| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Article |
Department of Pharmacology and Experimental Therapeutics, Loyola University Chicago, Strich School of Medicine (B.D.), Maywood, Illinois 60153; and Women’s Health Research Institute, Wyeth Research (I.M.), Collegeville, Pennsylvania 19426
Address all correspondence and requests for reprints to: Dr. Istvan Merchenthaler, Women’s Health Research Institute, Wyeth Research, 500 Arcola Road, Collegeville, Pennsylvania 19426. E-mail: merchei{at}wyeth.com.
Abstract
Gonadal functions are modulated by corticotropin-releasing factor (CRF) in the rat via direct suppression of LH-releasing hormone (LHRH) release. Although there is evidence of direct morphological contacts between the LHRH and CRF-immunoreactive (-IR) structures in the rat hypothalamus, little is known about the morphological base of CRF-influenced LHRH release in man. Thus, we studied the distribution of the CRF-IR and LHRH-IR systems in the human diencephalon and revealed putative CRF-LHRH juxtapositions using double label immunohistochemistry.
LHRH-IR cells were present mainly in the infundibular region and the medial preoptic area. CRF-IR neuronal structures were observed in the periventricular area, paraventricular nucleus, infundibular region, and median eminence. CRF-LHRH juxtapositions were found mainly in the infundibulum and median eminence. Few juxtapositions were detected in the medial preoptic area. In these regions, black diaminobenzidine/silver-labeled CRF-IR fibers abutted fusiform brown diaminobenzidine-labeled LHRH neurons, usually forming multiple contacts. Examination of semithin sections of these close associations with the aid of oil immersion revealed no cleft between CRF-IR nerve terminals contacting LHRH-IR structures.
These findings suggest that the juxtapositions between the LHRH-IR and CRF-IR neurons may be functional synapses forming the morphological substrate of the CRF-controlled LHRH secretion. Moreover, the wide distribution of CRF-IR elements suggests that CRF controls other diencephalic functions as well.
GONADAL FUNCTIONS ARE suppressed by prolonged stress (1, 2, 3, 4, 5). Although it is generally believed that this phenomenon involves the inhibition of LH release by catecholamines (6, 7, 8, 9, 10), there is growing evidence that corticotropin-releasing factor (CRF) plays a pivotal role in the stress-related suppression of reproductive functions (11, 12, 13, 14, 15, 16, 17).
Gonadal functions can be influenced at two different levels: 1) via direct modulation of LH secretion by the pituitary, and/or 2) via regulation of LH-releasing hormone (LHRH)-stimulated LH release. Although CRF directly influences LH release in rat and human pituitary cells in vitro (18, 19), there is a general consensus that the effect of CRF on gonadal functions is based on CRF-mediated inhibition of LH release via suppression of hypothalamic LHRH secretion. Administration of CRF into the third ventricle blocks the preovulatory LHRH-mediated LH surge in the ewe (20). CRF infusion into the medial preoptic area or into the median eminence of the rat inhibits LH secretion by altering LHRH release (21, 22, 23). CRF inhibits LHRH release from superfused rat hypothalami in vitro (24). Moreover, CRF directly suppresses LHRH transcription in cell culture (25).
Although several studies indicated that endogenous opiate peptides mediate the regulation of gonadal functions by CRF (18, 26, 27), there is growing evidence that CRF may directly regulate LHRH release in the hypothalamus. In rats, the inhibition of LH secretion elicited by intracerebroventricular infusion of CRF is not influenced by administration of the opiate agonist naltrexone (11), and suppression of LHRH release by CRF can also be observed in the presence of opiate antagonists (16). Moreover, synaptic connections have been found between CRF and LHRH-immunoreactive (-IR) elements in the medial preoptic area of the rat hypothalamus (28). In contrast to these findings, little is known about the role of the CRF in the regulation of reproductive functions in humans.
CRF-IR structures have been found in the rat (29, 30) and human brain (31, 32, 33, 34, 35, 36, 37, 38); however, the CRF system has not yet been precisely mapped in areas containing LHRH-IR elements in the human. In contrast, the morphology and location of LHRH-IR elements are well known in the human diencephalon (39, 40, 41, 42, 43, 44, 45, 46). Because the interaction between CRF-IR and LHRH-IR elements can be the morphological substrate of the inhibition of LHRH release by CRF, it is crucial to reveal the distribution of CRF-IR elements and examine their putative juxtapositions with LHRH-IR structures in the human diencephalon.
