Potential for Polysialylated Form of Neural Cell Adhesion Molecule ...

0013-7227/01/$03.00/0 Endocrinology Copyright © 2001 by The Endocrine Society

Vol. 142, No. 3 Printed in U.S.A.

Potential for Polysialylated Form of Neural Cell Adhesion Molecule-Mediated Neuroplasticity within the Gonadotropin-Releasing Hormone Neurosecretory System of the Ewe* ´ †, H. T. JANSEN‡, J. D. GLASS, M. WATANABE, H. J. BILLINGS, C. VIGUIE L. COOLEN, M. N. LEHMAN, AND F. J. KARSCH Department of Physiology (F.J.K.), Reproductive Sciences Program (C.V., H.J.B., F.J.K.), University of Michigan, Ann Arbor, Michigan 48109-0404; Department of Cell Biology, Neurobiology and Anatomy, University of Cincinnati College of Medicine (H.T.J., L.C., M.N.L.), Cincinnati, Ohio 45267-0521; Department of Biological Sciences, Kent State University (J.D.G.), Kent, Ohio 44240-0001; and Division of Pediatric Cardiology, Department of Pediatrics, Rainbow Babies and Childrens Hospital, Case Western Reserve University (M.W.), Cleveland, Ohio 44106-6011 ABSTRACT The GnRH neurosecretory system undergoes marked structural and functional changes throughout life. The initial goal of this study was to examine the neuroanatomical relationship between GnRH neurons and a glycoprotein implicated in neuroplasticity, the polysialylated form of neural cell adhesion molecule (PSA-NCAM). Using dual label immunocytochemistry in conjunction with confocal microscopy, we determined that fibers, terminals, and perikarya of GnRH neurons in adult ovariectomized ewes are intimately associated with PSA-NCAM. In the preoptic area, intense PSA-NCAM immunoreactivity was evident around the periphery of GnRH cell bodies. The second goal of this study was to determine whether PSA-NCAM expression associated with GnRH neurons varies in conjunction with seasonal changes in the activity of the GnRH neurosecretory system

M

ORPHOLOGICAL PLASTICITY of the neuroendocrine system during development is extensively documented. A striking example of this phenomenon is the migration of GnRH neurons from the olfactory placode to the diencephalon during fetal life (1, 2). Growing evidence indicates that neuroendocrine plasticity is not limited to prenatal development, but continues postnatally and even into adulthood. A well documented example is the neurohypophyseal-magnocellular system, which undergoes dramatic Received July 7, 2000. Address all correspondence and request for reprints to: Dr. F. J. Karsch, Reproductive Science Program, University of Michigan, 300 North Ingalls Building, Room 1101 SW, Ann Arbor, Michigan 481090404. E-mail: [email protected]. * This work was supported by USDA Grant 97–35203-4908 and NIH Grant HD-30773 (to F.J.K.), NIH Grant MH-057034 (to J.D.G.), the Center for the Study of Reproduction (NIH Grant P30-HD-18258), Standards and Reagents, Morphology and Sheep Research Core Facilities at the University of Michigan, and the Office of the Vice President for Research at the University of Michigan. † Present address: INSERM, U-501, IFR Jean Roche, Faculte´ Nord, boulevard Pierre Dramard, 13916 Marseilles Cedex 20, France. ‡ Present address: Department of Veterinary and Comparative Anatomy, Pharmacology and Physiology, Washington State University College of Veterinary Medicine, P.O. Box 646520, Pullman, Washington 99164-6520.

