- Email: [email protected]
Telencephalic projections to the goldfish hypothalamus: An anterograde degeneration study
Brain Research
Bderin,
Vol. 20, pp. 503-514.
o
Pergamon Press
plc, 1988. Printed
in the
U.S.A.
0361-9230/88
$3.00 + .OO
Telencephalic Projections to the Goldfish Hypothalamus: An Anterograde Degeneration Study MARK
J. AIRHART,*’
JAMES
0. SHIRK*
AND
RICHARD
M. KRIEBELt
*Department of Anatomy, Quillen-Dishner College of Medicine East Tennessee State University, Johnson City, TN 37614 and tDepartment of Anatomy and Neurobiology, University of Vermont College of Medicine, Burlington, VT 05404 Received
10 July
1987
M. J., J. 0. SHIRK AND R. M. KRIEBEL. Telencephalic projections to the goldfish hypothalamus: An degeneration study. BRAIN RES BULL 20(4) 503-514, 1988.-In this study, large areas of goldfish telencephalon were ablated including rostra1 nucleus preopticus periventriculare (rNPP), and degenerating axons were traced by a modified Fink and Heimer procedure. The lesioning procedure ablated large regions of area dorsalis telencephali pars medialis, centralis, and dorsolateral complex; and completely removed area ventralis telencephali pars dorsalis, ventralis, and lateralis. In addition, the supracommissural nucleus and rNPP were lesioned specifically because both nuclei have been thought to be involved in courtship behavior and endocrine control of reproduction. This investigation demonstrated extensive fiber projections from telencephalic nuclei and/or rNPP to the hypothalamus. Lesioned telencephalon and/or rNPP projected bilaterally to nucleus preopticus and the suprachiasmatic nucleus and unilaterally to the following tuberal nuclei: nucleus anterior tuberis, and the lateral hypothalamic nucleus. A much larger fiber projection to the inferior lobe nuclei was also observed with a large contralateral as well as ipsilateral input. AIRHART,
anterogrude
Anterograde degeneration techniques Preoptic and hypothalamic afferents
Goldfish
Telencephalic
SEVERAL studies have shown that the telencephalon projects fibers to the hypothalamus in teleosts [ 11, 17,26,29,33, 371. All teleosts, so far examined, show a relatively large number of telencephalohypothalamic fibers traveling within the lateral forebrain bundle (LFB) and projecting to areas just dorsal and ventromedial to nucleus recessus lateralis and to regions within the inferior lobe. The presence of telencephalic efferents to the tuberal region, however, depends on the species of fish studied. An extensive projection to the tuberal region has been shown in the hime salmon (0. nerka) [33] and the bichir (Polypterus palmas) [281. In contrast, in goldfish (C. auratus) and other teleosts this projection is either lacking or extremely sparse 111, 17, 26, 29, 371. The cumulative observations on most teleosts including goldfish suggest that telencephalic connections to the hypothalamus primarily terminate in nontuberal nuclei. Recent behavioral studies in goldfish, however, suggest that the telencephalon plays a major role in courtship behavior and spawning [21,22]. Combined lesion and behavioral
projections
studies have shown that bilateral ablation of two ventral telencephalic nuclei (area ventralis pars supracommissuralis, V, and/or area ventralis pars ventralis, V,,) or a rostral preoptic nucleus (nucleus preopticus perventriculare, NPP) severely impairs courtship behavior and spawning in male goldfish [21,22]. These same nuclei as well as tuberal nuclei have been shown to bind sex steroids in several teleosts including goldfish [8, 9, 19, 201. These studies suggest more extensive connections between telencephalon including rostral NPP and hypothalamus, and prompted us to reinvestigate telencephalohypothalamic fibers in goldfish. Silver degeneration techniques were used to trace the connections between goldfish telencephalon including rostral NPP and hypothalamus. A modified Fink and Heimer procedure was used to label lesioned efferent fibers [2]. This study has shown that a significant number of telencephalic and/or rostral NPP fibers project to preoptic and tuberal regions of goldfish hypothalamus, and that these fibers more extensively innervate the inferior lobes than previously reported [29].
‘Requests for reprints should be addressed to Mark J. Airhart, Ph.D., Department of Anatomy, Box 19%OA, Quillen-Dishner Medicine, East Tennessee State University, Johnson City, TN 37614.
503
College of
AIRHART. ABBREVIATIONS
I. AC 2. BV 3. cc 4. CE 5. CM 6. CN III 7. CT 8. D, 9. DF 10. DIG
11. D, 12. FVLTrO 13. H 14. HC 15. LFB 16. LH 17. MFB 18. NAH 19. NAT 20. NCH 21. NDL 22. NE 23. NG 24. NLTI 25. NLT, 26. NPGI 27. NPG,,,
anterior commissure blood vessel corpus of the cerebellum nucleus centralis of the inferior lobe corpus mamillare cranial nerve III connective tissue area dorsalis telencephali pars centralis nucleus diffusus of the inferior lobe dorsolateral nuclear complex area dorsalis telencephali pars medialis fasciculus ventrolaterahs tractus opticus habenula horizontal commissure lateral forebrain bundle nucleus lateralis hypothalami medial forebrain bundle nucleus anterioris hypothalamicus nucleus anterior tuberis nucleus cerebellosus hypothalami nucleus dorsolateralis thalami nucleus entopeduncularis nucleus glomerulosis nucleus lateral tuberis pars lateralis nucleus lateral tuberis pars posterioris nucleus preglomerulosus pars lateralis nucleus preglomerulosus pars medialis
‘8. NPO 29. NPO,, 30. NPP 31. NPP,
32. NPT 33. NR 34. NRL 3.5. NRP 36. NT 37. NTP 38. NVM 39. OT 40. OT, 41. PC 42. SCN 43. TE 44. TL 45. I-L,, 46. V 47. VB 48. VC 49. Vd 50. vHC 51. vp 53. V, 53. v,
METHOD
Animds
Common goldfish (Carassius auratus), 7-10 cm long, were obtained from Grassyfork Fisheries (Martinville, IN) and maintained in aquaria at 22-24°C under normal daynight laboratory conditions. Surgery
Fish were anesthetized in 0.03% MS-222 (Sandoz) and surgery performed under a dissecting microscope. The telencephalon was approached through a small bone flap incised in the calvaria. The lesioning procedure consisted of gently separating the telencephalic lobes to visualize the anterior commissure and transsecting at approximately a right angle the right telencephalon at the level of the anterior commissure with a microscalpel. The microscalpel was constructed from a razor blade 1 mm in width and held by a hemostat mounted on a micromanipulator. The approximate position of the transection is shown schematically in Fig. 1. To remove the telencephalon rostra1 to the transsection, the anterior commissure was sectioned and approximately twothirds of the rostral right telencephalon was ablated by aspiration. The extent of the ablation was determined by examining alternate cresyl violet stained sections of the remaining caudal telencephalic segment. Only animals showing a completely intact nucleus preopticus (NPO) were included in this study. Silver Degeneration
Procedures
Eight experimental fish (5 males, 3 females) were anesthetized in 0.03% MS-222 and killed by intracardial perfusion with 10.0% buffered formalin (pH=7.2) four days after
SHIRK AND KRIEBEI
--
nucleus preopticus nucleus preopticus pars parvocellulari\ nucleus preopticus periventricularis nucleus posterioris periventricularis nucleus posterior tuberis nucleus rotundus nucleus recessus laterahs nucleus recessus posterioris nucleus tenia nucleus posterioris thalami nucleus ventromedialis thalami optic tectum optic tract posterior commissure suprachiasmatic nucleus telencephalon torus longitudinalis nucleus of the torus lateralis ventricle ventral brachium valvules of the cerebellum area ventralis telencephali pars dorsalis ventral branch of the horizontal commissure area ventralis telencephali pars posterioris area ventralis telencephali pars supracommissuralis area ventralis telencephali pars ventralis
surgery. Brains were embedded in 20% gelatin, sectioned transversely at 33 pm with a freezing microtome, and alternating sections were stained with cresyl violet or a modification of the Fink and Heimer procedure [2] described by Airhart and Kriebel. Briefly, this modification consisted of a reduction in the silver nitrate concentration in the presoak step (0.4%) and the ammoniacal solution (1.4%), and extending treatment with potassium permanganate (O.OS%, 25 min). These modifications resulted in a decrease in background levels of non-specific silver grains. A survival time of four days was chosen in these studies because a higher density of degenerating profiles were observed in the hypothalamus at this time in preliminary experiments which compared 2, 4, and 6 days postsurgical fish. The modified Fink and Heimer technique was chosen for these studies because preliminary experiments with HRP suggested the presence of several reciprocal connections between telencephalon and hypothaHRP is transported both lamic nuclei [ 11. Since anterogradely and retrogradely, the afferent or efferent nature of labeled fibers could not be established unambiguously unless a specific anterograde tracer technique was employed. In all experimental animals, degenerating silver impregnated fibers, preterminals, and terminals in silverstained sections were plotted with a camera-lucida. The nomenclature used to describe nuclear regions in this report was compiled from several sources [6, 23, 27, 303. In most instances, however, we used the nomenclature of Peter and Gill [30] and have specifically indicated when nuclei were described in accordance with the terminology of others. Control Studies
In two animals,
a bone flap was reflected as described
TELENCEPHALOHYPOTHALAMIC
PROJECTIONS
FIG. 1. Schematic transverse sections through the telencephalon with the most rostral section (A) at the far right. The middle section (B) represents the level of telencephalic transection. Rostral NPP (section B) and NPO (section C) are separated along the rostrocaudal axis by 300 ym, and sections A and B separated by 400 pm. The nomenclature used to describe telencephalic
nuclei is derived from Niewenhuys
above and the surrounding ventricle of the telencephalic lobe was tom, but the telencephalon itself was left intact. Four days later, the animals were perfused, and brains were processed for silver impregnated elements using the modified Fink and Heimer procedure described above. Hy~th~amic sections from control animals showed only occasional, randomly placed, silver-impregnated profiles. The density of these profiles was no greater in hypothalamic regions than elsewhere in the control brains. An additional two animals were used to determine if efferents from the olfactory bulb projected beyond the telencephalon. A bone flap was reflected in these animals and the right olfactory tract cut. The right olfactory bulb and proximal olfactory tract were aspirated from the cranial cavity in both animals. Animals were killed 4 days later by intracardial perfusion and their brains processed by a modified Fink and Heimer procedure. Animals showed no extratelencephalic olfactopetal fibers. RESULTS
In all experimental animals, the right telencephalic lobe was severed at the level of the anterior commissure and telencephalon rostral to the cut removed (Fig. 1). This lesioning procedure was adopted to remove as much of telencephalon as possible without damaging the preoptic nuclei (Fig. 1). Lesioning nucleus preopticus m~oce~ul~s and/or nucleus preopticus parvocellularis would have compromised the interpretation of results because both nuclei send axons to the pituitary and the hypothesized axon trajectory is via the tuberal region [ 14,151.