In the present study we studied the morphology and distribution of CRF-IR and LHRH-IR elements in the human diencephalon. First, the sites of overlap between the two systems were extensively searched to reveal the juxtapositions between CRF-IR and LHRH-IR structures; then these putative synapses were quantitated according to their location and morphology. Because the long postmortem time of the human samples made the application of electron microscopic evaluation virtually impossible, semithin sections were used to study the fine structure of the putative CRF-LHRH synapses at the light microscopic level.
Materials and Methods
Brain samples
Brain samples from three women and two men (28–81 yr of age) were obtained from autopsies at 24–48 h postmortem. The clinical and pathological records of the individuals did not contain any neurological or neuroendocrinological diseases. The brain samples were taken in accordance with the regulation and permission of the ethics board of Szent-Györgyi Albert Medical University.
Fixation and section preparation
Hypothalami were dissected from the brains and fixed by immersion in 0.1 M phosphate-buffered (pH 7.4) 4% formaldehyde at 4 C for 2–24 wk. Each block contained half of the hypothalamus split at the midsagittal line. The samples were cryoprotected with 30% sucrose in phosphate buffer containing 0.9% sodium chloride (PBS) supplemented with 0.1% sodium-azide. Coronal sections (30 µm) were cut on a freezing microtome, collected in four series of wells of plastic 24-compartment plates with PBS containing 0.2% sodium azide, and stored at 4 C until processing. The adjacent sections were processed as follows: 1) Nissl staining with cresyl-violet, 2) immunohistochemical detection of LHRH, 3) immunohistochemical detection of CRF, and 4) double label immunohistochemistry of CRF and LHRH.
Immunohistochemistry
Single labeling was carried out using either the peroxidase-antiperoxidase or streptavidin-biotin methods combined with silver intensification introduced by Gallyas and co-workers (47). The samples were pretreated first with 10% thioglycolic acid for 30 min to suppress the endogenous tissue argentophilia, then with 0.2% Triton X-100 for 20 min, followed by 10% normal horse serum in PBS for 1 h at room temperature to block nonspecific staining. Thereafter, the sections were incubated in a primary antiserum solution composed of 10% normal horse serum, PBS, and 0.1% sodium azide for 24 h. The LHRH-containing structures were identified using a monoclonal mouse anti-LHRH antiserum (Chemicon, Temecula, CA) at a dilution of 1:20,000. The CRF-IR structures were visualized by rabbit anti-CRF serum (Peninsula Laboratories, Inc., Belmont, CA; 1:15,000). The sections were then incubated in a secondary antiserum containing unlabeled goat antirabbit IgG (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA; at a dilution of 1:50) to detect CRF or in biotin-labeled horse antimouse IgG (Vector Laboratories, Inc., Burlingame, CA; 1:500) for 2 h at room temperature to detect LHRH. In each case, the cross-reactions of the secondary antibodies with the endogenous human IgGs were eliminated by the addition of 2% normal human serum to these reagents. Finally, the sections were incubated for 2 h at room temperature in rabbit peroxidase-antiperoxidase (Jackson ImmunoResearch Laboratories, Inc.; at a dilution of 1:2,000) for the detection of CRF or in peroxidase-labeled streptavidin (Jackson ImmunoResearch Laboratories, Inc.; at a dilution of 1:500) for the detection of LHRH. The chromogen solution was composed of 0.05% diaminobenzidine (DAB), 0.125% nickel-ammonium sulfate, and 0.005% hydrogen peroxide in 0.1 M Tris-HCl (pH 7.6). The resulting DAB polymer was silver-intensified in a solution containing 0.1% silver nitrate, 0.1% ammonium nitrate, 1% silicotungstic acid, and 0.2% formaldehyde for 3–4 min (47).
Simultaneous detection of CRF-IR and LHRH-IR structures was performed using double label immunohistochemistry. First, CRF immunohistochemistry was carried out, then the LHRH-containing neuronal structures were immunolabeled as described above, but using DAB chromogen without nickel ions. The anti-LHRH serum was used in a 1:8000 dilution.
Some of the double-labeled sections were embedded in plastic to obtain thin (1-µm) sections. The large frozen sections were cut into smaller pieces, processed through graded ethanol series and propylene oxide, and then flat-embedded in Durcupan (Sigma, St. Louis, MO). Semithin sections were cut with an ultramicrotome.
In control sections, the primary antibodies were omitted or replaced by nonimmune rabbit serum at a dilution of 1:15,000 (CRF) or 1:20,000 (LHRH), respectively. Immunoreaction was not observed in either case.