in ovariectomized ewes treated with constant release implants of estradiol. During the breeding season when reproductive neuroendocrine activity was enhanced, the expression of PSA-NCAM immunoreactivity associated with GnRH neurons was significantly greater than that during anestrus when GnRH secretion was reduced. This difference, which occurred despite an unchanging ovarian steroid milieu, was not observed in preoptic area structures devoid of GnRH immunoreactivity, suggesting that the seasonal change is at least partially specific to the GnRH system. The close association between PSA-NCAM and GnRH neurons and the change in this relationship in conjunction with seasonal alterations in GnRH secretion provide anatomical evidence that this molecule may contribute to seasonal remodeling of the GnRH neurosecretory system of the adult. (Endocrinology 142: 1317–1324, 2001)

morphological change in response to physiological stimuli such as suckling and dehydration (3, 4). Another instance occurs within hypothalamic centers that govern reproduction. In the rat and monkey, estradiol induces a synaptic reorganization within the basal hypothalamus (5, 6). These changes in the rat occur during the course of the estrous cycle (7), and they have been related to generation of the preovulatory LH surge (8). With regard to GnRH neurons, the preovulatory GnRH surge of the rat is associated with decreased glial ensheathment of GnRH terminals that contact portal vascular loops in the median eminence (9). Gonadal steroids influence innervation and glial ensheathment of GnRH perikarya of monkeys (10), and synaptic input to GnRH neurons changes in association with puberty in monkeys (11) and with reproductive aging in rats (12). Morphological plasticity of the GnRH system is also expressed during changes in season. In this regard, the number of synaptic inputs to GnRH neurons in the ewe decreases between the breeding and the nonbreeding season in conjunction with the decreased pulsatile GnRH secretion that leads to anestrus (13). Despite extensive evidence for neuroplasticity within the adult GnRH system, the underlying molecular basis remains unknown. In the neurohypophyseal-magnocellular system, proteins that influence neuronal remodeling, in particular

1317

1318

PSA-NCAM AND SEASONAL GnRH

the neural cell adhesion molecule (NCAM), are implicated in plasticity during periods of altered neurosecretion (14 –16). Of interest, NCAM and its polysialylated isoform (PSANCAM), which promotes plasticity by modifying the stability of cell-cell contacts, are expressed along the prenatal migratory path of GnRH neurons (17, 18). The importance of these molecules to the migratory process is suggested by findings that GnRH neuronal migration is perturbed by immunoneutralization of NCAM or by enzymatic alteration of its PSA isoform (18 –20). Both NCAM and PSA-NCAM continue to be expressed in the adult brain and, of particular interest to the present study, are found in regions that contain GnRH perikarya, fibers, and terminals (21–23). Further, the expression of NCAM and PSA-NCAM in the preoptic area and hypothalamus of Siberian hamsters is influenced by photoperiodic manipulations that alter GnRH secretion (24). All of these findings are consistent with a role for NCAM and PSA-NCAM in remodeling of the GnRH system. Nevertheless, it has not been assessed whether these molecules exist within or immediately adjacent to GnRH neurons in the adult brain. In this study, we explored the potential for PSA-NCAMmediated plasticity in the GnRH neurosecretory system of the adult ewe, a model in which seasonal changes in both structure and function of the GnRH system are well documented (25). As a first step, we examined whether PSANCAM is expressed within or directly adjacent to GnRH neurons. Having found that it is, we determined whether PSA-NCAM expression associated with GnRH neurons varies in conjunction with seasonal alterations in GnRH neuronal function. Materials and Methods Animals Adult Suffolk ewes were maintained outdoors under natural photoperiod at the Sheep Research Facility (Ann Arbor, MI; 42°18⬘N). Animals were fed hay and had free access to water and mineral licks. A preliminary study was conducted on three long-term ovariectomized ewes to validate the PSA-NCAM immunocytochemical procedure and to gain an initial assessment of the relationship of GnRH to PSA immunoreactivity. The main study was conducted on eight long-term ovariectomized ewes allocated to two groups: 1) animals killed during the breeding season (December 23; n ⫽ 4), and 2) animals killed during anestrus (April 8; n ⫽ 4). Ovariectomy was performed under aseptic conditions. The University of Michigan Committee on the Use and Care of Animals approved all procedures.

