- Email: [email protected]
Pre- and post-natal ontogeny of thyrotropin-releasing-hormone in the rat spinal cord: an immunocytochemical study
245
Decelopmental Brain Research, 70 (1992) 245-257 © 1992 Elsevier Science Publishers B.V. All rights reserved 0165-3806/92/$05.00
BRESD 51549
Pre- and post-natal ontogeny of thyrotropin-releasing-hormone in the rat spinal cord: an immunocytochemical study P. Poulat a, A. L e g r a n d a, N. R a j a o f e t r a a, L. Marlier a, A. Privat a a n d C. Oliver b a INSERM U-336, DPVSN, Montpellier (France)and b INSERM U-297, Marseille (France)
(Accepted 28 July 1992)
Key words: Thyrotropin-releasing hormone; Spinal cord; Ontogeny; lmmun•cytochemistry;
Rat; lntermediolateral cell column
This work aimed at providing by means of immunocytochemical techniques a detailed study of the ontogeny of thyrotropin-releasing hormone (TRH) in the spinal cord of the rat. We report the first appearance of TRH-immunoreactive fibers in the ventral funiculus of thoracic and lumbar levels at embryonic day 17. At embryonic day 18, fibers penetrated the ventral gray matter towards the central canal. At embryonic day 19, the first immunoreactive fibers were seen in the intermediolateral cell column at upper thoracic levels. This region was invaded at lower thoracic levels on the day of birth. At this time, TRH-immunoreactive axodendritic synapses were observed in the ventral horn and in the intermediolateral cell column, lmmunoreactivity increased in these regions until post-natal day 21 when the adult pattern of TRH immunoreactivity was established in the sympathetic nuclei and in the ventral horn. However, a transient TRH-like immunoreactivity was detected in lamina ill of the dorsal horn between post-natal days 14 and 30: at ultrastructural level, immunoreactive varicosities were seen to establish axodendritic synapses, in conclusion, TRH is one of the earliest peptidergic systems established in the spinal cord and it presents extensive temporal and topographical similarities with the serotonergic system with which it could be colocalized.
INTRODUCTION Thyrotropin-releasing hormone (TRH) is a neuropeptide originally isolated from ovine and porcine hypothalamus tissue 3'6. However, within the last decade, numerous studies have revealed that two.thirds of the total CNS content of TRH are contained in extra-hypothalamic regions 3~'5°. TRH neurons have been involved in vegetative, motor and endocrine functions including hypophysiotropic function, thermoregulation, sleep and feeding behavior 3s. The majority of extrahypothalamic TRH-immunoreactive (IR) neurons are found in medullary raphe nuclei 22'4s which project mainly to the spinal cord 4. Coexistence of TRH and/or substance P (SP) with serotonin (5-HT) has been reported in this region using immunochemical methods I°'2s. Lesion and transport studies confirmed that these neurons project mainly to the intermediolatcral cell column (IML) and the ventral horn of the spinal c o r d 10'14'16'25. The presence of TRH in the spinal cord was first reported in 1975 when
H6kfelt et al. I'~ used an immunofluorescence technique to demonstrate TRH-like immunostaining (TRH-LI) in the ventral horn of the rat spinal cord. Since then, the presence of TRH has been reported in the spinal cord of various mammalian species such as rat t2, cat 4f', monkey 2"~and man 2. Physiological studies suggest that TRH in the spinal cord influences motor behavior 17 and spinal sympathetic outflow 13. Microiontophoretic applications of TRH modify the excitability of sympathetic preganglionic neurons (SPN) m and of motoneurons 2°,49. Recent studies reported the first appearance of TRH in the spinal cord at embryonic day (E) 2121 or E20 in the IML26. In keeping with a recent study of 5-HT ontogeny 37, we wished to provide a detailed immunocytochemical study of pre- and post-natal ontogeny of TRH innervation in the rat spinal cord at light and electron microscopy levels with special reference to IML. The latter has recently been the object of studies on enkephalin and SP post-natal ontogeny in sympathetic nuclei 2s'4t. Taken as a whole, these studies on
Correspondence: P. Poulat, INSERM U-336, DPVSN, Case courrier 106, U.S.T.L. Place Eugene Bataillon, 34095 Montpellier Cedex 05, France. Fax: (33) 67.14.33.18.
246 Transverse sections were made at cervical, high-, mid- and lowthoracic (TtI-L I) and at lumbo-sacral levels. For each age group, 20-30 sections were performed at each level. Horizontal sections were realized at high-thoracic, mid-thoracic and low-thoracic levels when possible. Sections were then processed for immunocytochemistry. During the following treatment, dilution and washing were carried out in a Tris (0.05 M)-SMB (0.05 M) buffer (T-SMB buffer) at pH 7.5. The sections were successively incubated for a variable time in a trypsin-EDTA (Gibco) 0.35 mM (10 s for E17-19 and 15 s for E20) or 0.7 mM (10 s for P0, 30 s for P4, 2 rain for P7, 4 rain for PI4 and 5 rain for P21 to adult), and in 0.5% H20 2 solution for 8 rain. They were then treated with 1% sodium borohydride 19 for 15 rain and incubated 36 h at 4°C with the primary antiserum solution containing 1% normal goat serum (rabbit antiserum against TRH diluted at I: 10,000 in T-SMB buffer and 0.1% Triton was added). This TRH antiserum, previously characterized31, showed no cross-reactions with TRH metabolites (TRH-OH, His-Pro, diketopiperazine) and all staining was abolished with incubation with 10/tg of TRH. Positive controls with this antibody were carried on sections of the rat for including the hypothalamus3s. For electron microscopy, Triton was omitted in order to preserve a good ultrastructure. Incubations with secondary antibodies and revelation were performed in Tris-saline buffer at pH 7.4. The sections were incubated in a goat anti-rabbit antiserum I : 100 for 30 rain followed by incubation in PAP complex
the ontogeny of transmitter systems innervating spinal cord, particularly autonomic regions, should provide a morphological basis for a better understanding of neuropeptide functions in the spinal cord. MATERIALS AND METHODS For this study, we used fetuses obtained from pregnant female Sprague-Dawley rats (Iffa-Credo) aged from embryonic day (E) 17 (day 0 being that of mating) and newborn to adult animals. For each embryonic and early post-natal stage, at least 4 animals were used. Different stages of development studied were: EIT, EIg, E20, E21 = IN}(post-natal day 0: day of birth), i:4, P7, P14, P21, 1:30 and adult. After laparotomy, embryos were sacrificed and perfused transcardialy with 5% glutaraldehyde in a s ~ u m cacodylate 0.05 M and sodium metabisulfite (SMB) I% buffer at pH 7.4. Post-natal animals were anesthetized and peffused transcardiacly with a prewash of the same buffer followed by 100 ml/100 g b.wt. of the same fixative solution of 5% glutaraldehyde. Spinal cords were dissected out and immersed in the same fixative at least overnight. 50 p.m (El7 to P14) or 30 p,m (P21 to adult) transverse and horizontal sections were performed at different levels of the spinal cord with a vibratome.
El7
EI8
P4
EI9
P7
E20
PO
P14
Fig. I, Camera lucida drawings of TRH-like immunoreactivity in the spinal cord at different ages and, for each stage at three different levels: cervical, thoracic, lumbar. Cross-sections at different stages from El7 to P14.
247 I:100 for 30 min and finally reacted with 0.04% diaminobenzidine and 0.007% H20 2 under visual control. For examination with light microscope, sections were mounted on slides, dehydrated, cleared in xylene and coverslipped using Gurr-DePeX as mounting medium. For electron microscopy preparations, sections were immersed in
P21
1% osmium tetroxide in 0.1 M phosphate buffer for 30 min, rinsed and dehydrated in a graded series of ethanol and acetone and flat-embedded in araldite between two siliconed glass slides. They were then reembedded in araldite-filled gelatin capsules. Ultrathin sections were picked up on copper grids and contrasted with uranyi
P30
Adult
•, ' : . :
,,,!,, .,... I
,.,~ ,~
,:. ,..:..,..,,,
,., ,, ,,:.
E20
P4
PO ~
~oo
q~m~mllm~
.
Npm
am,
~'-
w
It
__
P7
~
I
I
•
~
P 14
I
II
Adult
Fig. 2. Camera lucida drawings of TRH-like immunoreactivity in the spinal cord at different ages and, for each stage at three different levels: cervical, thoracic, lumbar. Cross-sections at different stages from P21 to adult and drawings from horizontal sections at the IML level from E20 to adult.
248 here findings on at least 4 animals for each group, Variations existed from animal to animal, especially in young fetuses, but they were always quantitative, i.e. related to the density of innervation of a given level or locus. Embryonic day 17 (Fig. 1). First TRH-IR fibers are seen at thoracic and lumbar level in the ventral and ventrolateral funiculi (Fig. 3).
amtate and lead citrate and examined in a ZEISS E M 900 electron microscope at 80 kV. RESULTS
Light microscopy For the sake of clarification, the results obtained with light microscopy, are assembled in Figs. 1 and 2 (drawn from a camera lucida). As a rule, we report
.,I !
t'
,'
I~
°
;4
++,p~+ L +
+.
[+
+++
,
__+,+,~
,
+r'
' '
+,,+,+~+++
'
~++++, , I ' ~
'
~,
.
;
. ¢
t
,-
,.
t.
'
,i
~++,
.
~,
.' ~ .,~',"
+ .
,~.+ ,
,=
P+
,+
++ ++ ~++~% +L+, "+, ...... .... +*+' + +;~++ ~+* . +','m~,
'
. . . . "+++' +'
.++
"+* +.~I
'.~+~,m+~
+'+~
+,
eo+ ++.. •
+
+.
,
•
•
+
: "
,,
- "7+ ++
+~' '
+ •
+~'~ ;.
°',~PP
,,
~++
,~.
"
~+t
"
+',+'+
.
i1,
+, +~+, ' ,
_
_
FiB. 3-6. TRH-like immunoreactivily in the spinal cord from El'/to E20. Bar - 25 ++m. ',-
"+..
Fig. 3. Thin TRH-immunoreactive fibers (arrows) in the ventral funiculus of a thoracic sections at ElT.
Fig. 4. At El8, this transverse section at low-thoracic level shows a thick TRH-immunoreactive fiber (arrowhead) penetrating the ventral gray matter. Arrows indicate thin TRH-immunoreactive fibers in the ventral funiculus. Fig. 5. At EI9, a single TRH-immunoreactive fiber penetrates the IML, as seen on a transversally cut high-thoracic section. Fig. 6. At E20, this horizontal section at the level of the ]ML shows fine punctate TRH-like immunoreactivi~ (arrowheads) in the ]ML directed towards the CC. (In the left upper corner, the large arrowhead indicates the rostro
249
Fig. 7-12. Light and electron microscopic pictures of TRH-L! at P0. Fig, 7. This transversally cut mid.thoracic section shows fine punctate TRH-Li in the IML directed medially. (The medial direction is indicated by the arrowhead in the right upper corner.) Bar -- 40 pro. Fig. 8. This electron micrograph shows a TRH.immunoreactive varicosities making an asymmetric synaptic contact with a dendrite, in the IML. Bar -. 250 nm. Fig. 9. TRH-LI in the ventral horn at mid.thoracic level. Bar = 40/~m, Fig. 10. In the ventral horn, a TRH.immunoreactive varicosity establishes an asymmetric synaptic contact with a dendrite (d), Note the peroxidase deposit concentrated in the large vesicles (arrow). Bar = 250 nm. Fig. 11. In the ventral horn, an immature looking symmetric synapse between a TRH.immunoreactive varicosity and a dendrite• Bar = 250 nm. Fig. 12. The same TRH-immunoreactive varicosity as in Fig. 10 is tilted in another plane to show a synapse with another dendrite (d'). Bar = 250 nm.
