor-active drugs (n = 5). There were no significant differences in either group.
[3H]Paroxetine binding Postnatal D18. Homogenate binding assays in D18 spinal cord indicated significantly greater specific binding in animals prenatally treated with phenylbiguanide (126.7 + 15.6 fmoles/mg, n = 7) than animals treated with saline (61.8 + 13.3, n = 7; P < 0.0086). Animals prenatally treated with M D L 72222 showed a significant increase (P < 0.0325) in binding of 3H-paroxetine compared to animals treated with D M S O (165 + 30, n = 9 vs 57.2 + 7, n = 6). These data are shown in Fig. 3. Postnatal D30. Binding of 3H-paroxetine to spinal cord was not significantly different in animals prenatally treated with PG (74.7 + 20 fmoles/mg, n = 5 vs 66.0 + 21, n = 5) or in animals treated prenatally with M D L 72222 (48 + 6, n = 7 vs 65 + 15, n = 5). These data are shown in Fig. 4. There were no significant differences in the weights of dams or offspring.
POSTNATAL DAY18
200,
150 B mox 100 fmoles/rng 50
0 SAL
PG
DMSO
MDL
Fig. 3. Amount of 3H-paroxetine binding (Bmax) in whole spinal cord homogenates, from 18-day-old rat pups treated prenatally with 5-HT 3 receptor-active drugs. Data are given as mean + S.E.M. *P < 0.03, **P < 0.008. Numbers of animals are given in the results section.
DISCUSSION
In the adult central nervous system, serotonin (5-HT) has long been known to function as a major neurotransmitter, but in the fetal CNS serotonin is believed to also function as a developmental signal or a nerve growth factor 9. Our laboratory is interested in the role of 5-HT in the regulation of neuronal development. In the past we have examined the effects of 5-hydroxytryptamine on ascending serotonergic tracts and found serotonin to serve an essential function as a signal for directing neuronal growth and receptor formation in the maturing nervous system. This was demonstrated through studies utilizing both cell cultures 19'2° and animal models 16'21, which indicated the involvement of the 5-HTla receptor and a release-regulating autoreceptor 2°. The 5-HTla receptor releases a trophic substance from astrocytes, S-10022. The mechanism by which the release-regulating receptor is involved is less clear. In this study we examined the changes in both nociceptive response and spinal cord terminal density during development after prenatal exposure to several 5-HT active drugs. There were no significant changes after prenatal treatment with 5-HTla or 5-HTlb receptor agonists or the serotonin releaser, fenfluramine. This is in contrast to our findings in receptor regulation of development of ascending serotonergic pathways. The 5-HT 3 receptor drugs, however, did have significant effects. Animals with chronic prenatal exposure to 5-HT 3 receptor-active drugs showed alterations in both behavior and terminal density as indicated by tail-flick analgesia measures and 3H-paroxetine binding. Behavioral observations revealed increases in analgesia for animals treated with PG while M D L produced opposite effects by decreasing analgesia. Both drugs, however, increased the apparent number of serotonin terminals, as assessed by
296 3H-paroxetine binding. The 5-HT 3 receptor has been located in several areas of the rat brain with the greatest density of sites occurring in the entorhinal cortex, area postrema, amygdala and hippocampus s. The location of 5-HT 3 receptors in these areas can help to explain the anxiolytic, antiemetic and other therapeutic properties of 5-HT 3 receptor-active drugs. 5-HT 3 binding sites have also been located on synaptosomal m e m b r a n e s from the spinal cord 5. More specifically, these receptors are located presynaptically on primary sensory fibres in the dorsal horn 7. 5-HTla receptors appear to be located similarly3. In tests of nociception, serotonin applied intrathecally acts as an analgesic, increasing tail-flick latency 6. This effect is reversed by 5-HT 3 antagonists. H o w e v e r , 5-HTta receptor-agonists decrease tail-flick latency 15. This suggests that these 2 types of serotonin receptors actually mediate opposite effects on nociception. A n understanding of the different signal transduction mechanisms for these receptors can then perhaps be used to explain why these receptors not only have different effects on pain threshold, but also on regulation of spinal cord d e v e l o p m e n t (i.e. the 5-HT 3 receptors have an effect, while the 5-HTta receptors apparently do not). 5-HT 3 receptors function independently of the G-protein complex mechanism of both the 5-HTla and 5-HT 2 receptors and a p p e a r instead to be ligand-gated ion channels 4 which conduct sodium, potassium and, to some extent, calcium 23. This may be important in understanding our results, since ions, like G - p r o t e i n coupled receptors, may play a role in directing d e v e l o p m e n t 18. This is particularly true for calcium w-t2. A l t h o u g h the 5-HT 3