Monitoring reproductive neuroendocrine state All ewes were ovariectomized at least 4 months before death. Two weeks before collection of tissues, a 3-cm sc estradiol implant (26) was placed into each ewe to restore a luteal phase concentration of circulating estradiol (2–3 pg/ml) (27). Blood was sampled daily by jugular venipuncture from the first day of estradiol replacement until death to monitor LH secretion as an index of seasonal reproductive state. During anestrus, the GnRH neurosecretory system is exquisitely sensitive to estradiol negative feedback and thus pulsatile GnRH and LH secretion are dramatically suppressed within a few days after placing the estradiol implants (27, 28). In contrast, animals in the breeding season are much less responsive to estradiol negative feedback, and GnRH and LH secretion remain high. Blood samples were centrifuged to remove cells, and serum was stored at ⫺20 C. LH was assayed in duplicate 20- to 200-␮l aliquots using a modification (29) of a previously reported RIA (30, 31). Concentrations are expressed in terms of NIH LH-S12. The sensitivity for 200 ␮l (95%

Endo • 2001 Vol. 142 • No. 3

confidence interval of buffer control) was 0.58 ng/ml. All samples were run in a single assay; mean intraassay coefficients of variation for serum pools displacing radiolabeled LH to approximately 30% and 65% of buffer control values were 6.7% and 14.2%, respectively.

Immunocytochemistry Sodium heparin (25,000 U, iv) was injected at 10 min and 1 min before euthanasia with a barbiturate overdose (Beuthanasia, Schering-Plough Animal Health Corp., Kenilworth, NJ). Ewes were immediately decapitated, both carotid arteries were catheterized, and the head was perfused with 4 liters 2% paraformaldehyde in 0.1 m sodium phosphate buffer (pH 7.4) containing 10,000 U/liter heparin and 0.01% sodium nitrite. After perfusion, the brain was removed, and a block containing the whole hypothalamus and preoptic area (from 5 mm rostral to the optic chiasm to the anterior pole of the mammillary bodies) was isolated and postfixed overnight at 4 C in the same fixative. The block was washed in PBS (pH 7.3) and immersed in 10% sucrose for 24 h at 4 C and then in 20% sucrose containing 0.01% sodium azide at 4 C until sectioning. Serial 50-␮m coronal sections were obtained using a freezing microtome. Sections were stored at ⫺20 C in a cryopreservative (32). Every fifth section was stained for GnRH and PSA-NCAM immunoreactivity. Sections were washed in 0.01 m PBS and then incubated for 30 min at room temperature in 0.1 m glycine in PBS and 0.2% Triton X-100. After washing, sections were preincubated for 45 min at room temperature in a blocking solution (PBS, 2% BSA, 3% normal goat serum, 0.2% Triton X-100, and 0.01% sodium azide). Sections were exposed to primary antiserum to GnRH (LR1, provided by Dr. Robert Benoit, McGill University, Montre´al, Que´bec, Canada) at a dilution of 1:10,000 and antibody to PSA-NCAM (5A5, monoclonal mouse IgM directed against the 2– 8-linked polysialic acid of NCAM, provided by Dr. Urs Rutishauser, Sloan–Kettering Institute, New York) at a dilution of 1:250 in the blocking solution. Sections were incubated for 72–96 h at 4 C. After washing, sections were incubated for 90 –120 min in a mixture of Alexa 594-conjugated goat antirabbit IgG (Molecular Probes, Inc., Eugene, OR) at a dilution of 1:500 and goat antimouse fluorescein isothiocyanateconjugated IgM ␮-specific (Sigma, St. Louis, MO) at a dilution of 1:250. Sections were washed, mounted, and coverslipped using Prolong (Molecular Probes, Inc.) mounting medium for fluorescence. Stained sections were examined with a ⫻60 oil immersion lens on a confocal microscope (MRC-600, Bio-Rad Laboratories, Inc., Hercules, CA). For each GnRH neuron, a series of serial optical sections (z-series, 1- to 1.5-␮m thickness) was generated. In addition, a series of GnRH confocal images was generated from the median eminence of one ewe in each group. In addition to aforementioned tissue processing, several additional sections through the median eminence from these ewes were processed using an avidin-biotin immunofluorescent procedure involving tyramide amplification (33). Specifically, sections were washed in 0.1 m PBS, blocked with 1% H2O2 in 0.1 m PBS for 10 min, and washed again with PBS. This was followed by incubation in a blocking solution of 0.1 m PBS containing 0.1% Triton X-100 and 4% normal donkey serum for 1 h. Sections were then incubated overnight (⬃18 h) at room temperature with a cocktail of primary antibodies to PSA-NCAM (5A5 diluted 1:250 in the blocking solution) and GnRH (LR-1 diluted 1:3000). After this, sections were washed in PBS and incubated with biotin-conjugated donkey antimouse IgG (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA) diluted 1:200 in the blocking solution for 1 h. The sections were again washed in PBS, incubated with an avidin-biotin complex (ABC-elite kit, Vector Laboratories, Inc., Burlingame, CA) diluted 1:200 in PBS for 1 h, washed again in PBS, and incubated in biotin-conjugated tyramide (NEN Life Science Products, Boston, MA) diluted 1:250 in PBS and 0.003% H2O2 for 10 min. This was followed by another wash, incubation in CY2-conjugated streptavidin (Jackson ImmunoResearch Laboratories, Inc.) diluted 1:100 in PBS for 30 min, another wash with PBS, and incubation for 30 min with CY3-conjugated donkey antirabbit IgG (Jackson ImmunoResearch Laboratories, Inc.) diluted 1:100 in PBS. Sections were then washed in 0.1 m phosphate buffer, mounted, and coverslipped. Images were obtained using a laser scanning confocal microscope (LSM 510, Carl Zeiss, New York, NY) at ⫻10 and ⫻63 magnification. Overall, findings using the initial procedure were confirmed, including the elimination of PSA-NCAM immunoreactivity by omission of the primary antibody or treatment of tissue sections with endoneuraminidase enzyme (see below). The modified