250
Embryonic day 18 (Fig. I). At cervical level, TRH-IR fibers are detected in the ventral funiculus but also in the gray matter, ventrolateral to the ependyma. At
, ~.
'17.
•
i,'
~ ~•
~!~7o~
. . . . •.
• ~ ~,~
~
~ ~'i~~
*~ , I :~'?~¸~ ~~ !~i~i '~'~~¸~~,~ .., •
-~. °
. ,~i~~ii~! ~ , ~ i ~ ~
thoracic level, some fibers invade the ventral gray matter laterally to the central canal (CC) (Fig. 4). Moreover, immunoreactive fibers appear in the lateral
251 funiculus at high- and mid-thoracic levels only. At lumbar level, TRH-IR fibers are directed towards the CC but do not invade the ventral gray matter. Embryonic day 19 (Fig. 1). At this stage, the cervical level is invaded by thin TRH-IR fibers in the ventromedial part of the ventral horn and some other fibers grow from the ventrolateral funiculus. At high-tboracic level, immunoreactive fibers arising from the lateral funiculus invade the IML (Fig. 5) or the nucleus intermediolateralis pars principalis (ILp) and grow towards the CC while the ILp is not invaded at other thoracic levels. The lumbar level shows TRH-LI in the ventral, ventrolateral and lateral funiculi and some ventral fibers enter the gray matter, reaching the CC. Fibers in the white matter are thick and strongly immunoreactire whereas those detected in the gray matter are thin and weakly immunoreactive. Embryonic day 20 (Figs. 1 and 2). At cervical level, TRH-IR fibers arising from the ventrolateral funiculus invade the ventrolateral part of the ventral horn. At thoracic level, T R H - H is increased in the ventral horn at the border between the white and the gray matter. This immunoreactivity appears as very fine dots whereas longitudinally cut fibers invade the remaining ventral horn. At high-and mid-thoracic levels, the ILp and the dorsal commissural nucleus (DCN), above the CC, display some TRH-LI. The DCN is invaded first by IR fibers of ventral origin. Fibers are also seen in the nucleus intercalatus (IC) (Fig. 6) between the IML and the CC, oriented towards the DCN as seen on horizontal sections. At lumbar level, longitudinally cut immunoreactive fibers are present in the ventral horn and converge towards the CC. Post.natal day 0 (day of birth) (Figs. 1 and 2). At cervical level, TRH-IR fibers invade the lateral part of the ventral horn while immunoreactivity is increased in the ventromedial part. At thoracic level, TRH-LI is
increased in the ventral horn (Fig. 9) and in the sympathetic nuclei (Fig. 7). TRH-LI observed in the DCN decreases from high-tboracic to low-tboracic level and on horizontal sections. This innervation appears discontinuous. At lumbar level, TRH-LI appears as fine dots localized in the ventromedial part of the gray matter but some thick, longitudinally cut fibers extended from the commissure to the CC. A few longitudinally cut fibers are seen above the CC or at the base of the dorsal horn. There are still numerous ascending fibers in the ventral funiculus. Post-natal day 4 (Figs. 1 and 2). At all levels, TRH-LI is increased in the ventral horn and particularly at lumbar level, the ventrolateral part of which is now innervated (Fig. 13). in the same way, TRH-L! is increased in the s~qnpathetic nuclei at thoracic level. On horizontal sections we could visualize the first immunoreactive connections (called ILp longitudinal connections) between each TRH-IR l l p (Fig. 14). TRH-LI in the DCN now appears continuous, forming an immunoreactive band above the CC. Post-natal day 7 (Figs. 1 and 2). At all levels, TRH-LI in the ventral horn is increased and noticeably in the ventromedial group of motoneurons at cervical and lumbar levels (Fig. 15). At these two levels, TRH-LI appears first in the lateral part of the base of the dorsal horn corresponding probably to the intermediate zone and then in the dorsal funiculus (Fig. 17). This immunoreactivity extends a little medially and sometimes fibers are seen growing towards the apex of the dorsal horn. At thoracic level, TRH-L! is increased in the sympathetic nuclei as shown on transverse and horizontal sections (Fig. 16). However, on horizontal sections, we saw that TRH-IR ILps were close to each other. Sometimes, we could see a few immunoreactive fibers leaving the lateral funiculus, penetrating the IML and growing towards the ventral horn. At high-
¢-.
Figs. 13-18. TRH-LI in the spinal cord at P4 and P7 stages.
Fie. 13. At P4, transversesectionshowingTRH-L!in the ventralhorn at lumbarlevel.Bar - 100 ~m. Fig. 14. At P4, this horizontalsection showsthe patternof TRH-LI in the symp~taI~tic regionsat high.thoraciclevel.TRH-LI is observed in the IML (single arrows), in the ILp longitudinalconnections (single arrowheads), in the nucleus intercalatus(double arrows) and in the dorsal commissuralnucleus(double arrowheads)(see text). Bar - 1 0 0 / ~ m . Fig. 15. At P7, transverse section showingTRH-L! in the ventral horn at high-lumbarlevel. Arrowheadsindicate TRH-L! in the cremaster nucleus and in its lateraland medialextensions.Arrowpointsto TRH-LI in the IML. Bar = 50/~m. Fig. 16. At PT, an horizontalsectionshowsthe pattern of TRH-LI in the sympatheticregionsat mid.thoraciclevel.Bar = 50/~m. Fig. 17. At P7, TRH-immunoreactivefibers in the base of the dorsal horn. Note the single TRH.immunoreactivefiber in the dorsolateral funiculus(arrow).Bar = 50/tm. Fig. 18. At PT, a TRH-immunoreactivefiber in dorsal colum, (arrow). Immunoreactivefibers above the CC (arrowheads)seem to be oriented towardsthis region.Bar = 50/~m.
252 lumbar level, these fibers innervated the putative eremaster nucleus {Fig. 15). Post.natal day 14 (Figs. I and 2). At all levels, TRH-LI increased strongly in the ventral horn and on transverse sections some bundles of TRH-LI were seen extending in the ventral or lateral white matter (Fig. 19). At this stage, a fine punctate immunostaining appears in the lateral part of the inner part of the
lamina II (lli) of the dorsal horn at cervical and lumbar levels, while this staining extends more medially at thoracic level (Fig. 19). Few fibers were seen in the intermediate zone and some others invaded the dorsal horn. At thoracic level, TRH-LI was strongly increased in the sympathetic nuclei. On horizontal sections, the dense immunoreactivity observed in the ILp, in the ILp longitudinal connections, in the intercalatus nucleus
19
+
~;I,,
~,i J~' ~'~
~'
Figs, 10-21. Light and electron microphotographs of TRH-L! at PI4. Fig. 19. Cross section of the spinal cord at low thoracic level showing TRH-LI in the lamina ill of the dorsal horn, in the IML, in the IC and in the DCN. Bar ~ 50/~m. Fig. 20. in the dorsal horn, a TRH-immunoreactive varicosity establishes an asymmetric synaptic contact with a dendrite (probably a future dendritic spine). Bar = 200 nm. Fig. 21. in the IML. two TRH-immunoreactive varicosities are appos(~donto a dendrite. One of them makes a synaptic contact. Bar = 200 nm.
253 intermediate zone, between the ventral horn and the base of the dorsal horn, composed of fine, longitudinally cut fibers. Post-natal day 21 (Fig. 2). At this stage, a few qualitative and quantitative changes were observed. At all levels, the staining appeared reduced in lamina lli of
and in the dorsal commissural nucleus sketches and the ladder-like pattern observed in the adult. Only the intercalatus nucleus pars ependymalis, located laterally to the CC, does not show TRH-LI. At lumbar level, TRH-IR fibers are now also detected above the CC. We also observed at this level a delicate staining in the
23
i
12S I
,,
. ,
, ) ,1",
t ij
',~,, ,
, ,,
• .~ii,'. ~ , ~,,~,~:,-......,;...~ .... ~ ,,~
'
'
*
~'J~..
,
~,
,
'~t¢. ,%
,
.
, ,
•
~,
~,
,
~i '~ , ,~':~ ' ~,, ,,:~:~,'~, ",,~
, ~ , : ~ .~. ~ .,~.
%., ,.
,
-,~,~
.~
Figs. 22-25, TRH-L[ pattern in the spinal cord of the adult rat. Bars = 75/zm. Fig. 22. Cross-section at mid.thoracic level showing TRH-L! in the rat spinal cord. Fig. 23. This horizontal section at mid thoracic level shows the TRH.immunoreactive pattern in the sympathetic nuclei. In the right upper corner the large arrowhead indicates the rostral direction and the small a'rrow head indicates the medial direction. Fig. 24. Lack of TRH immunoreactivity in the dorsal horn. Fig. 25. TRH-L] in the ventral horn at the level of the cremaster nucleus. Note that density of immunoreactive fibers is higher in this nucleus than in the remaining ventral horn.
254
the dorsal horn. Some fibers were seen leaving the dorsolateral funiculus to innervate lamina V at cervical and lumbar levels. At these levels, some TRH-IR fibers left the lateral funiculus to reach the ventral horn and, at thoraco-lumbar level, we saw TRH-IR fibers from both the lateral and the ventrolateral funiculi projecting to the cremaster nucleus. At thoracic level, TRH-LI was increased in the sympathetic nuclei while the innervation pattern was not modified. Post-natal day 30 (Fig. 2). We saw few changes compared with the previous stage and, even so, they were only quantitative. Indeed, the staining in the dorsal horn still appeared reduced while in the sympathetic nuclei it appeared increased and unchanged in the ventral horn. Adult (Figs. 2 and 22). The final pattern of TRH-LI in the spinal cord of the adult rat is as follows at all levels: (i) no immunoreactivity in lamina lli of the dorsal horn but occasional longitudinally cut immunereactive fibers were seen extending dorsally in this horn (Fig. 24). (ii) Some residual fibers persisted at the base of the dorsal horn on the white/gray border and in the dorsal column. (iii) TRH immunostaining was denser in the internal group of motoneurons than in the external group at cervical and lumbar levels (Fig. 25). Some bundles of TRH-L! extended in the ventral or lateral white matter. (iv) At thoracic level, sympathetic nuclei displayed dense TRH-L! including: nucleus intermediolateralis pars principalis and pars funicularis, ILp longitudinal connections, nucleus intercalatus and dorsal commissural nucleus. Moreover, the TRH-IR ILps appeared more distant than in the earlier stages and were linked by dense immunoreactive ILp longitudinal connections (Fig, 23). At high-lumbar level, in the ventral horn, the presumptive cremaster nucleus displayed a dense TRH-Li which seemed to originate partly from the lateral funiculus via the IML
strongly TRH-IR growth cones and some rare TRH-IR varicosities, containing small vesicles, which established asymmetric a,todendritic synapses (Fig. 8). At P14 in the dorsal horn, immunoreactive processes were axonal profiles which presented varicosities containing small vesicles. These varicosities could establish synaptic contact with a dendrite (Fig. 20). In the IML, we observed varicosities which contained small vesicles and large, strongly TRH-IR, vesicles. They often could establish synaptic contact with a dendrite and rarely with a perikaryon (Fig. 21). DISCUSSION The optimal detection of immunogens in tissue sections is the result of the preservation and the exposition of the immunogen together with the preservation of the structure of the tissue, in a previous study 34, the composition of the fixative was the object of an extensive trial, and the use of a 5% solution of glutaraldehyde was found to yield a good preservation of the ultrastructure of the tissue, as well as a sensitive detection of the antigen while trypsin-EDTA was found to increase significantly the sensitivity of detection. In a pilot developmental study 34, this pretreatment was omitted on prenatal specimen, as it yielded unsatisfac. tory preservation of the tissues. We could then detect the first appearance of TRH.IR fibers in the ventral funiculus at the day of birth "u. Results obtained in this laboratory disclosing a very early detection of 5-HT in the cord "~7 prompted a re-examination of TRH.L! in fetuses, with pretreatment with various concentrations of trypsin-EDTA. For each age group, including adults, a composition of various treatments was performed, and we selected the combination of concentration and
(Fig. 25).