Computer-assisted mapping
The hypothalamic sections were systematically scanned using a plain scanner (Hewlett-Packard Co., Palo Alto, CA), and the outlines of the sections were traced using CorelTrace software 4.0 (Corel Corp., Ottawa, Ontario, Canada). The computer-generated superimposition of eight consecutive sections was summarized in each slide in Fig. 1
, denoted by letters A–P. The neurons and fibers were marked on these figures using an Axiophot microscope with camera lucida (Carl Zeiss, New York, NY), and Adobe PhotoShop software (version 3.0; San Jose, CA).
|
The terminology of the diencephalic structures was adapted from Braak and Braak (48), Saper (49), and Silverman et al. (43).
Results
LHRH system
The distribution and morphology of the LHRH-IR elements were in good agreementment with our previous studies (44, 45, 46). The majority of the perikarya (89%) were concentrated in the medial 4-mm zone of the human diencephalon, mainly in the medial preoptic area, infundibular regions, and median eminence. Few perikarya were found along the diagonal band of Broca, mammillary nuclei, and the caudal part of the periventricular areas. Only 11% of the LHRH-IR cell bodies were located in the lateral 4- to 16-mm area of the human diencephalon, mainly in the supraoptic nuclei and the lateral zone of the mammillary region. Immunoreactive fibers were located in the diagonal band of Broca and periventricular and infundibular areas. The pattern and morphology of the LHRH-IR structures did not show any significant differences regarding the age, sex, and postmortem time of the samples.
CRF system
The vast majority of CRF-IR perikarya (97%) were located in the medial 4-mm zone of the human diencephalon. Frontally to the lamina terminalis, the CRF-IR cells were concentrated in the superior part of the diencephalon around the fornix and laterally around the anterior commissure (Fig. 1
, A–C). In the preoptic area the perikarya were periventricularly located spreading from the fornix to the optic chiasm (Fig. 1, E and F). In the tuberal region, the CRF-IR cell bodies were concentrated in four areas: 1) the majority of the cells were located in the paraventricular nuclei (Fig. 1, G–I), 2) cells were found in the periventricular region, surrounding the third ventricle from above the fornix to the infundibulum (Fig. 1, G–L), 3) a well defined group of CRF-IR cells was found in the dorsomedial subdivision of the ventromedial nucleus (Fig. 1, K and L), and 4) a few immunoreactive cells were located in the infundibular region (Fig. 1, H–J). The supraoptic nuclei did not contain any CRF-IR cell bodies (Fig. 1, F–I). In the posterior hypothalamus, CRF-IR perikarya were found in the periventricular region and around the mammillary body (Fig. 1, M–P). The majority of the CRF-IR neurons were fusiform in shape, with two to four presses emanating from the opposite poles. Numerous star-shaped multipolar cells were found in the periventricular region.
CRF-IR fibers were organized in several well defined pathways in the human diencephalon. Fibers originating from the infundibulum ran laterally, close to the basal surface of the hypothalamus (Fig. 1, I–L). This tract arched over the optic tract dorsally. The CRF-IR fibers arising from the paraventricular nucleus (PVN) followed three different directions: 1) a well defined tract ran horizontally from the PVN through the substantia innominata of the lateral hypothalamus (Fig. 1, H–K); 2) fibers projecting to the base of the diencephalon surrounded the fornix laterally (Fig. 1, H–J); and 3) this tract crossed the infundibular pathway in the tuberal region of the hypothalamus, medially to the optic tract. Axons from the PVN surrounded the fornix medially and projected toward the infundibulum/medial eminence (Fig. 1, H–J). An additional set of CRF-IR fibers was detected periventricularly. These fibers appeared to arise from the periventricularly arranged CRF-IR perikarya (Fig. 1, H–K). Finally, delicate fiber network was observed around the anterior commissure (Fig. 1, B–E) and the mammillary body (Fig. 1, M–P). The CRF-IR fibers formed dense network around the portal vessels in the tuberal region (Fig. 1, I–K, and Fig. 2D), and were often associated with small blood vessels in the PVN and the periventricular region (Fig. 2A). The examined diencephalon did not show any significant differences in the distribution and morphology of the CRF-IR elements related to the age and sex of the individuals.