1319

method, however, provided better definition of the PSA-NCAM immunoreactive signal. This procedure was used for the results presented in Fig. 4, which illustrates PSA-NCAM/GnRH distribution in the median eminence.

Validation of PSA-NCAM immunocytochemistry The specificity of PSA-NCAM staining was checked by alternatively omitting one or another of the primary antibodies in the presence of both secondary antibodies. In addition, the specificity of the 5A5 antibody for PSA-NCAM was verified by pretreating a set of sections with the endoneuraminidase enzyme, endo-N (provided by Dr. Urs Rutishauser), which specifically cleaves sialic acid polymers in chains of eight or longer (34). Different sets of preoptic and hypothalamic sections were incubated for 1 h at room temperature in phosphate buffer (0.1 m; pH 7.4) containing the enzyme at one of two dilutions (1:500 or 1:5000). Stained sections were observed under epifluorescence (Leica Corp., Orthoplan, Deerfield, IL) using a ⫻40 lens. Photographic images were obtained under the same conditions (ASA, exposure) for positive control and experimental sections.

Data analysis Confocal images from the z-series of each individual GnRH neuron were analyzed using NIH Image software. Both image channels (GnRH and PSA-NCAM) could be viewed simultaneously and thus analyzed at the same time. The extent of PSA-NCAM immunoreactivity associated with GnRH neurons was quantified using two separate methods by different investigators working independently of each other and without knowledge of the experimental group. For both methods, analysis of GnRH neurons was performed using three consecutive z-sections in which a nuclear profile was present. The first analysis determined the percentage of the perimeter of the GnRH cell body juxtaposed to PSA-NCAM immunoreactivity (cell body perimeter excluded interfaces of dendrites/soma and axon/soma). Fifteen to 32 GnRH somas in the preoptic area were analyzed for each ewe. Two parameters were compared between breeding and anestrous seasons using unpaired t test: 1) mean percentage of the perimeter of GnRH somas juxtaposed to PSA-NCAM immunoreactivity, and 2) percentage of GnRH somas contacted at least 90% by PSA-NCAM immunoreactivity. The second analysis was based on the estimated density of PSANCAM immunoreactivity associated with GnRH neurons. For each zsection, a 150 ⫻ 150-pixel circle was first centered over a GnRH neuron, and a threshold value based on the average pixel intensity for the entire GnRH channel was applied. Then, the number of pixels (pixel area) occupied by each GnRH neuron was calculated for each z-section. An area of the exact same size was analyzed in the corresponding PSANCAM channel with threshold values based on the average pixel density for that channel, and an estimate of the density of PSA-NCAM immunoreactivity was calculated as the number of pixels. This value was also used to calculate a ratio of PSA-NCAM density to GnRH somal area. In addition, an estimate of PSA-NCAM density was calculated in an area of the optical field devoid of GnRH immunoreactivity. For each ewe, mean values derived from the three consecutive z-sections examined were determined for 1) the ratio of PSA-NCAM density to GnRH somal area; 2) the density of PSA-NCAM immunoreactivity associated with GnRH neurons; and 3) the density of PSA-NCAM immunoreactivity not associated with GnRH neurons. Statistical comparisons between the two seasons were made using an unpaired t test. For both types of analysis, parameters that do not change with season and/or reproductive neuroendocrine state were also measured as an assessment of analysis consistency: 1) GnRH cell nuclear diameter for the first type of analysis; and 2) GnRH somal area (expressed as pixels) for the second method.