Electron.microscopy Electron-microscopy was carried out only on the day of birth since the omission of Triton dramatically reduced the immunostaining at earlier stages. At this stage, ultrastructural studies were made in the ventral horn and in the IML at thoracic level, in addition, we chose post-natal day 14 to carry out ultrastructural studies in the dorsal horn at lumbar level and the intermediolateral cell column of thoracic level. At ~ , we observed in the ventral horn some TRHIR processes appearing as a succession of straight segments and varicosities containing small vesicles which sometimes established asymmetric axodendritic contact (Figs. 10-12). At this age in the IML, we saw
TABLE I 'Exlx,rimentalcondilions ,,,
A~
m,,
Tr~osin.EDTA concentration 0.125%-0.35 mM
Treatment time
E20
0.125%-0.35 mM
15 s
P0 P4 P7 P14 P21 P30 Adult
0.25% -03 mM 0.25% -0.7 mM 0.25% -0.7 mM 0.25% -0.7 mM 0.25% -0.7 mM
10 s 30 s 2 rain 4 rain ~ rain
1~'17
El8 El9
I0 s
255 time which yielded the highest sensitivity, compatible with the preservation of the ultrastructure (Table I). With radioimmunoassay, Lamberton et al. 2t did not detect TRH-LI in the whole spinal cord until E21. This may be due to the very low concentration of the peptide before birth, when assayed on the whole cord. Magi et al. 26 did not see any TRH-IR fibers with immunocytochemistry in the ventral horn until P0. Conversely, they showed at P14 a transient expression of TRH-LI in ceil bodies, without colchicine treatment, and in fibers of lamina lli. The latter persisted in the adult. These differences could be due to different technical procedures. However, the detection of TRHLI in the dorsal horn remains a matter of debate, since Del Rio-Garcia et al. 7 mentioned the presence in the rat of 'TRH-like' peptides, which could or could not be separated from genuine TRH according to the extraction mode used. Indeed, in another study we showed TRH-LI in laminae l l i - l l l of the dorsal horn of the adult rat using another secondary antibody 35. Are all these differences related to technical procedures or to the detection of 'TRH-like' peptide different from genuine TRH? More investigations are needed in order to answer this question. We can compare the evaluation of TRH-LI revealed here with the maturation of motor, autonomic and sensory systems of the cord: (i) on the one hand, TRH-LI appears in the motor area before the cortico-spinal or the rubro-spinal tract which are established after birth44'47 Supraspinal axons are present in embryonic and neonatal spinal cord u some of which may be 'functional', since 5-HT and TRH axodendritic synapses were seen respectively at El8 and P0. On the other hand, motoneurons are post-mitotic at El2 3° and display choline acetyltransferase-Ll (ChAT-U) at El4 33. What then can be the significance of early TRH innervation? We can suppose that TRH has atrophic role as suggested previously. Indeed, TRH applied to ventral spinal cord cell cultures increased ChAT activity 4t and in vivo, TRH reduced natural motoneuron death 4x. In addition, motoneurons showed their adult ChAT-L! pattern at P14-21, like TRH-Li. Phelps et al. s2 described motoneuron bundles of ChAT-IR dendrites in the white matter of ventral funiculi which closely correspond to the extension of TRH-IR processes in the white matter. Thus, it seems that TRH-LI is closely associated with the motoneurons and their dendrites. (ii) In the sympathetic area, developmental studies have shown that preganglionic neurons have reached their adult locations at E15-19 3° and first TRH-LI was observed at El9. These SPN expressed ChAT-LI at P l - 2 32 but no specific changes were observed with
time. In another study, Schramm et al. 42 reported on SPN that a new longitudinal dendritic arbor is established in addition to a preexisting one medially oriented at P22. This is in agreement with Forehand's work 9 on morphology of SPN in neonates which described a transverse dendritic arbor. However, we did not see any qualitative changes in TRH-LI since the ladder-like pattern was established at P14. Many transmitters were found in this region and the supraspinal sources are involved in the development of SPN ~. It is then impossible to determine the role of TRH in this phenomenon if it exists. (iii) In the dorsal horn, TRH-LI appeared at P14 concomitantly with a great increase of TRH binding sites in the etitire spinal cord 36. These high levels of TRH binding sites also correspond with the period of TRH-L! increase in the ventral horn and in the sympathetic nuclei. In the adult, TRH-L! disappeared in the dorsal horn and this corresponds to a three-fold decrease in the density of spinal TRH binding sites 3~. Worth noting is that sensory afferents, if they are present at birth, are not functional until the second post-natal week s when TRH-LI appears in the dorsal horn. Thus, it is possible that the transient presence of TRH-LI in the dorsal horn corresponds to a critical period for maturation of the physiological functions of the dorsal horn in which TRH could play a role. Alternatively, we cannot exclude, in the absence of optimal HPLC identification of the peptide, that the immunogen present in the dorsal horn is a TRH related peptide (see discussion in Poulat et al.'~5). TRH-LI and substance P-L! have been detected in some spinal projecting serotonergic neurons of raphe nuclei t6. Moreover, we have evidenced with double immunocytochemical labeling the coexistence of TRHLI and 5-HT-LI in caudal raphe neurons m. Our results can be compared with those of Bregman et al. 5 and Rajaofetra et al. 37 on ontogeny of 5-HT-LI in the spinal cord. It is interesting to note that TRH-LI first appemed in the same regions as 5-HT-LI, respectively the ventral horn and the intermediolateral cell column, but 2 or 3 days later. Moreover, we observed a rt~trocaudal gradient in the appearance of TRH*LI in the IML at a different thoracic level as described for 5-HT-LI. In the ventral horn, TRH-LI, like 5-HT-Li, later became more dense in the ventromedial group of motoneurons than in the ventrolateral group. In addition, the adult pattern of both transmitters, present since P21, exhibits striking similarities in the ventral horn and in the sympathetic nuclei. This delayed detection of TRH-LI compared with 5-HT-L! in the spinal cord could reflect a physiological phenomenon such as a differed maturation of TRH but could also be due to
256 a different sensitivity of both antisera. Taken together, these data indicate that the innervation of 5-HT-LI a,d TRH-LI coincide closely in the sympathetic nuclei and in the ventral horn, thus bringing further support to 5-HT/TRH coexistence in this region. Conversely, TRH-L! and SP-L! or Enk-Ll present some differences in their development in the sympathetic nuclei27,2s,39,4o,43. In the ventral horn, the cremaster nucleus displays a strong 5-HT 12, SP 29 and TRH-LI (our results). In our preparations, we saw TRH fibers innervating cremaster nucleus coming from the lateral funiculus through the IML at P7 and this innervation increased until adulthood. The adult rat displayed a dense TRH-L! in the cremaster nucleus originating partly in the IML. These data strongly suggest the coexistence of 5-HT/TRH and/or SP in fibers of this nucleus. However, in view of the different ontogeny of 5-HT-L! and SP-L! in this nucleus, Newton et al. 29 suggested a virtual absence of 5-HT/SP coexistence in this nucleus. In conclusion, TRH-L! appeared at El7 in the ventral funiculi and invaded the ventral horn at El8. At high-thoracic level, IML was invaded at El9. At P0, we showed TRH-IR synapses on dendrites in the ventral ham a , d in the IML. Later TRH-L! increased progreuively to reach adult pattern at P14-21. In addition, we observed a transient TRH-Li in the lamina Ill of the dorsal horn between PI4 and P30. The early detection of TRH in the ventral horn and in the IML could suggest that TRH has a trophic role in these regions. Moreover, the sequence of TRH-L! development corresponds strikingly to that of 5-HT-LI in the IML and in the ventral horn. This brings further evidence for TRH and 5-HT coexistence in terminals of these regions. The transient expression of TRH-L! in the dorsal horn could suggest that TRH is involved in the maturation of the physiological functions of the dorsal horn. Acknowledgements. The authors acknowledge J.R. Teilhac for art work. This study has been supported by grants from IRME and AFM, a IRME fellowship to P.P., a D. Heumann fellowship to N.R. and a FRM fellowship to L.M. REFERENCES I Backman, S.B. and Henry, J.L., Effects of substance P and thyrotropin-releasin8 hormone on sympathetic preganglionic neutones in the upper thoracic intermediolateral nucleus of the cat, Can. J. Physiol. Pkarmacol., 62 (I984) 248-251. 2 Bennett, G.W., Nathan, P.A., Wang, K.K. and Marsden, C.A., Regional distribution of immunoreactive-thyrotrophin-releasing hormone and substance P, and indolamines in human spinal cord, J. Neurochem., 46 (1986) 1718-1724. 3 Boler, J., Enzmann, F., Folkers, K., Bowers, C.Y. and SchaUy, A.V., The identity of chemical and hormonal properties of the