1271and Levine and Dethier [23].
Histological examination of alternate cresyl violet stained sections through the remaining caudal telencephalon of all experimental animals showed that the ablation removed large regions of both dorsal and ventral telencephalic nuclei as well as rostral NPP. The specific telencephalic nuclei ablated included a major percentage of area dorsalis pars medialis, centralis, and dorsolateral complex; and within ventral telencephalon complete ablation of area ventralis pars dorsalis, ventralis, lateralis, a large part of supracommissuralis, and the rostral part of posterioris (Fig. 1). The projection pattern of lesioned nuclei was essentially the same in all experimental animals. The results have been summarized by line diagrams depicting the topographic route taken by efferent fibers to their target nuclei (Fig. 2) and photomicrographs showing examples of the putative terminal fields. The term putative is used to qualify “terminal fields” because at the light microscopic level degenerating profiles can not be accurately distinguished as terminals, preter~n~s, or fibers. The terms ipsilateral and contralateral relate to the side of the brain associated with the lesioned telencephalon. The projection from the lesioned telencephalon was bilateral with the majority of silver impregnated degenerating profiles ipsilateral to the lesioned telencephalon (Fig. 2). The contralateral projection originated, at least in part, from lesioned fibers in the ipsilateral telenceph~on crossing through the anterior commissure. Lateral
Forebrain
Bundle (LFB)
The course and projection pattern of lesioned fibers in the
zot!
N
THIS AND FOLLOWING PAGE FIG. 2. (A-F). Line diagrams of transverse sections through the preoptic region and hypothalmus illustrating the distribution of degenerating profiles four days after lesion. The side ipsilateraf to the lesioned teIe~ceph~on is on the left. Degenerating protiiles indicated by dots may represent terminals, preterminals, or fibers cut in cross section: dashed lines indicate fiber degeneration. The sections are organized sequentially rostral (A) to caudal (E).
TELENCEPHALOHYPOTHALAMIC
507
PROJECTIONS
-
NRL
NAT
ipsilateral LFB will be described followed by a description of degeneration in the contralateral LFB. Lesioned fibers within the ipsilateral LFB coursed through the lateral region of the preoptic area and continued along the medial margin of the optic tract (Figs. 2A; and 3a,b). Degenerating profiles were observed in the suprachiasmatic nucleus (SCN) but the source of this projection was not clear (Figs. 2A,B and 3a,c); fibers may have reached the nucleus from both the medial forebrain bundle (MFB) and LFB. The SCN has been shown to receive retinofugal fibers and because of its position to the optic chiasm has been named the suprachiasmatic nucleus [6, 34, 353. At the level of the horizontal commissure (HC) degenerating fibers in the ipsilateral LFB projected ven-
trolaterally to travel along the lateral margin of the rostra1 tuberal region. A relatively small number of degenerating fibers coursed towards the pituitary stalk with a sparse number of fibers reaching the dorsal part of nucleus lateralis tuberis pars lateralis (NLTl). Fibers were not seen reaching the origin of the pituitary stalk (Fig. 2B,C). Within this same region, degenerating fibers from the LFB were observed around the dorsolateral, ventrolateral, and ventral periphery of nucleus anterior tuberis (NAT) with a few degenerating profiles projecting into the nucleus (Figs. 2B,C and 4b). Telencephalic connections with the glomerular complex of nuclei which includes: nucleus preglomerulosis lateralis (NPGl), nucleus preglomerulosis medialis (NPGm), and nu-
X)X
AIRHART.
SHIRK AND KRlt+BEl.
FIG. 3. (a) A photomicrograph at low magnification of the ipsilateral preoptic area at the level of the SCN. Degenerating profiles can be observed in the SCN and LFB. Note the absence of degenerating profiles in the optic tract (OTr) that lies directIy adjacent to the LFB. Scale bar 100 pm. (b) An enlargement of the LFB from the area in section a delineated by the black rectangle. The degenerating profiles. examples indicated by arrow. show variability in size and many profiles are out of the plane of focus. Scale bar IS em. (c) AR enlargement of the degenerating profiles found within the SCN shown in section a. Scale bar IS grn.
TELENCEPHALOHYPOTHALAMIC
509
PROJECTIONS
cleus glomerulosis (NG) were not included in this report because embryonic studies by Bergquist [4,51 suggest that these nuclei are part of the posterior tuberculum of the thalamus. We are presently studying the connections between the telencephalon and thalamus and will publish our findings in a later report. At the rostral level of nucleus recessus lateralis (NRL) degenerating profiles from the ipsilateral LFB were found in neuropil between NPG, NG, and the dorsal part of NRL; and bordering the lateral, ventral, and ventro-medial parts of NRL (Figs. 2D, 5). Degenerating profdes were also observed in NRL and the dorsal part of nucleus di@usus of the inferior lobe as described by Braford and Northcutt (Figs. 2D,E, 5). This area was delineated by Peter and Gill [30] as nucleus diffusus torus lateralis. In this same region, a relatively small number of degenerating profiles were observed in the caudal part of NAT and in the dorsolateral half of nucleus lateralis hypothalami, LH [6] (Fig. 2D). In the caudal region of the hy~th~~us (Fig. 2E,F) a large number of degenerating elements, which appeared to come from the LFB, were found in nucleus cerebellosus hypothalamicus (NCH) and all nuclei of the ipsilateral inferior lobe. Degenerating profiles were dispersed throughout the following nuclei: NCH, corpus mammillare (CM), and nucleus centralis of the inferior lobe (CE), the later nucleus was described by Braford and Northcutt [6] (Fig. 2&F). A more regional dist~bution of degenerating elements was observed in the laterally positioned part of nucleus di&sus of the inferior lobe (DF). In this region, DF contained randomly distributed putative terminals throughout most of its volume, except at its lateral and ventrolateral periphery where in transverse section a region approximately one-third of its area was delete of silver impregnated profiles (Fig. 2E,F). Telenceph~ic efferent fibers in the contralateral LFB followed the same rostrocaudal route through the brain but a small difference in the density of degenerating profiles was noted between the two pathways, the contralateral LFB contained a lower density of degenerating fibers. Differences in the relative number of degenerating profiles were also observed between corresponding ipsilateral and contralateral nuclei (Fig. 2A-F). In the preoptic region there was a slightly lower density of degenerating profiles in the contralateral SCN; and in the tuberal region no degenerating profiles were observed in contralateral NLTl or NAT (Fig. 2B,C). Further caudally degenerating profiles were also absent from contralateral LH (Fig. 2D). Contralateral nuclei of the caudal hy~thalamus including: NCH, NRL, CM, CE, and DF demonstrated the same distribution of lab&d profiles but within each nucleus the density of labeled profiles was slightly lower than the corresponding ipsilateral nuclei (Figs. 2D,E,F and 6a,b). Median Forebrain
FIG. 4. (a) Degenerating profiles (arrow) within a fiber pathway o~~nating from the MFB and located just medial to NAT. The photograph is oriented such that medial is to the left and dorsal at the top. The complete extent of this pathway is shown in section 2C. Scale bar 15 pm. (b) A photomicrograph of NAT showing a relatively small number of degenerating profiles (arrows). Medial is to the left and dorsal at the top. Scale bar 15 Frn.