Results Reproductive neuroendocrine state

In anestrous ewes, estradiol implants caused the serum LH concentration to decrease dramatically, from greater than 10 to about 1 ng/ml within 7–10 days (Fig. 1). In contrast, LH

FIG. 1. Mean plasma LH concentration in ovariectomized ewes before and for 2 weeks after implantation of estradiol (arrow) during the breeding season (December; F) or anestrous season (March to April; E). Ewes were killed to collect brains for immunocytochemical analysis after the last blood sample.

did not decrease after estradiol treatment in breeding season animals. Specificity of staining

PSA-NCAM immunoreactivity was observed within the preoptic area and hypothalamus. When sections were pretreated with the endo-N enzyme, no PSA immunostaining above background was observed with either the 1:500 or 1:5000 dilution of enzyme (Fig. 2, A vs. B, lower panel shows example for median eminence; similar results were obtained in other areas of the hypothalamus). GnRH immunostaining was not eliminated by endo-N treatment (Fig. 2, A and B, upper panels), although the number of samples examined was insufficient to assess whether GnRH immunoreactivity was quantitatively altered. Omission of either primary antibody in the presence of both secondary antibodies eliminated all staining corresponding to the omitted primary antibody (not illustrated). Relationship of PSA-NCAM to GnRH neurons

PSA-NCAM immunoreactivity was widely distributed throughout the preoptic area, where the majority of GnRH neurons are located in the ewe (35). Almost all GnRH neurons examined throughout the preoptic area were closely associated with PSA-NCAM immunoreactivity (Fig. 3, A–D). Occasional GnRH neurons did not exhibit associated PSANCAM immunoreactivity even though they were located in the vicinity of GnRH neurons that did (Fig. 3B). PSA-NCAM immunoreactivity was associated primarily with the periphery of the GnRH cell soma and the large proximal and distal dendrites and less frequently around axons and small diameter dendrites (Fig. 3C). When GnRH neurons were juxtaposed, the interface between them, whether somatosomatic, dendro-somatic, or axo-somatic, was invariably PSA-NCAM immunoreactive (Fig. 3D). PSA-NCAM-positive structures that appeared as rings of immunoreactivity but did not exhibit GnRH immunoreactivity (e.g. asterisk in

1320

Endo • 2001 Vol. 142 • No. 3


F IG . 2. Epifluorescence microscopy digital images (⫻40) of GnRH (red, top) and PSA-NCAM (green, bottom) immunoreactivity within the external zone of the lateral median eminence of an ovariectomized ewe (the top of each panel is oriented dorsally). A, Control stained in usual conditions; B, sections preincubated with endo-N (1:5000), an enzyme that specifically cleaves sialic acid polymers from NCAM. Images were collected and processed in the same conditions. Endo-N pretreatment did not prevent GnRH immunostaining (B, top), but completely eliminated PSA-NCAM immunoreactivity (B, bottom).