thyrotropin-releasing hormone and pyroglutamyi-histidyl-proline amide, Biochem. Biophys. Res. Commun., 37 (1969) 705-710. 4 Bowker, R.M., Westlund, K.N., Sullivan, M.C., Wilber, J.F. and Coulter, J.D., Descending scrotonergic, peptidergic and cholinergic pathways from the rapbe nuclei: a multiple transmitter complex, Brain Res., 288 (1983) 33-48. 5 Bregman, B.S., Development of serotonin immunoreactivity in the rat spinal cord and its plasticity after neonatal spinal cord lesion, De~: Brain Res., 34 (1987) 245-263. 6 Burgus, R., Dunn, T.F., Desiderio, D., and Guillemin, R., Structure mol~culaire du facteur hypothalamique hypophysiotrope TRF d'origine ovine: miss en evidence par spectrometric de masse de la s~quence PCA-His-Pro-NH 2, C.R. Acad. Sci. Paris D269, (1969) 1870-1873. 7 DeI-Rio-Garcia, J. and Smyth, D.G., Distribution of pyroglutamylpeptide amides related to thyrotrophin-releasing hormone in the central nervous system and periphery of the rat, J. Endocrinol., 127 (1990) 445-450. 8 Fitzgerald, M., The post-natal development of cutaneous afferent fiber input and receptive field organization in the rat dorsal horn, J. Physiol., 364 (1985) 1-18. 9 Forehand, CJ., Morphology of sympathetic preganglionic neurons in the neonatal rat spinal cord: an intracellular horseradish peroxidase study, J. Camp. Neural., 298 (1990) 334-342. 10 Gilbert, R.F.T., Emson, P.C., Hunt, S.P., Bennett, G.W., Marsden, C.A., Sandberg, B.E.B., Steinbusch, H.W.M. and Verhofstad, A.AJ., The effects of monoamines neurotoxins on peptides in the rat spinal cord, Neuroscience, 7 (1982) 69-87. II Hamill, R.W., Bloom, E.M. and Black, I.B., Effects of spinal transection on tbe development of cholinergic and adrenergic sympathetic neurons, Brain Res., 134 (1977) 269-278. 12 Harkncss, D.H. and Brownfield, M.S., Segmental distribution of thyrotropin-releasing hormone in the rat spinal cord, Brain Res. Ball., 17 (1986) 11-19. 13 Helke, CJ. and Phillips, E.T., Thyrotropin-releasing hormone receptor activation in the spinal cord increases blood pre~ure and sympathetic tone to the vasculature and the adrenals, J. Pharmacol. F,xp. That., 245 (1988) 41-46. 14 Hirsch, M.D. and Helke, C,J., Bulbospinal thyrotropin-releasing hormone projections to the intermediolateral cell column: a doubit fluorescence immunohistochemical-retrograde tracing study in the rat, Ngurosclence, 25 (1988) 625-637. 15 H6kfelt, T., Fuxe, K., Johansson,O., Jeffcoate, S. and White, N., Thyrotropin-releasing hormone (TRH)-containinB nerve terminals in certain brainstem nuclei and in the spinal cord, Neurosci. Lett., I (1975) 133-139. 16 Johansson,O., H6kfelt, T., Pernow, B., Jeffcoate, S.L., White, N., Steinbusch, H.W.M., Verhofstad, A.A.J., Emson, P.C. and Spindel, E., Immunohistochemical support for three putative transmitters in one neuron: Coexistence of 5-hydroxytryptamine, substance P- end thyrotropin-releasing hormone-like immunoreactiv. i,ty in medullary neurons projecting to the spinal cord, Neuro. science, 6 (1981) 1857-1881. 17 Johnson, J.V., Fone, K,C.F., Havler, M.E., Tulloch, I.F., Bennett, G.W. and Marsden, C.A., A comparison of the motor behaviours produced by the intrathecal administration of thyrotropin-releasin8 hormone and thyrotropin-releasing hormone (TRH) analogues in the conscious rat, Neuroscience, 29 (1989) 463-470. 18 Kachidian, P,, Poulat, P., Marlier, L, and Privat, A , lmmunohistochemical evidence for the coexistence of substance P, thyrotropin-releasing hormone, GABA, methionin-enkephalin, and leucin-enkephalin in the seroto,ergic neuroni of the caudal raphe nuclei: a dual labeling in the rat, J. Neurosci. Res., 30 (1991) 521-530. 19 Kosaka, T., Nagatsu, !., Wu, J.Y. and Hama, K., Use of high concentrations of 81utaraldehyde for immunocytochemistry of transmitter-synthesizing enzymes in the central nervous system, Neuroscience, 18 (1986) 975-990. 20 Lacey, G., Nistri, A. and Rhys-Maitland, E.R., Large enhancement of excitatory postsynaptic potentials and currents by thyrotropin-releasing hormone (TRH) in frog spinal motoneurones, Brain Res., 488 (1989) 80-88.
257 2';, Lamberton, R.P., Lechan, R.M. and Jackson, I.M.D., Ontogeny of thyrotropin-releasing hormone and histidyl proline diketopiperazine in the rat central nervous system and the pancreas, Endocn'nology, 115 (1984) 2400-2405. 22 Lechan, R.M., Molitch, M.E. and Jackson, I.M.D., Distribution of immunoreactive human growth hormone-like material and thyrotropin-releasing hormone in the rat central nelvous system: evidence for their coexistence in the same neurons, Endocrinolo/]y, 112 (1983) 877-884. 23 Lechan, R.M., Snapper, S.B., Jacobson, S. and Jackson, I.M.D., The distribution of thyrotropin-releasing hormone (TRH) in the rhesus monkey spinal cord, Peptides, 5 (1984) 185-194. 24 Leong, S.K., Shieh, J.Y. and Wong, W.C., Localizing spinal cord projecting neurons in neonatal and immature albino rats, J. Comp. Neurol., 228 (1984) 18-23. 25 Lighton, C., Marsden, C.A. and Bennett, G.W., The effects of 5,7-dihydrox~ryptamin and p-chlorophenylalanin on thyrotropinreleasing hormone in regions of the brain and spinal cord of the rat, Neuropharmacology, 23 (1984) 55-60. 26 Marti, E., Gibson, S.J., Polak, J,M., Facer, P., Springall, D.R., Van Aswegen, G,, Aitchison, M. and Koltzenburg, M., Ontogeny of peptide- and amine-containing neurons in motor, sensory, and autonomic regions of rat and human spinal cord, dorsal root ganglia, and rat skin, J. Comp. Neurol., 266 (1987) 332-359. 27 Millhorn, D.E., H6kfelt, T., Verhofstad, A.A.J. and Terenius, L., Individual cells in the raphe nuclei of the medulla oblongata in the rat that contain immunoreaetivities for both serotonin and enkephalin project to the spinal cord, Exp. Brain Res., 75 (1989) 536-542. 28 Newton, B.W., Burkhart, A.B., Romagnano, M.A.A. and Hamill, R,W., The postnatal ontogeny of substance P-immunoreactive fibers in the sympathetic preganglionic nuclei of the rat, Dee'. Brain Res., 44 (1988) 161-180. 29 Newton, B.W., Romagnano, M.A.A. and Hamill, R.W., The ontogeny of substance P-like and serotonin,like immunoreactivities in the sexually dimorphic cremaster nucleus of the rat spinal cord, Dee,.Brain Res., 47 (1989) 227-242. 30 Homes, H.O. and Des, G.D., Temporal pattern of neurogenesis in spinal cord of the rat. !. An autoradiographic study-time and sites of origin and migration and settling patterns of neuroblasts, Brain Res., 73 (1974) 121-138. 31 Oliver, C., Eskay, R,L., Ben.Jonathan, N. and Porter, J.C., Distribution and concentration of TRH in the rat brain, Endocrinology, % (1974) 540-54~. 32 Phelps, P.E,, Barber, R,P., Houser, C.R., Crawford, O.O., Salvaterra, P,M. and Vaughn, ,I.E., Postnatal development of neurons containing choline acetyltransferase in the rat spinal cord: an immunocytocbemicalstudy, J. Comp. NeuraL, 229 (1984) 347-
36l. 33 Phelps, P.E,, Barber, R.P., Brennan, L.A., Maines, V.M., Salvaterra, P.M. and Vaughn, J.E., Embryonic development of 4 different subsets of cholinergic neurons in rat cervical spinal cord, J. Comp. Neurol., 291 (1990) 9-26. 34 Poulat, P., Rajaofetra, N., Marlier, L., Oliver, C. and Privat, A., Post-natal ontogeny of TRH innervation of the rat spinal cord with special reference to intermediolateral cell column, Fur. Neurosci. Ass., 12 (1989) 49 (Abstract.)
35 Poulat, P., Sandillon, F., Marlier, L., Rajaofetra, N., Oliver, C. and Privat, A., Distribution of thyrotropin-releasing hormone in the rat spinal cord with special reference to sympathetic nuclei: a light- and electron-microscopy immunocytochemical study, J. Neurocytol., 21 (1992) 157-170. 36 Prasad, C., Postnatal development of the spinal cord thyrotropin-releasing hormone receptor in male and female rats, Neurosci. Left., 102 (1989) 253-256. 37 Rajaofetra, N., Sandillon, F., Geffard, M. and Privat, A., Preand post-natal ontogeny of serotonergic projections to the rat spinal cord, J. Neurosci. Res., 22 (1989) 305-321. 38 Reichlin, S., Neural functions of TRH, Acta Endocrinol., Suppl. 276 (1986) 21-33. 39 Romagnano, M.A., Braiman, J., Loomis, M. and Hamill, R.W., Enkephalin fibers in autonomic nuclear regions: Intraspinal vs supraspinal origin, ]. Comp. Neurol., 266 (1987) 319-331. 40 Romagnano, M.A., Newton, B.W., Pierre, A.H. and Hamill, R.W., Postnatal ontogeny of enkephalin fibers in spinal sympathetic nuclei, Dec. Brain Res., 46 (1989) 263-279. 41 Schmidt-Achert, M.K., Askanas, V. and Engel, K.W., Thyrotropin-releasing hormone enhances choline acetyltransferase and creatine kinase in cultured spinal ventral horn neurons, J. Neurochem., 43 (1984) 586-589. 42 Schramm, L.P., Stribling, J.M. and Adair, J.R., Developmental reorientation of sympathetic preganglionic neurons in the rat, Brain Res., 106 (1976) 166-171. 43 Senba, E., Shiosaka, S., Hare, Y., Inagaki, S., Takatsuki, K., Kawai, Y. and Tohyama, M., Ontogeny of the peptidergic system in the rat spinal cord: immunocytochemical analysis, J. Comp. Neurol., 208 (1982) 54-66. 44 Shieh, J.Y., Leong, S.K. and Wong, W.C., Origin of the rubro-spinal tract in neonatal developing and mature rats, J. Comp. Neurol., 214 (1983) 79-86. 45 Tsuruo, Y., 'H6kfelt, T. and Visser, T., Thyrotropin releasing hormone (TRH).immunoreactive cell groups in the rat central nervous system, Exp. Brain Res., 68 (1987) 213-217. 46 Ulfhake, B., Arvidsson, U., Cullheim, S., H6kfelt, T., Brodin, E., Verhofstad, A. and Visser, T., An ultrastructural study of 5-hydroxytryptamine-, thyrotropin-releasing hormone- and substance P.immunoreactive axonal boutons in the motor nucleus of spinal cord segments L7-S! in the adult cat, Neuroscience, 23 (1987) 917-929. 47 Valentino, K.L., Jones, E.G. and Kane S.A.. Expression of glial fibrillary acidic protein immunoreactivity during development of long fiber tracts in the rat central nervou, system. Dec. Brain Res., 3 (1983) 317-336. 48 Weill, C.L., The prevention of natural motoneuron cell death by thyrotropin-releasing hormone, $oc. Neuroscl. Abstr., 13 (1987) 641. 49 White, S.R., A comparision of the effects of serotonin, substance P, and thyrotropin-releasin8 hormone on excitability of rat spinal motoneurons in vivo, Brain Res., 335 (1985) 63-70. 50 Winokur, A. and Utiger, R.D., Thyrotropin-releasing hormone: regional distribution in the rat brain, Science, 185 (1974) 265-267.