Bundle (MFB)
In the preoptic region, both ipsilateral and contralateral MFB’s contained degenerating fibers primarily located along their ventral borders. At the level of the SCN degenerating profiles from both MFB’s were observed projecting ventrally along the lateral borders of NPO with a few profiles observed within parts of NPO bitaterally (Fig. 2A,B). Degenerating fibers within ventral projections of both MFB’s appeared to terminate in SC nuclei with the ipsilateral SCN containing a slightly larger density of degenerating elements (Fig. 2A,B). Further caudally, degenerating profiles branching from both MFB’s projected ventrally along the lateral border of nucleus preopticus periventriculare (NPP,) with some
AIRHART.
SHIRK AND KRIEBEL
FIG. S. Photomicrograph of rostra1 NRL oriented such that lateral is to the right and dorsal at the top. Degenerating observed within the dorsal and lateral parts of NRL as well as around its dorsal, lateral, and ventromedial periphery. RL. contains a film of gelatin with silver precipitate on its surface. Scale bar 50 wrn.
profiles observed within the nucleus bilaterally (Figs. 2C, and 7a,b). Ipsilateral fibers projected further ventrally coursing along the medial border of NAT (Fig. 4a) with some degenerating profiles observed within the medial region of NAT (Figs. 2C and 4a). The projection from the telencephalon and/or NPP through the MFB’s appeared to end at the level of rostra1 NRL (Fig. 2D).
DISCUSSION
Our results strongly suggest the presence of new connections between ablated telencephalon and/or rostra1 NPP and preoptico-hypothalamic nuclei. These connections include efferents projecting bilaterally to NPO, SCN, NPP,,, NCH. and unilaterally to NAT, and LH. In addition, our results also show a much larger projection to the ipsilateral inferior lobe nuclei [ 11,291 and a previously unreported projection to contralateral inferior lobe nuclei (DF, CE, and CM). These results receive further support from a preliminary HRP study in which HRP was injected into a telencephalic lobe at the level of the anterior commissure in goldfish [ 11. The projecFOLLOWING
profiles (arrows) can be Part of recessus lateralis.
tion pattern of labeled fibers and nuclei containing labeled elements completely corresponded to the data presented in this paper. Because the ablated area includes several telencephalic nuclei in area dorsalis and ventralis and NPP neurons, we were unable to make conclusions regarding what specific nuclei were responsible for individual projections. Further studies involving more selective lesions are required as well as tracer studies from specific terminal fields. Three items effecting the interpretation of our results require further discussion. The first concerns the validity of the modified Fink and Heimer procedure. This procedure was tested in an earlier investigation examining telencephalotectal efferents by comparing the results of electron microscopic degeneration studies and the silver degeneration techniques [2]. This study showed a positive correlation between the presence of silver impregnated degenerating profiles in a tectal lamina and the presence of degenerating axons or terminals in the same lamina. A second item concerns the bilaterality of the projection. Correct interpretation of the bilaterality of the projection requires considering
PAGE
FIG. 6. (a) Photomicrograph of the contralateral inferior lobe showing degenerating profiles in CM, CE, and DF. Due to the low magnification, degenerating elements in NRL are not visible. Scale bar 100 pm. (b) A higher magnification of degenerating profiles in CE. Note the small punctate nature of the degenerating profiles suggesting the presence of terminal fields. Scale bar 50 km.
TELENCEPHALOHYPOTHALAMIC
PROJECTIONS
511
AIRHART.
FIG. 7. (a) Photomicrograph of rostra1 NPP,. ipsilateral to the lesion. showing the presence of a few degenerating profiles within the nucleus (arrow) and a larger density of degenerating elements along its lateral border. Scale bar I5 Wm. (b) Photomicrograph of both ipsilateral and contralateral NPP,, at a level caudal to section a, the ipsilateral side is on the left. Note the presence of degenerating profiles in both nuclei (arrows). Scale bar 15 pm.
whether the contralateral telencephalon was inadvertantly damaged during the lesioning procedure. In each experimental animal, the telencephalic lobe contralateral to the lesion was examined histologically for potential damage; no obvious indicators such as gliosis or cellular disorganization were observed. A third consideration is the possibility of confusing olfactory and telencephalic efferents. We addressed this problem by determining if olfactory efferents projected to targets outside the telencephalon. No bulbopetal fibers in extra-telencephalic regions were observed. This observation is in agreement with the findings of Oka and Ueda [29], but not with two recent HRP studies that traced olfactory bulb connections in goldfish. Both investigations showed a small number of bulbopetal fibers from one olfactory bulb projecting bilaterally through the MFB’s. These fibers passed through or terminated in NPP and an area just rostra1 to nucleus posterior tuberis designated as either the caudal olfactory nucleus or tuberal terminal field [3,23]. One
SHlRK
AND
KRIEBEt.
of these reports also eluded to the presence of a small number of HRP-labeled fibers in the vicinity of NLTI (231. We assume that these projections were not observed in our study because the survival time was not appropriate to detect argyrophilia in these extratelencephalic fibers. However, because of the discrepancy between our control study and the HRP studies, we have not drawn conclusions regarding projections which overlapped with the known olfactory bulb connections to NLTI. There have been only three studies examining telencephalic projections in goldfish. One of these studies by Oka and Ueda [29] used silver degeneration techniques. ‘Their results showed no telencephalic projection to preoptic or tuberal nuclei and a relatively small projection to the ipsilatera1 inferior lobe; nearly all telencephalohypothalamic efferents were observed in neuropil. The differences in results between the work of Oka and Ueda [29] and the present findings may be due to differences in survival times. Although Oka and Ueda used goldfish of approximately the same size as in the present investigation, the survival times were longer, i.e., 612 days. Previous silver degeneration studies have shown that the rate of phagocytosis and loss of argyrophilia is slower for axons than associated preterminals and terminals. Therefore, the longer survival times used by Oka and Ueda may have selected for axons and resulted in complete phagocytosis of degenerating terminals in hypothalamic nuclei. A second possible explanation is that our modification of the Fink and Heimer technique is more sensitive than the technique used by Oka and Ueda (Ebbesson’s rnodification of the Fink and Heimer method) [IO). Our results also differ from two previous HRP studies that traced connections between telencephalon and hypothalamus in C‘. UUIY~~U.) [I I] and C‘. c trrcr.stirt.t and C’xpr-iuu.v cwpio 1171. Both studies focussed on connections between telencephalon and thalamus, however, all putative efferent fibers and nuclei containing labeled cells were reported. Both studies showed labeled fibers in the ipsilateral LFB projecting to the medial part of the inferior lobe. No labeled fibers were observed in SCN or the tuberal region of the hypothalamus. The absence of such fibers may have resulted from the location of injection sites. In both studies, HRP was injected into only dorsal telencephalic nuclei. In the hime salmon (0. nc~&rr) ventral telencephalic nuclei were lesioned (V,. V,) [ 331 and silver degeneration techniques demonstrated an extensive bilateral projection to tuberal nuclei. Degenerating fibers and/or terminals were observed bilaterally in NLTI, nucleus lateralis tuberis. pars medialis (NLTm). LH, and NAT with a majority of degenerating profiles on the ipsilateral side. A similar projection was observed in the birchit-, P. &V(I.P [28]. Large HRP injections, including both dorsal and ventral telencephalic nuclei, labeled fibers projecting bilaterally to the tuberal region. These results support our findings and emphasize the potential importance of ventral telencephalic nuclei in modulating hypothalamic function. The control of circadian rhythms and secretion of pituitary hormones may involve several preoptic-hypothalamic nuclei that receive afferents from lesioned telencephalic nuclei and/or nPP. A nucleus homologous to the suprachiasmatic nucleus in mammals has been shown in goldfish. This nucleus receives projections from both eyes with the majority of fibers coming from the contralateral retina [34], as it is in most other species [2S.32]. In addition to retinal input, the present study has shown that SCN also receives a bilateral projection from each telencephalic lobe perhaps including NPP. with the majority of afferent fibers projecting to the
TELENCEPHALOHYPOTHALAMIC
513
PROJECTIONS
ipsilateral SCN. Such a projection has not been reported in other teleost species but has been observed in some mammals [24,32,36]. To suggest a function for these fibers would be premature since the physiology of SCN in teleosts is speculative [ 161. A number of reports using both immunohistochemical and biochemical techniques have suggested that a gonadotropin release-inhibitory factor, thought to be dopamine, is found in NPP neurons, and that these neurons project via the anterolateral tuberal region to the pituitary [7, 18, 311. Although the lesioned area in this study included rostra1 NPP, we did not observe degenerating profiles reaching the pituitary stalk region. Perhaps not enough of NPP was lesioned and/or the labeling techniques were not sensitive enough to detect the proposed fibers. Our results, however, have shown that at least some hypothalamic
nuclei that project
axons
to
the pituitary receive telencephahc and/or NPP afferents. These nuclei include NPP, and NRL [13,14]. Nucleus posterioris periventriculare and NRL compose part of the paraventricular organ of goldfish and both contain large ~pulations of ~atecholamine confining neurons projecting to all regions of the goldfish pituitary 1131. The presence of degenerating profiles in NPP, and NRL suggests that lesioned areas of telencephalon and/or NPP may modulate the activity of these nuclei and indirectly effect pituitary function.