Fig. 3A) were seen in proximity to some GnRH neurons (31% of GnRH neurons analyzed were juxtaposed to such structures). Whether this pattern of PSA-NCAM labeling represented immunoreactivity associated with non-GnRH neurons or glial cells was not clear. In the median eminence, a region of dense PSA-NCAM immunoreactivity was identified along the lateral border of the external zone (Fig. 4, A and C), associated with a region of intense GnRH-immunoreactive axons/terminals (Fig. 4, B and C). This region also contained fibers in which PSANCAM and GnRH immunoreactivity overlapped (Fig. 4D, long arrows) as well as PSA-NCAM-immunoreactive fibers not associated with GnRH labeling (Fig. 4D, large arrowheads) and GnRH immunoreactivity not associated with PSANCAM labeling (Fig. 4D, small arrowheads). Seasonal comparison

According to the first method of analysis, the percentage of the perimeter of GnRH somas juxtaposed to PSA-NCAM immunoreactivity (i.e. percentage of PSA contact) was significantly greater in the breeding season than in anestrus (P ⬍ 0.02), although the absolute difference was relatively small (Fig. 5A). Similarly, the percentage of GnRH somas contacted at least 90% by PSA-NCAM immunoreactivity was greater in the breeding season than in anestrus (42.7 ⫾ 5.0% of GnRH neurons in the breeding season vs. 16.5 ⫾ 5.3 in anestrus; P ⬍ 0.02; Fig. 5B). Seasonal differences were confirmed by the second analysis. The ratio of PSA-NCAM to GnRH labeling and the density of PSA-NCAM immunoreactivity associated with GnRH neurons were 23% and 38% greater (P ⬍ 0.05), respectively, in the breeding season than in anestrus (Fig. 5, C and D). However, the density of PSA-NCAM immunore-

activity in POA areas devoid of GnRH immunostaining did not exhibit any seasonal differences (Fig. 5E). Neither GnRH soma size nor nuclear diameter exhibited a seasonal difference (data not illustrated). Discussion

Prior studies have revealed that changes in GnRH secretion may be accompanied by a neuroplastic remodeling of the GnRH neurosecretory system. For example, in the rat, estradiol induces a synaptic remodeling within the arcuate nucleus, an alteration related to generation of the preovulatory LH surge (5, 7, 8). In addition, glial ensheathment of GnRH terminals in contact with portal capillaries in the median eminence decreases during the preovulatory LH surge of the rat (9). Further, synaptic inputs onto GnRH neurons change at the times of puberty, seasonal reproductive transitions, and reproductive neuroendocrine aging (11–13). The present study indicates that in the adult ewe, PSA-NCAM immunoreactivity is intimately associated with GnRH neurons at the levels of both terminals in the median eminence and perikarya in the preoptic area. As noted above, PSANCAM has been implicated in remodeling of both the developing (e.g. prenatal GnRH cell migration) (18 –20) and adult neuroendocrine systems (e.g. hypothalamo-neurohypophyseal system) (14 –16). These findings together with the widespread role of PSA-NCAM in promoting neuroplasticity (36 –38) lead us to hypothesize that PSA-NCAM may provide a molecular substrate for structural remodeling of the GnRH system of adults. Several aspects of the morphological relationship between PSA-NCAM immunoreactivity and GnRH neurons are noteworthy. The first is the extent of this relationship; nearly all