Decelopmental Brain Research, 70 (1992) 245-257 © 1992 Elsevier Science Publishers B.V. All rights reserved 0165-3806/92/$05.00
BRESD 51549
Pre- and post-natal ontogeny of thyrotropin-releasing-hormone in the rat spinal cord: an immunocytochemical study P. Poulat a, A. L e g r a n d a, N. R a j a o f e t r a a, L. Marlier a, A. Privat a a n d C. Oliver b a INSERM U-336, DPVSN, Montpellier (France)and b INSERM U-297, Marseille (France)
(Accepted 28 July 1992)
Key words: Thyrotropin-releasing hormone; Spinal cord; Ontogeny; lmmun•cytochemistry;
Rat; lntermediolateral cell column
This work aimed at providing by means of immunocytochemical techniques a detailed study of the ontogeny of thyrotropin-releasing hormone (TRH) in the spinal cord of the rat. We report the first appearance of TRH-immunoreactive fibers in the ventral funiculus of thoracic and lumbar levels at embryonic day 17. At embryonic day 18, fibers penetrated the ventral gray matter towards the central canal. At embryonic day 19, the first immunoreactive fibers were seen in the intermediolateral cell column at upper thoracic levels. This region was invaded at lower thoracic levels on the day of birth. At this time, TRH-immunoreactive axodendritic synapses were observed in the ventral horn and in the intermediolateral cell column, lmmunoreactivity increased in these regions until post-natal day 21 when the adult pattern of TRH immunoreactivity was established in the sympathetic nuclei and in the ventral horn. However, a transient TRH-like immunoreactivity was detected in lamina ill of the dorsal horn between post-natal days 14 and 30: at ultrastructural level, immunoreactive varicosities were seen to establish axodendritic synapses, in conclusion, TRH is one of the earliest peptidergic systems established in the spinal cord and it presents extensive temporal and topographical similarities with the serotonergic system with which it could be colocalized.
INTRODUCTION Thyrotropin-releasing hormone (TRH) is a neuropeptide originally isolated from ovine and porcine hypothalamus tissue 3'6. However, within the last decade, numerous studies have revealed that two.thirds of the total CNS content of TRH are contained in extra-hypothalamic regions 3~'5°. TRH neurons have been involved in vegetative, motor and endocrine functions including hypophysiotropic function, thermoregulation, sleep and feeding behavior 3s. The majority of extrahypothalamic TRH-immunoreactive (IR) neurons are found in medullary raphe nuclei 22'4s which project mainly to the spinal cord 4. Coexistence of TRH and/or substance P (SP) with serotonin (5-HT) has been reported in this region using immunochemical methods I°'2s. Lesion and transport studies confirmed that these neurons project mainly to the intermediolatcral cell column (IML) and the ventral horn of the spinal c o r d 10'14'16'25. The presence of TRH in the spinal cord was first reported in 1975 when
H6kfelt et al. I'~ used an immunofluorescence technique to demonstrate TRH-like immunostaining (TRH-LI) in the ventral horn of the rat spinal cord. Since then, the presence of TRH has been reported in the spinal cord of various mammalian species such as rat t2, cat 4f', monkey 2"~and man 2. Physiological studies suggest that TRH in the spinal cord influences motor behavior 17 and spinal sympathetic outflow 13. Microiontophoretic applications of TRH modify the excitability of sympathetic preganglionic neurons (SPN) m and of motoneurons 2°,49. Recent studies reported the first appearance of TRH in the spinal cord at embryonic day (E) 2121 or E20 in the IML26. In keeping with a recent study of 5-HT ontogeny 37, we wished to provide a detailed immunocytochemical study of pre- and post-natal ontogeny of TRH innervation in the rat spinal cord at light and electron microscopy levels with special reference to IML. The latter has recently been the object of studies on enkephalin and SP post-natal ontogeny in sympathetic nuclei 2s'4t. Taken as a whole, these studies on
Correspondence: P. Poulat, INSERM U-336, DPVSN, Case courrier 106, U.S.T.L. Place Eugene Bataillon, 34095 Montpellier Cedex 05, France. Fax: (33) 67.14.33.18.
246 Transverse sections were made at cervical, high-, mid- and lowthoracic (TtI-L I) and at lumbo-sacral levels. For each age group, 20-30 sections were performed at each level. Horizontal sections were realized at high-thoracic, mid-thoracic and low-thoracic levels when possible. Sections were then processed for immunocytochemistry. During the following treatment, dilution and washing were carried out in a Tris (0.05 M)-SMB (0.05 M) buffer (T-SMB buffer) at pH 7.5. The sections were successively incubated for a variable time in a trypsin-EDTA (Gibco) 0.35 mM (10 s for E17-19 and 15 s for E20) or 0.7 mM (10 s for P0, 30 s for P4, 2 rain for P7, 4 rain for PI4 and 5 rain for P21 to adult), and in 0.5% H20 2 solution for 8 rain. They were then treated with 1% sodium borohydride 19 for 15 rain and incubated 36 h at 4°C with the primary antiserum solution containing 1% normal goat serum (rabbit antiserum against TRH diluted at I: 10,000 in T-SMB buffer and 0.1% Triton was added). This TRH antiserum, previously characterized31, showed no cross-reactions with TRH metabolites (TRH-OH, His-Pro, diketopiperazine) and all staining was abolished with incubation with 10/tg of TRH. Positive controls with this antibody were carried on sections of the rat for including the hypothalamus3s. For electron microscopy, Triton was omitted in order to preserve a good ultrastructure. Incubations with secondary antibodies and revelation were performed in Tris-saline buffer at pH 7.4. The sections were incubated in a goat anti-rabbit antiserum I : 100 for 30 rain followed by incubation in PAP complex
the ontogeny of transmitter systems innervating spinal cord, particularly autonomic regions, should provide a morphological basis for a better understanding of neuropeptide functions in the spinal cord. MATERIALS AND METHODS For this study, we used fetuses obtained from pregnant female Sprague-Dawley rats (Iffa-Credo) aged from embryonic day (E) 17 (day 0 being that of mating) and newborn to adult animals. For each embryonic and early post-natal stage, at least 4 animals were used. Different stages of development studied were: EIT, EIg, E20, E21 = IN}(post-natal day 0: day of birth), i:4, P7, P14, P21, 1:30 and adult. After laparotomy, embryos were sacrificed and perfused transcardialy with 5% glutaraldehyde in a s ~ u m cacodylate 0.05 M and sodium metabisulfite (SMB) I% buffer at pH 7.4. Post-natal animals were anesthetized and peffused transcardiacly with a prewash of the same buffer followed by 100 ml/100 g b.wt. of the same fixative solution of 5% glutaraldehyde. Spinal cords were dissected out and immersed in the same fixative at least overnight. 50 p.m (El7 to P14) or 30 p,m (P21 to adult) transverse and horizontal sections were performed at different levels of the spinal cord with a vibratome.
El7
EI8
P4
EI9
P7
E20
PO
P14
Fig. I, Camera lucida drawings of TRH-like immunoreactivity in the spinal cord at different ages and, for each stage at three different levels: cervical, thoracic, lumbar. Cross-sections at different stages from El7 to P14.
247 I:100 for 30 min and finally reacted with 0.04% diaminobenzidine and 0.007% H20 2 under visual control. For examination with light microscope, sections were mounted on slides, dehydrated, cleared in xylene and coverslipped using Gurr-DePeX as mounting medium. For electron microscopy preparations, sections were immersed in
P21
1% osmium tetroxide in 0.1 M phosphate buffer for 30 min, rinsed and dehydrated in a graded series of ethanol and acetone and flat-embedded in araldite between two siliconed glass slides. They were then reembedded in araldite-filled gelatin capsules. Ultrathin sections were picked up on copper grids and contrasted with uranyi
P30
Adult
•, ' : . :
,,,!,, .,... I
,.,~ ,~
,:. ,..:..,..,,,
,., ,, ,,:.
E20
P4
PO ~
~oo
q~m~mllm~
.
Npm
am,
~'-
w
It
__
P7
~
I
I
•
~
P 14
I
II
Adult
Fig. 2. Camera lucida drawings of TRH-like immunoreactivity in the spinal cord at different ages and, for each stage at three different levels: cervical, thoracic, lumbar. Cross-sections at different stages from P21 to adult and drawings from horizontal sections at the IML level from E20 to adult.
248 here findings on at least 4 animals for each group, Variations existed from animal to animal, especially in young fetuses, but they were always quantitative, i.e. related to the density of innervation of a given level or locus. Embryonic day 17 (Fig. 1). First TRH-IR fibers are seen at thoracic and lumbar level in the ventral and ventrolateral funiculi (Fig. 3).
amtate and lead citrate and examined in a ZEISS E M 900 electron microscope at 80 kV. RESULTS
Light microscopy For the sake of clarification, the results obtained with light microscopy, are assembled in Figs. 1 and 2 (drawn from a camera lucida). As a rule, we report
.,I !
t'
,'
I~
°
;4
++,p~+ L +
+.
[+
+++
,
__+,+,~
,
+r'
' '
+,,+,+~+++
'
~++++, , I ' ~
'
~,
.
;
. ¢
t
,-
,.
t.
'
,i
~++,
.
~,
.' ~ .,~',"
+ .
,~.+ ,
,=
P+
,+
++ ++ ~++~% +L+, "+, ...... .... +*+' + +;~++ ~+* . +','m~,
'
. . . . "+++' +'
.++
"+* +.~I
'.~+~,m+~
+'+~
+,
eo+ ++.. •
+
+.
,
•
•
+
: "
,,
- "7+ ++
+~' '
+ •
+~'~ ;.
°',~PP
,,
~++
,~.
"
~+t
"
+',+'+
.
i1,
+, +~+, ' ,
_
_
FiB. 3-6. TRH-like immunoreactivily in the spinal cord from El'/to E20. Bar - 25 ++m. ',-
"+..
Fig. 3. Thin TRH-immunoreactive fibers (arrows) in the ventral funiculus of a thoracic sections at ElT.
Fig. 4. At El8, this transverse section at low-thoracic level shows a thick TRH-immunoreactive fiber (arrowhead) penetrating the ventral gray matter. Arrows indicate thin TRH-immunoreactive fibers in the ventral funiculus. Fig. 5. At EI9, a single TRH-immunoreactive fiber penetrates the IML, as seen on a transversally cut high-thoracic section. Fig. 6. At E20, this horizontal section at the level of the ]ML shows fine punctate TRH-like immunoreactivi~ (arrowheads) in the ]ML directed towards the CC. (In the left upper corner, the large arrowhead indicates the rostro
249
Fig. 7-12. Light and electron microscopic pictures of TRH-L! at P0. Fig, 7. This transversally cut mid.thoracic section shows fine punctate TRH-Li in the IML directed medially. (The medial direction is indicated by the arrowhead in the right upper corner.) Bar -- 40 pro. Fig. 8. This electron micrograph shows a TRH.immunoreactive varicosities making an asymmetric synaptic contact with a dendrite, in the IML. Bar -. 250 nm. Fig. 9. TRH-LI in the ventral horn at mid.thoracic level. Bar = 40/~m, Fig. 10. In the ventral horn, a TRH.immunoreactive varicosity establishes an asymmetric synaptic contact with a dendrite (d), Note the peroxidase deposit concentrated in the large vesicles (arrow). Bar = 250 nm. Fig. 11. In the ventral horn, an immature looking symmetric synapse between a TRH.immunoreactive varicosity and a dendrite• Bar = 250 nm. Fig. 12. The same TRH-immunoreactive varicosity as in Fig. 10 is tilted in another plane to show a synapse with another dendrite (d'). Bar = 250 nm.