ACKNOWLEDGEMENTS
The authors thank Cindy Canter for typing the manusc~pt. This work was supported by an ETSU Institutional Grant to M.J.A. and PHS 5429-16-19 to R.M.K.
REFERENCES 1. Airhart, M. J. Axonal connections between telencephalon and hypothalamus in goldfish. Anat Ret 218: SA, 1987. 2. Airhart, M. J. and R. M. Kriebel. Telencephalic terminals in the major retinal synaptic lamina of the goldfish optic tectum. Brain Res 336: 363-367,
1985.
3. Bartheld, C. S. von, D. L. Meyer, E. Fiebig and S. 0. E. Ebbesson. Central connections of the olfactory bulb in the goldfish, Carassius auratus. Cell Tissue Res 238: 475-487, 1984. 4. Bergquist, H. Zur Morphologie des Zwischenhims bei niederen Wirbeltieren. Acto Zoo1 13: 57-304, 1932. 5. Bergquist, H. Ontogenesis of diencephalic nuclei in vertebrates. A comparative study. Lunds ArssKr Aud 2 50: I-34, 1954. 6. Bradford, M. R. and R. G. Northcutt. Or~nization of the diencephalon and pretectum of the ray-finned fishes. In: Fish Neurobiology, ~01 II, edited by R. E. Davis and R. G. Northcutt. Ann Arbor: University of Michigan press, 1983, pp. 117163. 7. Chang, J. P. and R. E. Peter. Effects of dopamine on gonado. . .r . .._. tropm release m tern&e goldtish, carassms auratus. Neuroendocrinology 36: 351-357, 1983. 8. Davis, R. E., J. I. Morel1 and D. W. Pfaff. Autoradiographic localization of sex steroid-concentrating cells in the brain of the teleost Macrop~us opercularis (Oste~chthyes: Belontiidae). Gen Camp Endocrinol33: 496-505, 1977. 9. Demski, L. S. Neuroanatomical substrates of reproductive behavior in male sunfish (Genus Lepomis). Ann Biol Anim Biochime Biophys 18: 831-836,
1979.
10. Ebbesson, S. 0. E. The selective impregnation of degenerating axons and their synaptic endings in nonmammalian soecies. In: Contemporary Research Methids in Neurcranatomy,‘edited by W. J. H. Nauta and S. 0. E. Ebbesson. Berlin: Springer-Verlag, 1970, pp. 132-161. 11. Echteler, S. M. and W. M. Saidel. Forebrain connections in the goldfish support telencephalic homologies with land vertebrates. Science 212: 683-685, 1981. 12. Fink, R. P. and L. Heimer. Two methods for selective silver impregnation for degenerating axons and the synaptic endings in the central nervous system. Brain Res 4: 369-374, 1967. 13. Fryer, J. N., C. Boudreault-Chateauve~ and R. P. Kirby. Pituitary aBerents originating in the paraventricular organ (PVO) of the goldfish hypothalalmus. .I Comp Neurol242: 475-484, 1985. 14. Fryer, J. N. and L. Maler. Hypophysiotropic neurons in the gold&h hy~t~amus demonst~ted by retrograde transport of horseradish peroxidase. Cell Tissue Res 218: 93-102, 1981. 15. Gregory, W. A. and C. D. Tweedle. Horseradish peroxidase evidence for a spinal projection from the preoptic area of the goldfish, a light and electron microscopic study. Brain Res 341: 82-91,
1985.
16. Hontela, A. and R. E. Peter. Effects of pinealectomy, blinding and sexual conditions on serum gonadotropin levels in the goldfish. Gen Camp ~ffdu(.rin~~i 40: 168-179, 1980. 17. Ito, H., T. Murakami and Y. Morita. An indirect telencephalocerebellar pathway and its relay nucleus in teleosts. Brain RL’S 249: I-13, 1982. 18. Kah, O., P. Chambolle,
of dopaminergic
J. Thibault and M. Geffard. Existence neurons in the preoptic region of the goldfish.
Neurosci Lett 48: 293-298. 1984. 19. Kim, Y. S., W. E. Stumpf and M. Sar. Topography
20.
21.
22.
23.
of estrogen target cells in the forebrain of goldfish, Carassius auratus. J Camp Nrurol 182: 61 l-620, 1978. Kim, Y. S., W. E. Stumpf, M. Sar and M. C. Ma~ine2-Vargas. Estrogen and androgen target ceIIs in the brain of fishes, reptiles and birds: Phylogeny and ontogeny. Am Zoot 18: 425-433,1978. Koyama, Y., M. Satou, Y. Oka and K. Ueda. Involvement of the telencephalic hemispheres and the preoptic area in sexual behavior of the male goldfish, Carassius auratus: a brain-lesion study. Behav Neural Bioi 40: 70-86, 1984. Kyle, A. L. and R. E. Peter. Effects of forebrain lesions on spawning behavior in the male goldfish. Phvsiol Behav 28: 1103-l 109, 1982. Levine, R. L. and S. Dethier. The connections between the olfactory bulb and the brain in the aoldfish. .I Coma Neural 237:
427-444; 1985. 24. Meiback, R. C. and A. Siegel. Efferent connections of the hippocampal formation in the rat. Bruin Res 124: 197-224, 1977.
25. Moore, R. Y. Central neural control of circadian rhythms. In: Frontiers in Newoendocrinology, vol 5, edited by W. F. Ganong and L. Martini. New York: Raven Press, 1978, pp. 185-206. 26. Murakami, T., Y. Morita and H. Ito. Extrinsic and intrinsic fiber connections of the telenceph~on in a teleost, Sebastiscus marmoratus. f Camp Neurol 216: 115-131, 1983. 27. Nieuwenhuys, R. The comparative anatomy of the actinopterygian forebrain. J Hirnforsch 6: 171-192, 1963. 28. Northcutt, R. G. Localization of neurons afferent to the telencephalon in a primitive bony fish, Polypterus palmas. Neurosci Lett 22: 219-222, 1981. 29. Oka, Y. and K. Ueda. Telencephalic
projections in the goldfish (Carassius auratus): an anterograde degeneration study. / Fuc Sci Univ Tokyo 15: 1-8, 1981. 30. Peter, R. E. and V. E. Gill. A stereotaxic atlas and technique for forebrain nuclei of the goldfish, Carassius auratus. J Camp Neural 159: 69-102, 1975. 31. Peter, R. E. and C. R. Paulencu. Involvement of the preoptic region in gonadotropin release-inhibition in goldfish, Carassius au&us. Ne~4roendo~.r~noiogy 31: 133-141, 1980.
AIRHAR’I‘.
iI
32. Pickard. nucleus
Ci. E. The afferent connections of the golden hamster with
of the suprachiasmatic emphasis on the ret-
inohypothalamic projection. .I C‘ojtrp .Y~lrr~j/ 211: C-83. 1982. 33. Shiga, ‘T.. Y. Oka. M. Satou, N. Okumoto and I<. Ueda. Effercnts from the supracommissural ventral telencephalon in the Hime Salmon (landlocked red salmon. Oncorhynchus nerka): An anterograde degeneration study. Hrtrirl KC,.\ H/i// 14: 55-61. 19x.
SHIRK
AND
KRIkBEL.
34. Springer. A. D. and .I. S. Gaffney. Retinal projections 111the goldfish: 3 study using cobaltous-lysine, .! ( ~J,J,,I ~\‘~,~frr~/203: 401-424. 1981. 35. Springer. A. D. and A. S. Mednick. Selective Innervation ofthe goldfish suprachiasmatic nucleus by ventral retinal ganglion cell axons. Brtri/, KC,.\ 323: 293-796. 1984. 36. Swamon. L. W. and W. M. Cowen. The connections of the scptal region in the rat. .I C‘CJ!,I/>,l’~,rlro/ 186: Q-656, 1979. 37. Vanegas. H. and S. 0. E. Ebbesson. Telencephalic projections in two teleost species. .I ( ‘~j/fr/> h’c,rlr~~/ 165: IXI-196. 1976.