|
Superimposition of the CRF and LHRH maps revealed that these systems overlapped mainly in the infundibular region, median eminence, and medial preoptic area, whereas partial overlap was detected in the periventricular area of the tuberal region and in the basal part of the lateral preoptic area. Examination of the overlapping areas revealed black CRF-IR fibers juxtaposing brown LHRH-IR neurons (Fig. 3). Characteristically, the CRF-IR fiber varicosities abutted fusiform LHRH-IR cells forming multiple axosomatic and axodendritic contacts as they passed along the neuron. Close examination of these CRF-LHRH associations, using semithin sections of Durcupan-embedded double-labeled sections examined with oil immersion, revealed no traces of gap between the juxtaposing elements.
|
, E, F, I, J, and K). Here, 22% of the LHRH-IR cells were heavily innervated by CRF-IR fibers receiving three to five contacts per cell. Additionally, 42% of the LHRH-IR neurons were lightly innervated by the CRF system (one to three contacts), and only 36% of the LHRH-IR perikarya were detected without any associated CRF-IR elements in these regions. The CRF-LHRH juxtapositions were less numerous in the preoptic region of the hypothalamus (11%), located periventricularly in the medial preoptic area (Fig. 3H) and in the anteroventral periventricular nucleus (Fig. 3, A and G). An additional population of juxtapositions was detected in the basal part of the lateral hypothalamic region at the medial side of the optic tract (7%; Fig. 3, C and D). No contacts were detected between CRF-IR fibers and LHRH-IR cells in the rostral part of the diencephalon, anteriorly to the lamina terminalis. Few juxtapositions were observed between CRF-IR and LHRH-IR fibers. In some cases, CRF-IR and LHRH-IR axons targeted the same blood vessel in the periventricular area, forming multiple contacts around the vessel wall (Fig. 2A). All of the examined brains showed the similar pattern of the CRF-LHRH associations regardless of age, sex and postmortem time.
Juxtapositions between CRF-IR neurons
CRF-IR elements often received juxtaposing CRF-IR fibers in the paraventricular nucleus and periventricular area. These juxtapositions were mainly axosomatic contacts. The CRF-IR axons targeted both fusiform (Fig. 2C) and multipolar (Fig. 2B) CRF-IR perikarya.
Discussion
A body of evidence is accumulating that gonadal functions are controlled by CRF in several species, including ewe, rat, monkey, and human (14, 17, 20, 21, 22, 23). Although numerous studies suggest that other neurotransmitters are involved in the CRF-influenced LH regulation (18, 24, 27), there is growing evidence that CRF may directly regulate gonadal physiology via a direct synaptic mechanism between the CRF and LHRH systems (11, 15, 28). The present report is the first study to visualize close juxtapositions between the LHRH and CRF-IR elements in humans, which may be the morphological substrate of CRF-controlled LHRH/LH release. Additionally, we precisely mapped and characterized the LHRH and CRF systems in human diencephalon.
Our findings regarding the morphology and distribution of the LHRH system were in good agreement with our previous observations (44, 45, 46). Fusiform LHRH-IR perikarya were located in the periventricular and infundibular regions and in the median eminence. Small numbers of LHRH cells were detected along the diagonal band of Broca and around the mammillary bodies. LHRH-IR fibers ran periventricularly and along the base of the hypothalamus.
In our studies the vast majority of CRF-IR perikarya were found in the PVN. This finding is in good agreement with previous reports (31, 33, 34, 36), describing CRF-IR cell bodies in the PVN, the periventricular area, and the infundibular region. We detected an additional CRF-IR cell group located rostrally to the lamina terminalis and dorsally to the anterior commissure. These cells may be the source of the CRF-IR fibers surrounding the anterior commissure.
Previous studies described a dense, CRF-IR fiber network projecting from the paraventricular nucleus toward the infundibulum and the median eminence (31, 34). In addition, we detected a well defined set of CRF-IR fibers, seemingly originating from the paraventricular nucleus and projecting mediolaterally, through the hypothalamus. These fiber varicosities appear to target areas outside of the hypothalamus.
The infundibular region and the median eminence are targeted by CRF-IR fibers from several different directions. Fibers running periventricularly seem to originate from the periventricularly arranged perikarya, whereas the CRF-IR axon varicosities arching around the fornix arise from the paraventricular nucleus. The source of a delicate set of CRF-IR fibers running mediolaterally at the basal surface of the diencephalon might be extrahypothalamic; this pathway crosses the set of axons projecting from the paraventricular nucleus and extends toward the lateral hypothalamic area. CRF-IR fibers were also observed forming a dense network around the portal vessels of the tuberal region. In addition to these findings, CRF-IR fiber varicosities were detected in and around the walls of blood vessels in the paraventricular nucleus and periventricular area, suggesting that CRF controls distant diencephalic functions not only directly, but via the bloodstream as well.