250
Embryonic day 18 (Fig. I). At cervical level, TRH-IR fibers are detected in the ventral funiculus but also in the gray matter, ventrolateral to the ependyma. At
, ~.
'17.
•
i,'
~ ~•
~!~7o~
. . . . •.
• ~ ~,~
~
~ ~'i~~
*~ , I :~'?~¸~ ~~ !~i~i '~'~~¸~~,~ .., •
-~. °
. ,~i~~ii~! ~ , ~ i ~ ~
thoracic level, some fibers invade the ventral gray matter laterally to the central canal (CC) (Fig. 4). Moreover, immunoreactive fibers appear in the lateral
251 funiculus at high- and mid-thoracic levels only. At lumbar level, TRH-IR fibers are directed towards the CC but do not invade the ventral gray matter. Embryonic day 19 (Fig. 1). At this stage, the cervical level is invaded by thin TRH-IR fibers in the ventromedial part of the ventral horn and some other fibers grow from the ventrolateral funiculus. At high-tboracic level, immunoreactive fibers arising from the lateral funiculus invade the IML (Fig. 5) or the nucleus intermediolateralis pars principalis (ILp) and grow towards the CC while the ILp is not invaded at other thoracic levels. The lumbar level shows TRH-LI in the ventral, ventrolateral and lateral funiculi and some ventral fibers enter the gray matter, reaching the CC. Fibers in the white matter are thick and strongly immunoreactire whereas those detected in the gray matter are thin and weakly immunoreactive. Embryonic day 20 (Figs. 1 and 2). At cervical level, TRH-IR fibers arising from the ventrolateral funiculus invade the ventrolateral part of the ventral horn. At thoracic level, T R H - H is increased in the ventral horn at the border between the white and the gray matter. This immunoreactivity appears as very fine dots whereas longitudinally cut fibers invade the remaining ventral horn. At high-and mid-thoracic levels, the ILp and the dorsal commissural nucleus (DCN), above the CC, display some TRH-LI. The DCN is invaded first by IR fibers of ventral origin. Fibers are also seen in the nucleus intercalatus (IC) (Fig. 6) between the IML and the CC, oriented towards the DCN as seen on horizontal sections. At lumbar level, longitudinally cut immunoreactive fibers are present in the ventral horn and converge towards the CC. Post.natal day 0 (day of birth) (Figs. 1 and 2). At cervical level, TRH-IR fibers invade the lateral part of the ventral horn while immunoreactivity is increased in the ventromedial part. At thoracic level, TRH-LI is
increased in the ventral horn (Fig. 9) and in the sympathetic nuclei (Fig. 7). TRH-LI observed in the DCN decreases from high-tboracic to low-tboracic level and on horizontal sections. This innervation appears discontinuous. At lumbar level, TRH-LI appears as fine dots localized in the ventromedial part of the gray matter but some thick, longitudinally cut fibers extended from the commissure to the CC. A few longitudinally cut fibers are seen above the CC or at the base of the dorsal horn. There are still numerous ascending fibers in the ventral funiculus. Post-natal day 4 (Figs. 1 and 2). At all levels, TRH-LI is increased in the ventral horn and particularly at lumbar level, the ventrolateral part of which is now innervated (Fig. 13). in the same way, TRH-L! is increased in the s~qnpathetic nuclei at thoracic level. On horizontal sections we could visualize the first immunoreactive connections (called ILp longitudinal connections) between each TRH-IR l l p (Fig. 14). TRH-LI in the DCN now appears continuous, forming an immunoreactive band above the CC. Post-natal day 7 (Figs. 1 and 2). At all levels, TRH-LI in the ventral horn is increased and noticeably in the ventromedial group of motoneurons at cervical and lumbar levels (Fig. 15). At these two levels, TRH-LI appears first in the lateral part of the base of the dorsal horn corresponding probably to the intermediate zone and then in the dorsal funiculus (Fig. 17). This immunoreactivity extends a little medially and sometimes fibers are seen growing towards the apex of the dorsal horn. At thoracic level, TRH-L! is increased in the sympathetic nuclei as shown on transverse and horizontal sections (Fig. 16). However, on horizontal sections, we saw that TRH-IR ILps were close to each other. Sometimes, we could see a few immunoreactive fibers leaving the lateral funiculus, penetrating the IML and growing towards the ventral horn. At high-
¢-.
Figs. 13-18. TRH-LI in the spinal cord at P4 and P7 stages.
Fie. 13. At P4, transversesectionshowingTRH-L!in the ventralhorn at lumbarlevel.Bar - 100 ~m. Fig. 14. At P4, this horizontalsection showsthe patternof TRH-LI in the symp~taI~tic regionsat high.thoraciclevel.TRH-LI is observed in the IML (single arrows), in the ILp longitudinalconnections (single arrowheads), in the nucleus intercalatus(double arrows) and in the dorsal commissuralnucleus(double arrowheads)(see text). Bar - 1 0 0 / ~ m . Fig. 15. At P7, transverse section showingTRH-L! in the ventral horn at high-lumbarlevel. Arrowheadsindicate TRH-L! in the cremaster nucleus and in its lateraland medialextensions.Arrowpointsto TRH-LI in the IML. Bar = 50/~m. Fig. 16. At PT, an horizontalsectionshowsthe pattern of TRH-LI in the sympatheticregionsat mid.thoraciclevel.Bar = 50/~m. Fig. 17. At P7, TRH-immunoreactivefibers in the base of the dorsal horn. Note the single TRH.immunoreactivefiber in the dorsolateral funiculus(arrow).Bar = 50/tm. Fig. 18. At PT, a TRH-immunoreactivefiber in dorsal colum, (arrow). Immunoreactivefibers above the CC (arrowheads)seem to be oriented towardsthis region.Bar = 50/~m.
252 lumbar level, these fibers innervated the putative eremaster nucleus {Fig. 15). Post.natal day 14 (Figs. I and 2). At all levels, TRH-LI increased strongly in the ventral horn and on transverse sections some bundles of TRH-LI were seen extending in the ventral or lateral white matter (Fig. 19). At this stage, a fine punctate immunostaining appears in the lateral part of the inner part of the
lamina II (lli) of the dorsal horn at cervical and lumbar levels, while this staining extends more medially at thoracic level (Fig. 19). Few fibers were seen in the intermediate zone and some others invaded the dorsal horn. At thoracic level, TRH-LI was strongly increased in the sympathetic nuclei. On horizontal sections, the dense immunoreactivity observed in the ILp, in the ILp longitudinal connections, in the intercalatus nucleus
19
+
~;I,,
~,i J~' ~'~
~'
Figs, 10-21. Light and electron microphotographs of TRH-L! at PI4. Fig. 19. Cross section of the spinal cord at low thoracic level showing TRH-LI in the lamina ill of the dorsal horn, in the IML, in the IC and in the DCN. Bar ~ 50/~m. Fig. 20. in the dorsal horn, a TRH-immunoreactive varicosity establishes an asymmetric synaptic contact with a dendrite (probably a future dendritic spine). Bar = 200 nm. Fig. 21. in the IML. two TRH-immunoreactive varicosities are appos(~donto a dendrite. One of them makes a synaptic contact. Bar = 200 nm.
253 intermediate zone, between the ventral horn and the base of the dorsal horn, composed of fine, longitudinally cut fibers. Post-natal day 21 (Fig. 2). At this stage, a few qualitative and quantitative changes were observed. At all levels, the staining appeared reduced in lamina lli of
and in the dorsal commissural nucleus sketches and the ladder-like pattern observed in the adult. Only the intercalatus nucleus pars ependymalis, located laterally to the CC, does not show TRH-LI. At lumbar level, TRH-IR fibers are now also detected above the CC. We also observed at this level a delicate staining in the
23
i
12S I
,,
. ,
, ) ,1",
t ij
',~,, ,
, ,,
• .~ii,'. ~ , ~,,~,~:,-......,;...~ .... ~ ,,~
'
'
*
~'J~..
,
~,
,
'~t¢. ,%
,
.
, ,
•
~,
~,
,
~i '~ , ,~':~ ' ~,, ,,:~:~,'~, ",,~
, ~ , : ~ .~. ~ .,~.
%., ,.
,
-,~,~
.~
Figs. 22-25, TRH-L[ pattern in the spinal cord of the adult rat. Bars = 75/zm. Fig. 22. Cross-section at mid.thoracic level showing TRH-L! in the rat spinal cord. Fig. 23. This horizontal section at mid thoracic level shows the TRH.immunoreactive pattern in the sympathetic nuclei. In the right upper corner the large arrowhead indicates the rostral direction and the small a'rrow head indicates the medial direction. Fig. 24. Lack of TRH immunoreactivity in the dorsal horn. Fig. 25. TRH-L] in the ventral horn at the level of the cremaster nucleus. Note that density of immunoreactive fibers is higher in this nucleus than in the remaining ventral horn.
254
the dorsal horn. Some fibers were seen leaving the dorsolateral funiculus to innervate lamina V at cervical and lumbar levels. At these levels, some TRH-IR fibers left the lateral funiculus to reach the ventral horn and, at thoraco-lumbar level, we saw TRH-IR fibers from both the lateral and the ventrolateral funiculi projecting to the cremaster nucleus. At thoracic level, TRH-LI was increased in the sympathetic nuclei while the innervation pattern was not modified. Post-natal day 30 (Fig. 2). We saw few changes compared with the previous stage and, even so, they were only quantitative. Indeed, the staining in the dorsal horn still appeared reduced while in the sympathetic nuclei it appeared increased and unchanged in the ventral horn. Adult (Figs. 2 and 22). The final pattern of TRH-LI in the spinal cord of the adult rat is as follows at all levels: (i) no immunoreactivity in lamina lli of the dorsal horn but occasional longitudinally cut immunereactive fibers were seen extending dorsally in this horn (Fig. 24). (ii) Some residual fibers persisted at the base of the dorsal horn on the white/gray border and in the dorsal column. (iii) TRH immunostaining was denser in the internal group of motoneurons than in the external group at cervical and lumbar levels (Fig. 25). Some bundles of TRH-L! extended in the ventral or lateral white matter. (iv) At thoracic level, sympathetic nuclei displayed dense TRH-L! including: nucleus intermediolateralis pars principalis and pars funicularis, ILp longitudinal connections, nucleus intercalatus and dorsal commissural nucleus. Moreover, the TRH-IR ILps appeared more distant than in the earlier stages and were linked by dense immunoreactive ILp longitudinal connections (Fig, 23). At high-lumbar level, in the ventral horn, the presumptive cremaster nucleus displayed a dense TRH-Li which seemed to originate partly from the lateral funiculus via the IML
strongly TRH-IR growth cones and some rare TRH-IR varicosities, containing small vesicles, which established asymmetric a,todendritic synapses (Fig. 8). At P14 in the dorsal horn, immunoreactive processes were axonal profiles which presented varicosities containing small vesicles. These varicosities could establish synaptic contact with a dendrite (Fig. 20). In the IML, we observed varicosities which contained small vesicles and large, strongly TRH-IR, vesicles. They often could establish synaptic contact with a dendrite and rarely with a perikaryon (Fig. 21). DISCUSSION The optimal detection of immunogens in tissue sections is the result of the preservation and the exposition of the immunogen together with the preservation of the structure of the tissue, in a previous study 34, the composition of the fixative was the object of an extensive trial, and the use of a 5% solution of glutaraldehyde was found to yield a good preservation of the ultrastructure of the tissue, as well as a sensitive detection of the antigen while trypsin-EDTA was found to increase significantly the sensitivity of detection. In a pilot developmental study 34, this pretreatment was omitted on prenatal specimen, as it yielded unsatisfac. tory preservation of the tissues. We could then detect the first appearance of TRH.IR fibers in the ventral funiculus at the day of birth "u. Results obtained in this laboratory disclosing a very early detection of 5-HT in the cord "~7 prompted a re-examination of TRH.L! in fetuses, with pretreatment with various concentrations of trypsin-EDTA. For each age group, including adults, a composition of various treatments was performed, and we selected the combination of concentration and
(Fig. 25).