Bderin,
Vol. 20, pp. 503-514.
o
Pergamon Press
plc, 1988. Printed
in the
U.S.A.
0361-9230/88
$3.00 + .OO
Telencephalic Projections to the Goldfish Hypothalamus: An Anterograde Degeneration Study MARK
J. AIRHART,*’
JAMES
0. SHIRK*
AND
RICHARD
M. KRIEBELt
*Department of Anatomy, Quillen-Dishner College of Medicine East Tennessee State University, Johnson City, TN 37614 and tDepartment of Anatomy and Neurobiology, University of Vermont College of Medicine, Burlington, VT 05404 Received
10 July
1987
M. J., J. 0. SHIRK AND R. M. KRIEBEL. Telencephalic projections to the goldfish hypothalamus: An degeneration study. BRAIN RES BULL 20(4) 503-514, 1988.-In this study, large areas of goldfish telencephalon were ablated including rostra1 nucleus preopticus periventriculare (rNPP), and degenerating axons were traced by a modified Fink and Heimer procedure. The lesioning procedure ablated large regions of area dorsalis telencephali pars medialis, centralis, and dorsolateral complex; and completely removed area ventralis telencephali pars dorsalis, ventralis, and lateralis. In addition, the supracommissural nucleus and rNPP were lesioned specifically because both nuclei have been thought to be involved in courtship behavior and endocrine control of reproduction. This investigation demonstrated extensive fiber projections from telencephalic nuclei and/or rNPP to the hypothalamus. Lesioned telencephalon and/or rNPP projected bilaterally to nucleus preopticus and the suprachiasmatic nucleus and unilaterally to the following tuberal nuclei: nucleus anterior tuberis, and the lateral hypothalamic nucleus. A much larger fiber projection to the inferior lobe nuclei was also observed with a large contralateral as well as ipsilateral input. AIRHART,
anterogrude
Anterograde degeneration techniques Preoptic and hypothalamic afferents
Goldfish
Telencephalic
SEVERAL studies have shown that the telencephalon projects fibers to the hypothalamus in teleosts [ 11, 17,26,29,33, 371. All teleosts, so far examined, show a relatively large number of telencephalohypothalamic fibers traveling within the lateral forebrain bundle (LFB) and projecting to areas just dorsal and ventromedial to nucleus recessus lateralis and to regions within the inferior lobe. The presence of telencephalic efferents to the tuberal region, however, depends on the species of fish studied. An extensive projection to the tuberal region has been shown in the hime salmon (0. nerka) [33] and the bichir (Polypterus palmas) [281. In contrast, in goldfish (C. auratus) and other teleosts this projection is either lacking or extremely sparse 111, 17, 26, 29, 371. The cumulative observations on most teleosts including goldfish suggest that telencephalic connections to the hypothalamus primarily terminate in nontuberal nuclei. Recent behavioral studies in goldfish, however, suggest that the telencephalon plays a major role in courtship behavior and spawning [21,22]. Combined lesion and behavioral
projections
studies have shown that bilateral ablation of two ventral telencephalic nuclei (area ventralis pars supracommissuralis, V, and/or area ventralis pars ventralis, V,,) or a rostral preoptic nucleus (nucleus preopticus perventriculare, NPP) severely impairs courtship behavior and spawning in male goldfish [21,22]. These same nuclei as well as tuberal nuclei have been shown to bind sex steroids in several teleosts including goldfish [8, 9, 19, 201. These studies suggest more extensive connections between telencephalon including rostral NPP and hypothalamus, and prompted us to reinvestigate telencephalohypothalamic fibers in goldfish. Silver degeneration techniques were used to trace the connections between goldfish telencephalon including rostral NPP and hypothalamus. A modified Fink and Heimer procedure was used to label lesioned efferent fibers [2]. This study has shown that a significant number of telencephalic and/or rostral NPP fibers project to preoptic and tuberal regions of goldfish hypothalamus, and that these fibers more extensively innervate the inferior lobes than previously reported [29].
‘Requests for reprints should be addressed to Mark J. Airhart, Ph.D., Department of Anatomy, Box 19%OA, Quillen-Dishner Medicine, East Tennessee State University, Johnson City, TN 37614.
503
College of
AIRHART. ABBREVIATIONS
I. AC 2. BV 3. cc 4. CE 5. CM 6. CN III 7. CT 8. D, 9. DF 10. DIG
11. D, 12. FVLTrO 13. H 14. HC 15. LFB 16. LH 17. MFB 18. NAH 19. NAT 20. NCH 21. NDL 22. NE 23. NG 24. NLTI 25. NLT, 26. NPGI 27. NPG,,,
anterior commissure blood vessel corpus of the cerebellum nucleus centralis of the inferior lobe corpus mamillare cranial nerve III connective tissue area dorsalis telencephali pars centralis nucleus diffusus of the inferior lobe dorsolateral nuclear complex area dorsalis telencephali pars medialis fasciculus ventrolaterahs tractus opticus habenula horizontal commissure lateral forebrain bundle nucleus lateralis hypothalami medial forebrain bundle nucleus anterioris hypothalamicus nucleus anterior tuberis nucleus cerebellosus hypothalami nucleus dorsolateralis thalami nucleus entopeduncularis nucleus glomerulosis nucleus lateral tuberis pars lateralis nucleus lateral tuberis pars posterioris nucleus preglomerulosus pars lateralis nucleus preglomerulosus pars medialis
‘8. NPO 29. NPO,, 30. NPP 31. NPP,
32. NPT 33. NR 34. NRL 3.5. NRP 36. NT 37. NTP 38. NVM 39. OT 40. OT, 41. PC 42. SCN 43. TE 44. TL 45. I-L,, 46. V 47. VB 48. VC 49. Vd 50. vHC 51. vp 53. V, 53. v,
METHOD
Animds
Common goldfish (Carassius auratus), 7-10 cm long, were obtained from Grassyfork Fisheries (Martinville, IN) and maintained in aquaria at 22-24°C under normal daynight laboratory conditions. Surgery
Fish were anesthetized in 0.03% MS-222 (Sandoz) and surgery performed under a dissecting microscope. The telencephalon was approached through a small bone flap incised in the calvaria. The lesioning procedure consisted of gently separating the telencephalic lobes to visualize the anterior commissure and transsecting at approximately a right angle the right telencephalon at the level of the anterior commissure with a microscalpel. The microscalpel was constructed from a razor blade 1 mm in width and held by a hemostat mounted on a micromanipulator. The approximate position of the transection is shown schematically in Fig. 1. To remove the telencephalon rostra1 to the transsection, the anterior commissure was sectioned and approximately twothirds of the rostral right telencephalon was ablated by aspiration. The extent of the ablation was determined by examining alternate cresyl violet stained sections of the remaining caudal telencephalic segment. Only animals showing a completely intact nucleus preopticus (NPO) were included in this study. Silver Degeneration
Procedures
Eight experimental fish (5 males, 3 females) were anesthetized in 0.03% MS-222 and killed by intracardial perfusion with 10.0% buffered formalin (pH=7.2) four days after
SHIRK AND KRIEBEI
--
nucleus preopticus nucleus preopticus pars parvocellulari\ nucleus preopticus periventricularis nucleus posterioris periventricularis nucleus posterior tuberis nucleus rotundus nucleus recessus laterahs nucleus recessus posterioris nucleus tenia nucleus posterioris thalami nucleus ventromedialis thalami optic tectum optic tract posterior commissure suprachiasmatic nucleus telencephalon torus longitudinalis nucleus of the torus lateralis ventricle ventral brachium valvules of the cerebellum area ventralis telencephali pars dorsalis ventral branch of the horizontal commissure area ventralis telencephali pars posterioris area ventralis telencephali pars supracommissuralis area ventralis telencephali pars ventralis
surgery. Brains were embedded in 20% gelatin, sectioned transversely at 33 pm with a freezing microtome, and alternating sections were stained with cresyl violet or a modification of the Fink and Heimer procedure [2] described by Airhart and Kriebel. Briefly, this modification consisted of a reduction in the silver nitrate concentration in the presoak step (0.4%) and the ammoniacal solution (1.4%), and extending treatment with potassium permanganate (O.OS%, 25 min). These modifications resulted in a decrease in background levels of non-specific silver grains. A survival time of four days was chosen in these studies because a higher density of degenerating profiles were observed in the hypothalamus at this time in preliminary experiments which compared 2, 4, and 6 days postsurgical fish. The modified Fink and Heimer technique was chosen for these studies because preliminary experiments with HRP suggested the presence of several reciprocal connections between telencephalon and hypothaHRP is transported both lamic nuclei [ 11. Since anterogradely and retrogradely, the afferent or efferent nature of labeled fibers could not be established unambiguously unless a specific anterograde tracer technique was employed. In all experimental animals, degenerating silver impregnated fibers, preterminals, and terminals in silverstained sections were plotted with a camera-lucida. The nomenclature used to describe nuclear regions in this report was compiled from several sources [6, 23, 27, 303. In most instances, however, we used the nomenclature of Peter and Gill [30] and have specifically indicated when nuclei were described in accordance with the terminology of others. Control Studies
In two animals,
a bone flap was reflected as described
TELENCEPHALOHYPOTHALAMIC
PROJECTIONS
FIG. 1. Schematic transverse sections through the telencephalon with the most rostral section (A) at the far right. The middle section (B) represents the level of telencephalic transection. Rostral NPP (section B) and NPO (section C) are separated along the rostrocaudal axis by 300 ym, and sections A and B separated by 400 pm. The nomenclature used to describe telencephalic
nuclei is derived from Niewenhuys
above and the surrounding ventricle of the telencephalic lobe was tom, but the telencephalon itself was left intact. Four days later, the animals were perfused, and brains were processed for silver impregnated elements using the modified Fink and Heimer procedure described above. Hy~th~amic sections from control animals showed only occasional, randomly placed, silver-impregnated profiles. The density of these profiles was no greater in hypothalamic regions than elsewhere in the control brains. An additional two animals were used to determine if efferents from the olfactory bulb projected beyond the telencephalon. A bone flap was reflected in these animals and the right olfactory tract cut. The right olfactory bulb and proximal olfactory tract were aspirated from the cranial cavity in both animals. Animals were killed 4 days later by intracardial perfusion and their brains processed by a modified Fink and Heimer procedure. Animals showed no extratelencephalic olfactopetal fibers. RESULTS
In all experimental animals, the right telencephalic lobe was severed at the level of the anterior commissure and telencephalon rostral to the cut removed (Fig. 1). This lesioning procedure was adopted to remove as much of telencephalon as possible without damaging the preoptic nuclei (Fig. 1). Lesioning nucleus preopticus m~oce~ul~s and/or nucleus preopticus parvocellularis would have compromised the interpretation of results because both nuclei send axons to the pituitary and the hypothesized axon trajectory is via the tuberal region [ 14,151.