Microinjection of CRF into the third ventricle attenuated LHRH-mediated LH release in ewe (20). Similar findings were observed when CRF was injected into the medial preoptic area and the median eminence of the rat (21, 22, 23). These findings suggest that CRF control of the LHRH system could be localized periventricularly in these areas. Although synaptic contacts were found between the CRF and LHRH systems in the medial preoptic area of the rat, the existence of these close associations in humans has not yet been elucidated. The present study is the first that describes juxtapositions between LHRH and CRF systems in man and analyzes their morphology and pattern in the human diencephalon. Superimposition of the CRF and LHRH maps demonstrated that these systems overlap mainly in the infundibulum/median eminence and the medial preoptic area. Indeed, close examination of these regions revealed that the contacts between the CRF-IR fibers and LHRH-IR perikarya were most numerous in these regions. Because no gap has been detected between the juxtaposing LHRH-IR and CRF-IR elements, these intimate contacts may be functional synapses.
Our previous studies have shown that LHRH-IR perikarya juxtapose with neuropeptide Y (NPY)-IR, catecholaminergic, and substance P (SP)-IR elements mainly in the tuberal region and the medial preoptic area (44, 45, 46). The similar patterns of the described CRF-LHRH juxtapositions suggest that these neurotransmitters might colocalize and interact in these areas and thereby modify LHRH release at the level of the same neuron. Moreover, an additional subset of LHRH perikarya receiving CRF-IR fibers was observed in the basal part of the lateral preoptic area. Here, the LHRH-IR cell bodies appear to form intimate associations not only with CRF-IR fiber varicosities projecting from the paraventricular nucleus, but also with CRF-IR axons extending from the base of the diencephalon. Because the CRF-IR fibers running at the basal surface of the hypothalamus and seemingly projecting to the infundibulum and median eminence might have an extrahypothalamic origin, the CRF-LHRH juxtapositions in this area raise the possibility of extrahypothalamic control of hypothalamic LHRH release.
CRF-IR elements often received juxtaposing CRF-IR fibers in the PVN and periventricular area. These juxtapositions were mainly axosomatic contacts and may function as real synapses suggesting CRF-controlled CRF release in the hypothalamus.
In conclusion, the juxtapositions between the LHRH-IR and CRF-IR elements may represent functional synapses and may be the morphological substrate of CRF-controlled LHRH release by the human hypothalamus. Similar intimate contacts observed between the CRF-IR elements raise the possibility of CRF autoregulation in man. LHRH-CRF juxtapositions show similar pattern as those between the neuropeptide Y-LHRH, SP-LHRH, and catecholaminergic-LHRH elements. Thus, these neurotransmitters may be colocalized in the medial preoptic and tuberal areas, regulating LHRH release at the level of the same neuron. Moreover, because LHRH-IR perikarya in the rostral diencephalon do not seem to receive innervation from CRF-IR, NPY-IR, SP-IR, or catecholaminergic elements, this subset of the LHRH neurons may have an entirely different function and may not play a role in LHRH control of gonadal functions.
Acknowledgments
We thank Csaba Bohata (Albert Szent-Gyorgyi University, Szeged, Hungary) for the computer work, Gabriella Kovács Bohata (Albert Szent-Gyorgyi University, Szeged, Hungary) for expert technical assistance, and Dr. Thackery Gray (Department of Cell Biology, Neurobiology and Anatomy, Loyola University Chicago, Maywood, IL) for letting us use his microscope.
Footnotes
Abbreviations: CRF, Corticotropin-releasing factor; DAB, diaminobenzidine; -IR, immunoreactive; LHRH, LH-releasing hormone; NPY, neuropeptide Y; PVN, paraventricular nucleus; SP, substance P.
Received July 2, 2002.
Accepted September 4, 2002.
References
This article has been cited by other articles:
 |
|
 |
|
  J. C. Mitchell, X. F. Li, L. Breen, J.-C. Thalabard, and K. T. O'Byrne The Role of the Locus Coeruleus in Corticotropin-Releasing Hormone and Stress-Induced Suppression of Pulsatile Luteinizing Hormone Secretion in the Female Rat Endocrinology, January 1, 2005; 146(1): 323 - 331. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Dudas and I. Merchenthaler Topography and Associations of Leu-Enkephalin and Luteinizing Hormone-Releasing Hormone Neuronal Systems in the Human Diencephalon J. Clin. Endocrinol. Metab., April 1, 2003; 88(4): 1842 - 1848. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