Electron.microscopy Electron-microscopy was carried out only on the day of birth since the omission of Triton dramatically reduced the immunostaining at earlier stages. At this stage, ultrastructural studies were made in the ventral horn and in the IML at thoracic level, in addition, we chose post-natal day 14 to carry out ultrastructural studies in the dorsal horn at lumbar level and the intermediolateral cell column of thoracic level. At ~ , we observed in the ventral horn some TRHIR processes appearing as a succession of straight segments and varicosities containing small vesicles which sometimes established asymmetric axodendritic contact (Figs. 10-12). At this age in the IML, we saw
TABLE I 'Exlx,rimentalcondilions ,,,
A~
m,,
Tr~osin.EDTA concentration 0.125%-0.35 mM
Treatment time
E20
0.125%-0.35 mM
15 s
P0 P4 P7 P14 P21 P30 Adult
0.25% -03 mM 0.25% -0.7 mM 0.25% -0.7 mM 0.25% -0.7 mM 0.25% -0.7 mM
10 s 30 s 2 rain 4 rain ~ rain
1~'17
El8 El9
I0 s
255 time which yielded the highest sensitivity, compatible with the preservation of the ultrastructure (Table I). With radioimmunoassay, Lamberton et al. 2t did not detect TRH-LI in the whole spinal cord until E21. This may be due to the very low concentration of the peptide before birth, when assayed on the whole cord. Magi et al. 26 did not see any TRH-IR fibers with immunocytochemistry in the ventral horn until P0. Conversely, they showed at P14 a transient expression of TRH-LI in ceil bodies, without colchicine treatment, and in fibers of lamina lli. The latter persisted in the adult. These differences could be due to different technical procedures. However, the detection of TRHLI in the dorsal horn remains a matter of debate, since Del Rio-Garcia et al. 7 mentioned the presence in the rat of 'TRH-like' peptides, which could or could not be separated from genuine TRH according to the extraction mode used. Indeed, in another study we showed TRH-LI in laminae l l i - l l l of the dorsal horn of the adult rat using another secondary antibody 35. Are all these differences related to technical procedures or to the detection of 'TRH-like' peptide different from genuine TRH? More investigations are needed in order to answer this question. We can compare the evaluation of TRH-LI revealed here with the maturation of motor, autonomic and sensory systems of the cord: (i) on the one hand, TRH-LI appears in the motor area before the cortico-spinal or the rubro-spinal tract which are established after birth44'47 Supraspinal axons are present in embryonic and neonatal spinal cord u some of which may be 'functional', since 5-HT and TRH axodendritic synapses were seen respectively at El8 and P0. On the other hand, motoneurons are post-mitotic at El2 3° and display choline acetyltransferase-Ll (ChAT-U) at El4 33. What then can be the significance of early TRH innervation? We can suppose that TRH has atrophic role as suggested previously. Indeed, TRH applied to ventral spinal cord cell cultures increased ChAT activity 4t and in vivo, TRH reduced natural motoneuron death 4x. In addition, motoneurons showed their adult ChAT-L! pattern at P14-21, like TRH-Li. Phelps et al. s2 described motoneuron bundles of ChAT-IR dendrites in the white matter of ventral funiculi which closely correspond to the extension of TRH-IR processes in the white matter. Thus, it seems that TRH-LI is closely associated with the motoneurons and their dendrites. (ii) In the sympathetic area, developmental studies have shown that preganglionic neurons have reached their adult locations at E15-19 3° and first TRH-LI was observed at El9. These SPN expressed ChAT-LI at P l - 2 32 but no specific changes were observed with
time. In another study, Schramm et al. 42 reported on SPN that a new longitudinal dendritic arbor is established in addition to a preexisting one medially oriented at P22. This is in agreement with Forehand's work 9 on morphology of SPN in neonates which described a transverse dendritic arbor. However, we did not see any qualitative changes in TRH-LI since the ladder-like pattern was established at P14. Many transmitters were found in this region and the supraspinal sources are involved in the development of SPN ~. It is then impossible to determine the role of TRH in this phenomenon if it exists. (iii) In the dorsal horn, TRH-LI appeared at P14 concomitantly with a great increase of TRH binding sites in the etitire spinal cord 36. These high levels of TRH binding sites also correspond with the period of TRH-L! increase in the ventral horn and in the sympathetic nuclei. In the adult, TRH-L! disappeared in the dorsal horn and this corresponds to a three-fold decrease in the density of spinal TRH binding sites 3~. Worth noting is that sensory afferents, if they are present at birth, are not functional until the second post-natal week s when TRH-LI appears in the dorsal horn. Thus, it is possible that the transient presence of TRH-LI in the dorsal horn corresponds to a critical period for maturation of the physiological functions of the dorsal horn in which TRH could play a role. Alternatively, we cannot exclude, in the absence of optimal HPLC identification of the peptide, that the immunogen present in the dorsal horn is a TRH related peptide (see discussion in Poulat et al.'~5). TRH-LI and substance P-L! have been detected in some spinal projecting serotonergic neurons of raphe nuclei t6. Moreover, we have evidenced with double immunocytochemical labeling the coexistence of TRHLI and 5-HT-LI in caudal raphe neurons m. Our results can be compared with those of Bregman et al. 5 and Rajaofetra et al. 37 on ontogeny of 5-HT-LI in the spinal cord. It is interesting to note that TRH-LI first appemed in the same regions as 5-HT-LI, respectively the ventral horn and the intermediolateral cell column, but 2 or 3 days later. Moreover, we observed a rt~trocaudal gradient in the appearance of TRH*LI in the IML at a different thoracic level as described for 5-HT-LI. In the ventral horn, TRH-LI, like 5-HT-Li, later became more dense in the ventromedial group of motoneurons than in the ventrolateral group. In addition, the adult pattern of both transmitters, present since P21, exhibits striking similarities in the ventral horn and in the sympathetic nuclei. This delayed detection of TRH-LI compared with 5-HT-L! in the spinal cord could reflect a physiological phenomenon such as a differed maturation of TRH but could also be due to
256 a different sensitivity of both antisera. Taken together, these data indicate that the innervation of 5-HT-LI a,d TRH-LI coincide closely in the sympathetic nuclei and in the ventral horn, thus bringing further support to 5-HT/TRH coexistence in this region. Conversely, TRH-L! and SP-L! or Enk-Ll present some differences in their development in the sympathetic nuclei27,2s,39,4o,43. In the ventral horn, the cremaster nucleus displays a strong 5-HT 12, SP 29 and TRH-LI (our results). In our preparations, we saw TRH fibers innervating cremaster nucleus coming from the lateral funiculus through the IML at P7 and this innervation increased until adulthood. The adult rat displayed a dense TRH-L! in the cremaster nucleus originating partly in the IML. These data strongly suggest the coexistence of 5-HT/TRH and/or SP in fibers of this nucleus. However, in view of the different ontogeny of 5-HT-L! and SP-L! in this nucleus, Newton et al. 29 suggested a virtual absence of 5-HT/SP coexistence in this nucleus. In conclusion, TRH-L! appeared at El7 in the ventral funiculi and invaded the ventral horn at El8. At high-thoracic level, IML was invaded at El9. At P0, we showed TRH-IR synapses on dendrites in the ventral ham a , d in the IML. Later TRH-L! increased progreuively to reach adult pattern at P14-21. In addition, we observed a transient TRH-Li in the lamina Ill of the dorsal horn between PI4 and P30. The early detection of TRH in the ventral horn and in the IML could suggest that TRH has a trophic role in these regions. Moreover, the sequence of TRH-L! development corresponds strikingly to that of 5-HT-LI in the IML and in the ventral horn. This brings further evidence for TRH and 5-HT coexistence in terminals of these regions. The transient expression of TRH-L! in the dorsal horn could suggest that TRH is involved in the maturation of the physiological functions of the dorsal horn. Acknowledgements. The authors acknowledge J.R. Teilhac for art work. This study has been supported by grants from IRME and AFM, a IRME fellowship to P.P., a D. Heumann fellowship to N.R. and a FRM fellowship to L.M. REFERENCES I Backman, S.B. and Henry, J.L., Effects of substance P and thyrotropin-releasin8 hormone on sympathetic preganglionic neutones in the upper thoracic intermediolateral nucleus of the cat, Can. J. Physiol. Pkarmacol., 62 (I984) 248-251. 2 Bennett, G.W., Nathan, P.A., Wang, K.K. and Marsden, C.A., Regional distribution of immunoreactive-thyrotrophin-releasing hormone and substance P, and indolamines in human spinal cord, J. Neurochem., 46 (1986) 1718-1724. 3 Boler, J., Enzmann, F., Folkers, K., Bowers, C.Y. and SchaUy, A.V., The identity of chemical and hormonal properties of the