1271and Levine and Dethier [23].
Histological examination of alternate cresyl violet stained sections through the remaining caudal telencephalon of all experimental animals showed that the ablation removed large regions of both dorsal and ventral telencephalic nuclei as well as rostral NPP. The specific telencephalic nuclei ablated included a major percentage of area dorsalis pars medialis, centralis, and dorsolateral complex; and within ventral telencephalon complete ablation of area ventralis pars dorsalis, ventralis, lateralis, a large part of supracommissuralis, and the rostral part of posterioris (Fig. 1). The projection pattern of lesioned nuclei was essentially the same in all experimental animals. The results have been summarized by line diagrams depicting the topographic route taken by efferent fibers to their target nuclei (Fig. 2) and photomicrographs showing examples of the putative terminal fields. The term putative is used to qualify “terminal fields” because at the light microscopic level degenerating profiles can not be accurately distinguished as terminals, preter~n~s, or fibers. The terms ipsilateral and contralateral relate to the side of the brain associated with the lesioned telencephalon. The projection from the lesioned telencephalon was bilateral with the majority of silver impregnated degenerating profiles ipsilateral to the lesioned telencephalon (Fig. 2). The contralateral projection originated, at least in part, from lesioned fibers in the ipsilateral telenceph~on crossing through the anterior commissure. Lateral
Forebrain
Bundle (LFB)
The course and projection pattern of lesioned fibers in the
zot!
N
THIS AND FOLLOWING PAGE FIG. 2. (A-F). Line diagrams of transverse sections through the preoptic region and hypothalmus illustrating the distribution of degenerating profiles four days after lesion. The side ipsilateraf to the lesioned teIe~ceph~on is on the left. Degenerating protiiles indicated by dots may represent terminals, preterminals, or fibers cut in cross section: dashed lines indicate fiber degeneration. The sections are organized sequentially rostral (A) to caudal (E).
TELENCEPHALOHYPOTHALAMIC
507
PROJECTIONS
-
NRL
NAT
ipsilateral LFB will be described followed by a description of degeneration in the contralateral LFB. Lesioned fibers within the ipsilateral LFB coursed through the lateral region of the preoptic area and continued along the medial margin of the optic tract (Figs. 2A; and 3a,b). Degenerating profiles were observed in the suprachiasmatic nucleus (SCN) but the source of this projection was not clear (Figs. 2A,B and 3a,c); fibers may have reached the nucleus from both the medial forebrain bundle (MFB) and LFB. The SCN has been shown to receive retinofugal fibers and because of its position to the optic chiasm has been named the suprachiasmatic nucleus [6, 34, 353. At the level of the horizontal commissure (HC) degenerating fibers in the ipsilateral LFB projected ven-
trolaterally to travel along the lateral margin of the rostra1 tuberal region. A relatively small number of degenerating fibers coursed towards the pituitary stalk with a sparse number of fibers reaching the dorsal part of nucleus lateralis tuberis pars lateralis (NLTl). Fibers were not seen reaching the origin of the pituitary stalk (Fig. 2B,C). Within this same region, degenerating fibers from the LFB were observed around the dorsolateral, ventrolateral, and ventral periphery of nucleus anterior tuberis (NAT) with a few degenerating profiles projecting into the nucleus (Figs. 2B,C and 4b). Telencephalic connections with the glomerular complex of nuclei which includes: nucleus preglomerulosis lateralis (NPGl), nucleus preglomerulosis medialis (NPGm), and nu-
X)X
AIRHART.
SHIRK AND KRlt+BEl.
FIG. 3. (a) A photomicrograph at low magnification of the ipsilateral preoptic area at the level of the SCN. Degenerating profiles can be observed in the SCN and LFB. Note the absence of degenerating profiles in the optic tract (OTr) that lies directIy adjacent to the LFB. Scale bar 100 pm. (b) An enlargement of the LFB from the area in section a delineated by the black rectangle. The degenerating profiles. examples indicated by arrow. show variability in size and many profiles are out of the plane of focus. Scale bar IS em. (c) AR enlargement of the degenerating profiles found within the SCN shown in section a. Scale bar IS grn.
TELENCEPHALOHYPOTHALAMIC
509
PROJECTIONS
cleus glomerulosis (NG) were not included in this report because embryonic studies by Bergquist [4,51 suggest that these nuclei are part of the posterior tuberculum of the thalamus. We are presently studying the connections between the telencephalon and thalamus and will publish our findings in a later report. At the rostral level of nucleus recessus lateralis (NRL) degenerating profiles from the ipsilateral LFB were found in neuropil between NPG, NG, and the dorsal part of NRL; and bordering the lateral, ventral, and ventro-medial parts of NRL (Figs. 2D, 5). Degenerating profdes were also observed in NRL and the dorsal part of nucleus di@usus of the inferior lobe as described by Braford and Northcutt (Figs. 2D,E, 5). This area was delineated by Peter and Gill [30] as nucleus diffusus torus lateralis. In this same region, a relatively small number of degenerating profiles were observed in the caudal part of NAT and in the dorsolateral half of nucleus lateralis hypothalami, LH [6] (Fig. 2D). In the caudal region of the hy~th~~us (Fig. 2E,F) a large number of degenerating elements, which appeared to come from the LFB, were found in nucleus cerebellosus hypothalamicus (NCH) and all nuclei of the ipsilateral inferior lobe. Degenerating profiles were dispersed throughout the following nuclei: NCH, corpus mammillare (CM), and nucleus centralis of the inferior lobe (CE), the later nucleus was described by Braford and Northcutt [6] (Fig. 2&F). A more regional dist~bution of degenerating elements was observed in the laterally positioned part of nucleus di&sus of the inferior lobe (DF). In this region, DF contained randomly distributed putative terminals throughout most of its volume, except at its lateral and ventrolateral periphery where in transverse section a region approximately one-third of its area was delete of silver impregnated profiles (Fig. 2E,F). Telenceph~ic efferent fibers in the contralateral LFB followed the same rostrocaudal route through the brain but a small difference in the density of degenerating profiles was noted between the two pathways, the contralateral LFB contained a lower density of degenerating fibers. Differences in the relative number of degenerating profiles were also observed between corresponding ipsilateral and contralateral nuclei (Fig. 2A-F). In the preoptic region there was a slightly lower density of degenerating profiles in the contralateral SCN; and in the tuberal region no degenerating profiles were observed in contralateral NLTl or NAT (Fig. 2B,C). Further caudally degenerating profiles were also absent from contralateral LH (Fig. 2D). Contralateral nuclei of the caudal hy~thalamus including: NCH, NRL, CM, CE, and DF demonstrated the same distribution of lab&d profiles but within each nucleus the density of labeled profiles was slightly lower than the corresponding ipsilateral nuclei (Figs. 2D,E,F and 6a,b). Median Forebrain
FIG. 4. (a) Degenerating profiles (arrow) within a fiber pathway o~~nating from the MFB and located just medial to NAT. The photograph is oriented such that medial is to the left and dorsal at the top. The complete extent of this pathway is shown in section 2C. Scale bar 15 pm. (b) A photomicrograph of NAT showing a relatively small number of degenerating profiles (arrows). Medial is to the left and dorsal at the top. Scale bar 15 Frn.