REFERENCES 1 Cohan, C.S., Haydon, P.G. and Kater, S.B., Single channel activity differs in growing and nongrowing growth cones of isolated identified neurons of helisoma, J. Neurosci. Res., 13 (1985) 285-300. 2 Costall, B,, Naylor, R.J. and Tyers, M.B., The psychopharmacology of 5-HT3 receptors, Pharmacol. Ther., 47 (1990) 181202. 3 Daval, G., Verge, D., Basbaum, A.I., Bourgoin, S. and Hamon, M., Autoradiographic evidence of serotonin~ binding sites on primary afferent fibres in the dorsal horn of the rat spinal cord, Neurosci. Lett., 83 (1987) 71-76. 4 Derkach, V., Suprenant, A. and North, R.A., 5-HT3 receptors are membrane ion channels, Nature, 339 (1989) 706-709. 5 Glaum, S.R. and Anderson, E.G., Identification of 5-HT 3 binding sites in rat spinal cord synaptosomal membranes, Eur. J. Pharmacol., 156 (1988) 287-290. 6 Glaum, S.R., Proudfit, H.K. and Anderson, E.G., 5-HT 3 receptors modulate spinal nociceptive reflexes, Brain Research, 510 (1990) 12-18. 7 Hamon, M., Gallissot, M.C., Menard, E, Gozlan, H., Bourgoin, S. and Verge, D., 5-HT3 receptors are on capsaicin-sensitive fibres in the rat spinal cord, Eur. J. Pharmacol., 164
receptor is associated predominantly with other ions in the mature brain, its ion conductance characteristics have never been examined in the immature brain or in spinal cord. It is possible that calcium is indeed involved in the effects of 5-HT 3 drugs in development, since the hybrid mouse neuroblastoma-glioma cell line, NG-108-15, derived partly from a spinal tumor, expresses 5-HT 3 receptors linked to a calcium channel 13. Moreover. in another model commonly used to study serotonin as a developmental signal, the snail helisoma, an as yet unidentified serotonin receptor has been shown to regulate a calcium channel I . We are currently expanding our findings into 2 further studies. Firstly, we are performing immunochemistry for serotonin neurons in the spinal cord, to determine whether the changes produced by the agonist and antagonist could be regionally different, since in the whole spinal cord h o m o g e n a t e they apparently had similar effects on the n u m b e r of serotonin terminals. Secondly, we are interested in determining what neurochemical changes may have occurred which resulted in the changes in tail-flick latency, since these changes could have been due to differences in the amount of serotonin innervation, or in the sensitivity of a serotonin receptor. In conclusion, this study shows that the serotonin receptors which influence the d e v e l o p m e n t of descending serotonergic fibres are distinct from those which influence the d e v e l o p m e n t of ascending fibres.
Acknowledgements. The authors wish to thank Joshua Sherman and Christine Stark for their assistance. The work described in this paper was supported by a grant from The National Institute for Child Health and Human Development, to P.M.W.-A.
(1989) 315-322. 8 Kilpatrick, G.J., Jones, B.J. and Tyers, M.B., Identification and distribution of 5-HT3 receptors in the rat brain using radioligand binding, Nature, 330 (1987) 746-748. 9 Lauder, J.M., Hormonal and humoral influences on brain development, Psychoneuroendocrinology, 8 (1981) 121-155. 10 Mattson, M.P., Murain, M. and Guthrie, P.B., Localized calcium influx orients axon formation in embryonic hippocampal pyramidal neurons, Dev. Brain Res., 52 (1990) 201-209. 11 McCobb, D.P., Best, P.M. and Beam, K.G., Development alters the expression of calcium currents in chick limb motoneurons, Neuron, 2 (1989) 1633-1643. 12 McMahon, D., Chemical messengers in development: a hypothesis, Science, 185 (1974) 1012-1021. 13 Reiser, G., Donie, F. and Binmoller, EJ., Serotonin regulates cytosolic Ca2+ activity and membrane potential in a neuronal and in a gtial cell line via 5-HT 3 and 5-HT2 receptors by different mechanisms, J. Cell Sci., 93 (1989) 545-555. 14 Richardson, B.P. and Engel, G., The pharmacology and function of 5-HT 3 receptors, Trends Neurosci., 99 (1986) 424-428. 15 Richardson, B.P., Serotonin and nociception, Ann. N. Y. Acad. Sci., 600 (1990) 511-520. 16 Shemer, A.V., Azmitia, E.C. and Whitaker-Azmitia, P.M., Dose-related effects of prenatal 5-methoxytryptamine (5-MT)
297
17
18 19 20
on development of serotonin terminal density and behavior, Dev. Brain Res., 59 (1991) 59-64. Spear, L.P., Enters, E.K., Aswad, M.A. and Louzan, M., Drug and environmentally induced manipulations of the opiate and serotonergic systems alter nociception in neonatal rat pups, Behay. Neural Biol., 44 (1985) 1-22. Whitaker-Azmitia, P.M., Role of neurotransmitter receptors in development: the basis for developmental pharmacology, Pharmacol. Rev., in press. Whitaker-Azmitia, P.M. and Azmitia, E.C., Autoregulation of fetal serotonergic neuronal development: role of high-affinity serotonin receptors, Neurosci. Lett., 67 (1986) 307-311. Whitaker-Azmitia, P.M. and Azmitia, E.C., Tissue culture of
MDMA toxicity. In S.J. Peroutka (Ed.), M D M A Ecstasy or Human Toxin, Kuweiler Press Inc., 1988, pp. 201-211. 21 Whitaker-Azmitia, P.M., Shemer, A.V., Caruso, J., Molino, L., and Azmitia, E.C., Role of high-affinity serotonin receptors in development, Ann. N. Y. Acad. Sci., 600 (1990) 315-330. 22 Whitaker-Azmitia, P.M., Murphy, R. and Azmitia, E.C., S-100 protein is released from astroglial cells by stimulation of 5-HTI~ receptors and regulates development of serotonin neurons, Brain Research, 528 (1990) 155-159. 23 Yakel, J.L., Shao, X.M. and Jackson, M.B., The selectivity of the channel coupled to the 5-HT3 receptor, Brain Research, 533 (1990) 46-52.