FIG. 3. Confocal microscopy localization of PSA-NCAM (green) and GnRH (red) immunoreactivity (⫻60, oil immersion) within the preoptic area of estradiol-treated ovariectomized ewes. Yellow indicates areas of overlap of PSA-NCAM and GnRH immunoreactivities. PSANCAM immunoreactivity (arrowheads) was prominent around the periphery of most GnRH cell bodies and their large proximal dendrites (A–D). Occasional GnRH neurons were seen that were not closely associated with PSA-NCAM immunoreactivity (e.g. right cell in B). Axonal varicosities of GnRH neurons (arrows) were less frequently associated with PSA immunoreactivity. Note the PSA-NCAM immunoreactivity at the interface between two GnRH neurons (B and D). The asterisk in A indicates an example of a ring-like, PSA-NCAMpositive structure devoid of GnRH.

GnRH neurons examined in the preoptic area were associated with PSA-NCAM immunoreactivity. Second, PSANCAM immunoreactivity was observed at the periphery of GnRH perikarya, consistent with NCAM being a cell surface protein that is polysialylated on its extracellular domain (38). However, an ultrastructural study will be required to determine whether PSA-NCAM is contained within the plasma membrane of GnRH neurons or within other cells or neural elements in apposition to GnRH neurons. If PSA-NCAM is, in fact, expressed on the extracellular surface of GnRH perikarya, this might explain why GnRH neurons are relatively less innervated than adjacent nonidentified neurons (13, 39), given evidence to suggest that PSA-NCAM impedes the formation and stabilization of synapses (40). Our finding that PSA-NCAM is rarely associated with small diameter GnRH dendrites complements earlier observations that these processes are the major sites of synaptic inputs to ovine GnRH neurons (39).

1321

The present and previous observations are consistent with the possibility that PSA-NCAM is functionally significant to seasonal plasticity of the GnRH neurosecretory system of adults. In the Siberian hamster, photoperiodic manipulations that cause seasonal shifts in gonadotropin secretion also alter the relative amounts of PSA and other NCAM isoforms in the basal hypothalamus, a region important to GnRH regulation (24). Our present findings in sheep extend those observations by demonstrating that the expression of PSA-NCAM is intimately associated with preoptic GnRH neurons and that this expression varies with the seasonal shift in reproductive neuroendocrine activity. Specifically, we observed a significant difference in three indexes of the PSA-NCAM/GnRH relationship: the extent to which GnRH neurons are juxtaposed to PSA-NCAM immunoreactivity, the ratio of PSANCAM to GnRH immunoreactivity, and the density of PSANCAM immunoreactivity associated with GnRH neurons. Each of these measures was greater during the breeding season than during anestrus. Further, this seasonal difference appeared to be at least partially specific to GnRH neurons, because PSA-NCAM immunoreactivity in adjacent preoptic areas devoid of GnRH neurons and fibers did not change with the season. Although our findings are consistent with the possibility that PSA-NCAM may contribute to the seasonality of the GnRH system, it should be noted that in themselves they do not demonstrate that this molecule plays a role in seasonal changes in either the structure or function of GnRH neurons. Such a conclusion requires evidence that seasonal alterations in the GnRH system can be prevented by perturbation of endogenous PSA-NCAM expression by endoneuraminidase or other means. It is important to note that our study was conducted in ovariectomized ewes treated with constant release estradiol implants. Thus, the seasonal difference in PSA-NCAM expression associated with GnRH neurons was not the consequence of an altered ovarian steroid milieu, which in itself can induce hypothalamic neuroplasticity (5– 8, 10). Rather, it reflects an intrinsic seasonality in expression of the molecule. Prior work in this animal model has demonstrated that in the face of an unvarying gonadal steroid milieu, there are marked seasonal changes in both synaptic inputs to GnRH neurons and the pulsatile secretion of GnRH into pituitary portal blood (13, 28). The seasonal difference in GnRH secretion was confirmed indirectly in the present study by serum LH concentrations; values were not reduced by estradiol in the breeding season, but they were profoundly suppressed during anestrus. Both the direction and the magnitude of the seasonal difference in PSA-NCAM associated with GnRH neurons are noteworthy. Regarding magnitude, this seasonal difference was relatively small. Although it could be argued that such a subtle difference is not likely to have significant functional consequences, it is important to note that a minute change in NCAM polysialylation can dramatically alter the physical and functional properties of the molecule (38, 41). Further, it should be noted that the tissues analyzed in the present study were obtained several months before or after the seasonal reproductive transition, not during the actual transition, when the shift in GnRH secretion occurs and the PSA-NCAM difference might be greater. In addition, it is important to