Bundle (MFB)
In the preoptic region, both ipsilateral and contralateral MFB’s contained degenerating fibers primarily located along their ventral borders. At the level of the SCN degenerating profiles from both MFB’s were observed projecting ventrally along the lateral borders of NPO with a few profiles observed within parts of NPO bitaterally (Fig. 2A,B). Degenerating fibers within ventral projections of both MFB’s appeared to terminate in SC nuclei with the ipsilateral SCN containing a slightly larger density of degenerating elements (Fig. 2A,B). Further caudally, degenerating profiles branching from both MFB’s projected ventrally along the lateral border of nucleus preopticus periventriculare (NPP,) with some
AIRHART.
SHIRK AND KRIEBEL
FIG. S. Photomicrograph of rostra1 NRL oriented such that lateral is to the right and dorsal at the top. Degenerating observed within the dorsal and lateral parts of NRL as well as around its dorsal, lateral, and ventromedial periphery. RL. contains a film of gelatin with silver precipitate on its surface. Scale bar 50 wrn.
profiles observed within the nucleus bilaterally (Figs. 2C, and 7a,b). Ipsilateral fibers projected further ventrally coursing along the medial border of NAT (Fig. 4a) with some degenerating profiles observed within the medial region of NAT (Figs. 2C and 4a). The projection from the telencephalon and/or NPP through the MFB’s appeared to end at the level of rostra1 NRL (Fig. 2D).
DISCUSSION
Our results strongly suggest the presence of new connections between ablated telencephalon and/or rostra1 NPP and preoptico-hypothalamic nuclei. These connections include efferents projecting bilaterally to NPO, SCN, NPP,,, NCH. and unilaterally to NAT, and LH. In addition, our results also show a much larger projection to the ipsilateral inferior lobe nuclei [ 11,291 and a previously unreported projection to contralateral inferior lobe nuclei (DF, CE, and CM). These results receive further support from a preliminary HRP study in which HRP was injected into a telencephalic lobe at the level of the anterior commissure in goldfish [ 11. The projecFOLLOWING
profiles (arrows) can be Part of recessus lateralis.
tion pattern of labeled fibers and nuclei containing labeled elements completely corresponded to the data presented in this paper. Because the ablated area includes several telencephalic nuclei in area dorsalis and ventralis and NPP neurons, we were unable to make conclusions regarding what specific nuclei were responsible for individual projections. Further studies involving more selective lesions are required as well as tracer studies from specific terminal fields. Three items effecting the interpretation of our results require further discussion. The first concerns the validity of the modified Fink and Heimer procedure. This procedure was tested in an earlier investigation examining telencephalotectal efferents by comparing the results of electron microscopic degeneration studies and the silver degeneration techniques [2]. This study showed a positive correlation between the presence of silver impregnated degenerating profiles in a tectal lamina and the presence of degenerating axons or terminals in the same lamina. A second item concerns the bilaterality of the projection. Correct interpretation of the bilaterality of the projection requires considering
PAGE
FIG. 6. (a) Photomicrograph of the contralateral inferior lobe showing degenerating profiles in CM, CE, and DF. Due to the low magnification, degenerating elements in NRL are not visible. Scale bar 100 pm. (b) A higher magnification of degenerating profiles in CE. Note the small punctate nature of the degenerating profiles suggesting the presence of terminal fields. Scale bar 50 km.
TELENCEPHALOHYPOTHALAMIC
PROJECTIONS
511
AIRHART.
FIG. 7. (a) Photomicrograph of rostra1 NPP,. ipsilateral to the lesion. showing the presence of a few degenerating profiles within the nucleus (arrow) and a larger density of degenerating elements along its lateral border. Scale bar I5 Wm. (b) Photomicrograph of both ipsilateral and contralateral NPP,, at a level caudal to section a, the ipsilateral side is on the left. Note the presence of degenerating profiles in both nuclei (arrows). Scale bar 15 pm.
whether the contralateral telencephalon was inadvertantly damaged during the lesioning procedure. In each experimental animal, the telencephalic lobe contralateral to the lesion was examined histologically for potential damage; no obvious indicators such as gliosis or cellular disorganization were observed. A third consideration is the possibility of confusing olfactory and telencephalic efferents. We addressed this problem by determining if olfactory efferents projected to targets outside the telencephalon. No bulbopetal fibers in extra-telencephalic regions were observed. This observation is in agreement with the findings of Oka and Ueda [29], but not with two recent HRP studies that traced olfactory bulb connections in goldfish. Both investigations showed a small number of bulbopetal fibers from one olfactory bulb projecting bilaterally through the MFB’s. These fibers passed through or terminated in NPP and an area just rostra1 to nucleus posterior tuberis designated as either the caudal olfactory nucleus or tuberal terminal field [3,23]. One
SHlRK
AND
KRIEBEt.
of these reports also eluded to the presence of a small number of HRP-labeled fibers in the vicinity of NLTI (231. We assume that these projections were not observed in our study because the survival time was not appropriate to detect argyrophilia in these extratelencephalic fibers. However, because of the discrepancy between our control study and the HRP studies, we have not drawn conclusions regarding projections which overlapped with the known olfactory bulb connections to NLTI. There have been only three studies examining telencephalic projections in goldfish. One of these studies by Oka and Ueda [29] used silver degeneration techniques. ‘Their results showed no telencephalic projection to preoptic or tuberal nuclei and a relatively small projection to the ipsilatera1 inferior lobe; nearly all telencephalohypothalamic efferents were observed in neuropil. The differences in results between the work of Oka and Ueda [29] and the present findings may be due to differences in survival times. Although Oka and Ueda used goldfish of approximately the same size as in the present investigation, the survival times were longer, i.e., 612 days. Previous silver degeneration studies have shown that the rate of phagocytosis and loss of argyrophilia is slower for axons than associated preterminals and terminals. Therefore, the longer survival times used by Oka and Ueda may have selected for axons and resulted in complete phagocytosis of degenerating terminals in hypothalamic nuclei. A second possible explanation is that our modification of the Fink and Heimer technique is more sensitive than the technique used by Oka and Ueda (Ebbesson’s rnodification of the Fink and Heimer method) [IO). Our results also differ from two previous HRP studies that traced connections between telencephalon and hypothalamus in C‘. UUIY~~U.) [I I] and C‘. c trrcr.stirt.t and C’xpr-iuu.v cwpio 1171. Both studies focussed on connections between telencephalon and thalamus, however, all putative efferent fibers and nuclei containing labeled cells were reported. Both studies showed labeled fibers in the ipsilateral LFB projecting to the medial part of the inferior lobe. No labeled fibers were observed in SCN or the tuberal region of the hypothalamus. The absence of such fibers may have resulted from the location of injection sites. In both studies, HRP was injected into only dorsal telencephalic nuclei. In the hime salmon (0. nc~&rr) ventral telencephalic nuclei were lesioned (V,. V,) [ 331 and silver degeneration techniques demonstrated an extensive bilateral projection to tuberal nuclei. Degenerating fibers and/or terminals were observed bilaterally in NLTI, nucleus lateralis tuberis. pars medialis (NLTm). LH, and NAT with a majority of degenerating profiles on the ipsilateral side. A similar projection was observed in the birchit-, P. &V(I.P [28]. Large HRP injections, including both dorsal and ventral telencephalic nuclei, labeled fibers projecting bilaterally to the tuberal region. These results support our findings and emphasize the potential importance of ventral telencephalic nuclei in modulating hypothalamic function. The control of circadian rhythms and secretion of pituitary hormones may involve several preoptic-hypothalamic nuclei that receive afferents from lesioned telencephalic nuclei and/or nPP. A nucleus homologous to the suprachiasmatic nucleus in mammals has been shown in goldfish. This nucleus receives projections from both eyes with the majority of fibers coming from the contralateral retina [34], as it is in most other species [2S.32]. In addition to retinal input, the present study has shown that SCN also receives a bilateral projection from each telencephalic lobe perhaps including NPP. with the majority of afferent fibers projecting to the
TELENCEPHALOHYPOTHALAMIC
513
PROJECTIONS
ipsilateral SCN. Such a projection has not been reported in other teleost species but has been observed in some mammals [24,32,36]. To suggest a function for these fibers would be premature since the physiology of SCN in teleosts is speculative [ 161. A number of reports using both immunohistochemical and biochemical techniques have suggested that a gonadotropin release-inhibitory factor, thought to be dopamine, is found in NPP neurons, and that these neurons project via the anterolateral tuberal region to the pituitary [7, 18, 311. Although the lesioned area in this study included rostra1 NPP, we did not observe degenerating profiles reaching the pituitary stalk region. Perhaps not enough of NPP was lesioned and/or the labeling techniques were not sensitive enough to detect the proposed fibers. Our results, however, have shown that at least some hypothalamic
nuclei that project
axons
to
the pituitary receive telencephahc and/or NPP afferents. These nuclei include NPP, and NRL [13,14]. Nucleus posterioris periventriculare and NRL compose part of the paraventricular organ of goldfish and both contain large ~pulations of ~atecholamine confining neurons projecting to all regions of the goldfish pituitary 1131. The presence of degenerating profiles in NPP, and NRL suggests that lesioned areas of telencephalon and/or NPP may modulate the activity of these nuclei and indirectly effect pituitary function.