thyrotropin-releasing hormone and pyroglutamyi-histidyl-proline amide, Biochem. Biophys. Res. Commun., 37 (1969) 705-710. 4 Bowker, R.M., Westlund, K.N., Sullivan, M.C., Wilber, J.F. and Coulter, J.D., Descending scrotonergic, peptidergic and cholinergic pathways from the rapbe nuclei: a multiple transmitter complex, Brain Res., 288 (1983) 33-48. 5 Bregman, B.S., Development of serotonin immunoreactivity in the rat spinal cord and its plasticity after neonatal spinal cord lesion, De~: Brain Res., 34 (1987) 245-263. 6 Burgus, R., Dunn, T.F., Desiderio, D., and Guillemin, R., Structure mol~culaire du facteur hypothalamique hypophysiotrope TRF d'origine ovine: miss en evidence par spectrometric de masse de la s~quence PCA-His-Pro-NH 2, C.R. Acad. Sci. Paris D269, (1969) 1870-1873. 7 DeI-Rio-Garcia, J. and Smyth, D.G., Distribution of pyroglutamylpeptide amides related to thyrotrophin-releasing hormone in the central nervous system and periphery of the rat, J. Endocrinol., 127 (1990) 445-450. 8 Fitzgerald, M., The post-natal development of cutaneous afferent fiber input and receptive field organization in the rat dorsal horn, J. Physiol., 364 (1985) 1-18. 9 Forehand, CJ., Morphology of sympathetic preganglionic neurons in the neonatal rat spinal cord: an intracellular horseradish peroxidase study, J. Camp. Neural., 298 (1990) 334-342. 10 Gilbert, R.F.T., Emson, P.C., Hunt, S.P., Bennett, G.W., Marsden, C.A., Sandberg, B.E.B., Steinbusch, H.W.M. and Verhofstad, A.AJ., The effects of monoamines neurotoxins on peptides in the rat spinal cord, Neuroscience, 7 (1982) 69-87. II Hamill, R.W., Bloom, E.M. and Black, I.B., Effects of spinal transection on tbe development of cholinergic and adrenergic sympathetic neurons, Brain Res., 134 (1977) 269-278. 12 Harkncss, D.H. and Brownfield, M.S., Segmental distribution of thyrotropin-releasing hormone in the rat spinal cord, Brain Res. Ball., 17 (1986) 11-19. 13 Helke, CJ. and Phillips, E.T., Thyrotropin-releasing hormone receptor activation in the spinal cord increases blood pre~ure and sympathetic tone to the vasculature and the adrenals, J. Pharmacol. F,xp. That., 245 (1988) 41-46. 14 Hirsch, M.D. and Helke, C,J., Bulbospinal thyrotropin-releasing hormone projections to the intermediolateral cell column: a doubit fluorescence immunohistochemical-retrograde tracing study in the rat, Ngurosclence, 25 (1988) 625-637. 15 H6kfelt, T., Fuxe, K., Johansson,O., Jeffcoate, S. and White, N., Thyrotropin-releasing hormone (TRH)-containinB nerve terminals in certain brainstem nuclei and in the spinal cord, Neurosci. Lett., I (1975) 133-139. 16 Johansson,O., H6kfelt, T., Pernow, B., Jeffcoate, S.L., White, N., Steinbusch, H.W.M., Verhofstad, A.A.J., Emson, P.C. and Spindel, E., Immunohistochemical support for three putative transmitters in one neuron: Coexistence of 5-hydroxytryptamine, substance P- end thyrotropin-releasing hormone-like immunoreactiv. i,ty in medullary neurons projecting to the spinal cord, Neuro. science, 6 (1981) 1857-1881. 17 Johnson, J.V., Fone, K,C.F., Havler, M.E., Tulloch, I.F., Bennett, G.W. and Marsden, C.A., A comparison of the motor behaviours produced by the intrathecal administration of thyrotropin-releasin8 hormone and thyrotropin-releasing hormone (TRH) analogues in the conscious rat, Neuroscience, 29 (1989) 463-470. 18 Kachidian, P,, Poulat, P., Marlier, L, and Privat, A , lmmunohistochemical evidence for the coexistence of substance P, thyrotropin-releasing hormone, GABA, methionin-enkephalin, and leucin-enkephalin in the seroto,ergic neuroni of the caudal raphe nuclei: a dual labeling in the rat, J. Neurosci. Res., 30 (1991) 521-530. 19 Kosaka, T., Nagatsu, !., Wu, J.Y. and Hama, K., Use of high concentrations of 81utaraldehyde for immunocytochemistry of transmitter-synthesizing enzymes in the central nervous system, Neuroscience, 18 (1986) 975-990. 20 Lacey, G., Nistri, A. and Rhys-Maitland, E.R., Large enhancement of excitatory postsynaptic potentials and currents by thyrotropin-releasing hormone (TRH) in frog spinal motoneurones, Brain Res., 488 (1989) 80-88.
257 2';, Lamberton, R.P., Lechan, R.M. and Jackson, I.M.D., Ontogeny of thyrotropin-releasing hormone and histidyl proline diketopiperazine in the rat central nervous system and the pancreas, Endocn'nology, 115 (1984) 2400-2405. 22 Lechan, R.M., Molitch, M.E. and Jackson, I.M.D., Distribution of immunoreactive human growth hormone-like material and thyrotropin-releasing hormone in the rat central nelvous system: evidence for their coexistence in the same neurons, Endocrinolo/]y, 112 (1983) 877-884. 23 Lechan, R.M., Snapper, S.B., Jacobson, S. and Jackson, I.M.D., The distribution of thyrotropin-releasing hormone (TRH) in the rhesus monkey spinal cord, Peptides, 5 (1984) 185-194. 24 Leong, S.K., Shieh, J.Y. and Wong, W.C., Localizing spinal cord projecting neurons in neonatal and immature albino rats, J. Comp. Neurol., 228 (1984) 18-23. 25 Lighton, C., Marsden, C.A. and Bennett, G.W., The effects of 5,7-dihydrox~ryptamin and p-chlorophenylalanin on thyrotropinreleasing hormone in regions of the brain and spinal cord of the rat, Neuropharmacology, 23 (1984) 55-60. 26 Marti, E., Gibson, S.J., Polak, J,M., Facer, P., Springall, D.R., Van Aswegen, G,, Aitchison, M. and Koltzenburg, M., Ontogeny of peptide- and amine-containing neurons in motor, sensory, and autonomic regions of rat and human spinal cord, dorsal root ganglia, and rat skin, J. Comp. Neurol., 266 (1987) 332-359. 27 Millhorn, D.E., H6kfelt, T., Verhofstad, A.A.J. and Terenius, L., Individual cells in the raphe nuclei of the medulla oblongata in the rat that contain immunoreaetivities for both serotonin and enkephalin project to the spinal cord, Exp. Brain Res., 75 (1989) 536-542. 28 Newton, B.W., Burkhart, A.B., Romagnano, M.A.A. and Hamill, R,W., The postnatal ontogeny of substance P-immunoreactive fibers in the sympathetic preganglionic nuclei of the rat, Dee'. Brain Res., 44 (1988) 161-180. 29 Newton, B.W., Romagnano, M.A.A. and Hamill, R.W., The ontogeny of substance P-like and serotonin,like immunoreactivities in the sexually dimorphic cremaster nucleus of the rat spinal cord, Dee,.Brain Res., 47 (1989) 227-242. 30 Homes, H.O. and Des, G.D., Temporal pattern of neurogenesis in spinal cord of the rat. !. An autoradiographic study-time and sites of origin and migration and settling patterns of neuroblasts, Brain Res., 73 (1974) 121-138. 31 Oliver, C., Eskay, R,L., Ben.Jonathan, N. and Porter, J.C., Distribution and concentration of TRH in the rat brain, Endocrinology, % (1974) 540-54~. 32 Phelps, P.E,, Barber, R,P., Houser, C.R., Crawford, O.O., Salvaterra, P,M. and Vaughn, ,I.E., Postnatal development of neurons containing choline acetyltransferase in the rat spinal cord: an immunocytocbemicalstudy, J. Comp. NeuraL, 229 (1984) 347-
36l. 33 Phelps, P.E,, Barber, R.P., Brennan, L.A., Maines, V.M., Salvaterra, P.M. and Vaughn, J.E., Embryonic development of 4 different subsets of cholinergic neurons in rat cervical spinal cord, J. Comp. Neurol., 291 (1990) 9-26. 34 Poulat, P., Rajaofetra, N., Marlier, L., Oliver, C. and Privat, A., Post-natal ontogeny of TRH innervation of the rat spinal cord with special reference to intermediolateral cell column, Fur. Neurosci. Ass., 12 (1989) 49 (Abstract.)
35 Poulat, P., Sandillon, F., Marlier, L., Rajaofetra, N., Oliver, C. and Privat, A., Distribution of thyrotropin-releasing hormone in the rat spinal cord with special reference to sympathetic nuclei: a light- and electron-microscopy immunocytochemical study, J. Neurocytol., 21 (1992) 157-170. 36 Prasad, C., Postnatal development of the spinal cord thyrotropin-releasing hormone receptor in male and female rats, Neurosci. Left., 102 (1989) 253-256. 37 Rajaofetra, N., Sandillon, F., Geffard, M. and Privat, A., Preand post-natal ontogeny of serotonergic projections to the rat spinal cord, J. Neurosci. Res., 22 (1989) 305-321. 38 Reichlin, S., Neural functions of TRH, Acta Endocrinol., Suppl. 276 (1986) 21-33. 39 Romagnano, M.A., Braiman, J., Loomis, M. and Hamill, R.W., Enkephalin fibers in autonomic nuclear regions: Intraspinal vs supraspinal origin, ]. Comp. Neurol., 266 (1987) 319-331. 40 Romagnano, M.A., Newton, B.W., Pierre, A.H. and Hamill, R.W., Postnatal ontogeny of enkephalin fibers in spinal sympathetic nuclei, Dec. Brain Res., 46 (1989) 263-279. 41 Schmidt-Achert, M.K., Askanas, V. and Engel, K.W., Thyrotropin-releasing hormone enhances choline acetyltransferase and creatine kinase in cultured spinal ventral horn neurons, J. Neurochem., 43 (1984) 586-589. 42 Schramm, L.P., Stribling, J.M. and Adair, J.R., Developmental reorientation of sympathetic preganglionic neurons in the rat, Brain Res., 106 (1976) 166-171. 43 Senba, E., Shiosaka, S., Hare, Y., Inagaki, S., Takatsuki, K., Kawai, Y. and Tohyama, M., Ontogeny of the peptidergic system in the rat spinal cord: immunocytochemical analysis, J. Comp. Neurol., 208 (1982) 54-66. 44 Shieh, J.Y., Leong, S.K. and Wong, W.C., Origin of the rubro-spinal tract in neonatal developing and mature rats, J. Comp. Neurol., 214 (1983) 79-86. 45 Tsuruo, Y., 'H6kfelt, T. and Visser, T., Thyrotropin releasing hormone (TRH).immunoreactive cell groups in the rat central nervous system, Exp. Brain Res., 68 (1987) 213-217. 46 Ulfhake, B., Arvidsson, U., Cullheim, S., H6kfelt, T., Brodin, E., Verhofstad, A. and Visser, T., An ultrastructural study of 5-hydroxytryptamine-, thyrotropin-releasing hormone- and substance P.immunoreactive axonal boutons in the motor nucleus of spinal cord segments L7-S! in the adult cat, Neuroscience, 23 (1987) 917-929. 47 Valentino, K.L., Jones, E.G. and Kane S.A.. Expression of glial fibrillary acidic protein immunoreactivity during development of long fiber tracts in the rat central nervou, system. Dec. Brain Res., 3 (1983) 317-336. 48 Weill, C.L., The prevention of natural motoneuron cell death by thyrotropin-releasing hormone, $oc. Neuroscl. Abstr., 13 (1987) 641. 49 White, S.R., A comparision of the effects of serotonin, substance P, and thyrotropin-releasin8 hormone on excitability of rat spinal motoneurons in vivo, Brain Res., 335 (1985) 63-70. 50 Winokur, A. and Utiger, R.D., Thyrotropin-releasing hormone: regional distribution in the rat brain, Science, 185 (1974) 265-267.