ACKNOWLEDGEMENTS
The authors thank Cindy Canter for typing the manusc~pt. This work was supported by an ETSU Institutional Grant to M.J.A. and PHS 5429-16-19 to R.M.K.
REFERENCES 1. Airhart, M. J. Axonal connections between telencephalon and hypothalamus in goldfish. Anat Ret 218: SA, 1987. 2. Airhart, M. J. and R. M. Kriebel. Telencephalic terminals in the major retinal synaptic lamina of the goldfish optic tectum. Brain Res 336: 363-367,
1985.
3. Bartheld, C. S. von, D. L. Meyer, E. Fiebig and S. 0. E. Ebbesson. Central connections of the olfactory bulb in the goldfish, Carassius auratus. Cell Tissue Res 238: 475-487, 1984. 4. Bergquist, H. Zur Morphologie des Zwischenhims bei niederen Wirbeltieren. Acto Zoo1 13: 57-304, 1932. 5. Bergquist, H. Ontogenesis of diencephalic nuclei in vertebrates. A comparative study. Lunds ArssKr Aud 2 50: I-34, 1954. 6. Bradford, M. R. and R. G. Northcutt. Or~nization of the diencephalon and pretectum of the ray-finned fishes. In: Fish Neurobiology, ~01 II, edited by R. E. Davis and R. G. Northcutt. Ann Arbor: University of Michigan press, 1983, pp. 117163. 7. Chang, J. P. and R. E. Peter. Effects of dopamine on gonado. . .r . .._. tropm release m tern&e goldtish, carassms auratus. Neuroendocrinology 36: 351-357, 1983. 8. Davis, R. E., J. I. Morel1 and D. W. Pfaff. Autoradiographic localization of sex steroid-concentrating cells in the brain of the teleost Macrop~us opercularis (Oste~chthyes: Belontiidae). Gen Camp Endocrinol33: 496-505, 1977. 9. Demski, L. S. Neuroanatomical substrates of reproductive behavior in male sunfish (Genus Lepomis). Ann Biol Anim Biochime Biophys 18: 831-836,
1979.
10. Ebbesson, S. 0. E. The selective impregnation of degenerating axons and their synaptic endings in nonmammalian soecies. In: Contemporary Research Methids in Neurcranatomy,‘edited by W. J. H. Nauta and S. 0. E. Ebbesson. Berlin: Springer-Verlag, 1970, pp. 132-161. 11. Echteler, S. M. and W. M. Saidel. Forebrain connections in the goldfish support telencephalic homologies with land vertebrates. Science 212: 683-685, 1981. 12. Fink, R. P. and L. Heimer. Two methods for selective silver impregnation for degenerating axons and the synaptic endings in the central nervous system. Brain Res 4: 369-374, 1967. 13. Fryer, J. N., C. Boudreault-Chateauve~ and R. P. Kirby. Pituitary aBerents originating in the paraventricular organ (PVO) of the goldfish hypothalalmus. .I Comp Neurol242: 475-484, 1985. 14. Fryer, J. N. and L. Maler. Hypophysiotropic neurons in the gold&h hy~t~amus demonst~ted by retrograde transport of horseradish peroxidase. Cell Tissue Res 218: 93-102, 1981. 15. Gregory, W. A. and C. D. Tweedle. Horseradish peroxidase evidence for a spinal projection from the preoptic area of the goldfish, a light and electron microscopic study. Brain Res 341: 82-91,
1985.
16. Hontela, A. and R. E. Peter. Effects of pinealectomy, blinding and sexual conditions on serum gonadotropin levels in the goldfish. Gen Camp ~ffdu(.rin~~i 40: 168-179, 1980. 17. Ito, H., T. Murakami and Y. Morita. An indirect telencephalocerebellar pathway and its relay nucleus in teleosts. Brain RL’S 249: I-13, 1982. 18. Kah, O., P. Chambolle,
of dopaminergic
J. Thibault and M. Geffard. Existence neurons in the preoptic region of the goldfish.
Neurosci Lett 48: 293-298. 1984. 19. Kim, Y. S., W. E. Stumpf and M. Sar. Topography
20.
21.
22.
23.
of estrogen target cells in the forebrain of goldfish, Carassius auratus. J Camp Nrurol 182: 61 l-620, 1978. Kim, Y. S., W. E. Stumpf, M. Sar and M. C. Ma~ine2-Vargas. Estrogen and androgen target ceIIs in the brain of fishes, reptiles and birds: Phylogeny and ontogeny. Am Zoot 18: 425-433,1978. Koyama, Y., M. Satou, Y. Oka and K. Ueda. Involvement of the telencephalic hemispheres and the preoptic area in sexual behavior of the male goldfish, Carassius auratus: a brain-lesion study. Behav Neural Bioi 40: 70-86, 1984. Kyle, A. L. and R. E. Peter. Effects of forebrain lesions on spawning behavior in the male goldfish. Phvsiol Behav 28: 1103-l 109, 1982. Levine, R. L. and S. Dethier. The connections between the olfactory bulb and the brain in the aoldfish. .I Coma Neural 237:
427-444; 1985. 24. Meiback, R. C. and A. Siegel. Efferent connections of the hippocampal formation in the rat. Bruin Res 124: 197-224, 1977.
25. Moore, R. Y. Central neural control of circadian rhythms. In: Frontiers in Newoendocrinology, vol 5, edited by W. F. Ganong and L. Martini. New York: Raven Press, 1978, pp. 185-206. 26. Murakami, T., Y. Morita and H. Ito. Extrinsic and intrinsic fiber connections of the telenceph~on in a teleost, Sebastiscus marmoratus. f Camp Neurol 216: 115-131, 1983. 27. Nieuwenhuys, R. The comparative anatomy of the actinopterygian forebrain. J Hirnforsch 6: 171-192, 1963. 28. Northcutt, R. G. Localization of neurons afferent to the telencephalon in a primitive bony fish, Polypterus palmas. Neurosci Lett 22: 219-222, 1981. 29. Oka, Y. and K. Ueda. Telencephalic
projections in the goldfish (Carassius auratus): an anterograde degeneration study. / Fuc Sci Univ Tokyo 15: 1-8, 1981. 30. Peter, R. E. and V. E. Gill. A stereotaxic atlas and technique for forebrain nuclei of the goldfish, Carassius auratus. J Camp Neural 159: 69-102, 1975. 31. Peter, R. E. and C. R. Paulencu. Involvement of the preoptic region in gonadotropin release-inhibition in goldfish, Carassius au&us. Ne~4roendo~.r~noiogy 31: 133-141, 1980.
AIRHAR’I‘.
iI
32. Pickard. nucleus
Ci. E. The afferent connections of the golden hamster with
of the suprachiasmatic emphasis on the ret-
inohypothalamic projection. .I C‘ojtrp .Y~lrr~j/ 211: C-83. 1982. 33. Shiga, ‘T.. Y. Oka. M. Satou, N. Okumoto and I<. Ueda. Effercnts from the supracommissural ventral telencephalon in the Hime Salmon (landlocked red salmon. Oncorhynchus nerka): An anterograde degeneration study. Hrtrirl KC,.\ H/i// 14: 55-61. 19x.
SHIRK
AND
KRIkBEL.
34. Springer. A. D. and .I. S. Gaffney. Retinal projections 111the goldfish: 3 study using cobaltous-lysine, .! ( ~J,J,,I ~\‘~,~frr~/203: 401-424. 1981. 35. Springer. A. D. and A. S. Mednick. Selective Innervation ofthe goldfish suprachiasmatic nucleus by ventral retinal ganglion cell axons. Brtri/, KC,.\ 323: 293-796. 1984. 36. Swamon. L. W. and W. M. Cowen. The connections of the scptal region in the rat. .I C‘CJ!,I/>,l’~,rlro/ 186: Q-656, 1979. 37. Vanegas. H. and S. 0. E. Ebbesson. Telencephalic projections in two teleost species. .I ( ‘~j/fr/> h’c,rlr~~/ 165: IXI-196. 1976.