1322


Endo • 2001 Vol. 142 • No. 3

FIG. 4. A–C, Low power confocal images (⫻10, 10-␮m thickness) of PSANCAM (green) and GnRH (red) immunoreactivity in the median eminence at the level of the tubero-infundibular sulcus of an estradiol-treated ovariectomized ewe in the anestrous season. Intense PSA-NCAM (A) and GnRH (B) immunoreactivity was located along the external layer of the lateral portion of the median eminence. Note that the area where PSA-NCAM and GnRH immunoreactivities overlap appears yellow in the overlay of the two images (C, arrows). D, Higher power confocal image (⫻63, water immersion, 1-␮m thickness) of the area indicated by the square in C. Fibers in which both PSA-NCAM and GnRH immunoreactivities overlapped (long arrows) were present along with fibers expressing either PSANCAM (large arrowheads) or GnRH (small arrowheads) immunoreactivity. V3, Third ventricle; PT, pars tuberalis.

emphasize that a seasonal change in PSA-NCAM itself might not be critical. Rather, its mere presence in or around GnRH neurons may promote neuroplastic remodeling, as suggested for the neurohypophyseal magnocellular system, where its expression is constitutive and its action permissive (15, 16). With regard to direction of the seasonal change, we observed greater expression of PSA-NCAM associated with GnRH neurons in the breeding season than during anestrus. Our results in sheep are consistent with findings in Siberian hamsters that the expression of PSA-NCAM in hypothalamic regions that regulate reproduction is more prevalent under a stimulatory photoperiod than under an inhibitory day length (24). It is of interest that synaptic input to GnRH neurons is also greater during the breeding season than during anestrus (13). This raises questions related to the role that PSA-NCAM may play in promoting seasonal changes in the innervation of GnRH neurons. Although it is recognized that the PSA isoform of NCAM promotes neuroplasticity by modifying the stability of cell to cell contacts, its specific role in synaptic rearrangements is not well understood. On the one hand, it has recently been suggested that PSA-NCAM can impede the formation of synapses or reduce their stability (40). This was based on observations that genetic deletion of PSA-NCAM or enzymatic removal of PSA from NCAM enhanced synaptic number in the pyramidal cell layer of the hippocampus. On the other hand, it has been suggested that PSA-NCAM can pro-

mote the formation of synapses onto neuroendocrine cells of the hypothalamic magnocellular system, where PSA-NCAM was found to be present in both neurons and adjacent glial cells (16). In this system, lactation and dehydration induce both the withdrawal of glial processes surrounding oxytocin and vasopressin neurons and an increase in synaptic input to these neurons; this is associated with enhanced hormone release (16). Enzymatic removal of PSA from NCAM blocked both the withdrawal of glial processes and the increased synaptic input. It was suggested, therefore, that PSA-NCAM promotes the innervation of magnocellular neurons by acting in either a permissive or a stimulatory fashion, possibly involving glial withdrawal (16). It is also important to note that the expression of PSA-NCAM is activity dependent (15, 16, 38). Therefore, the higher levels of PSA-NCAM immunoreactivity associated with GnRH neurons in the breeding season might be a consequence, rather than a cause, of the increased GnRH neurosecretory activity at this time of year. In addition to influencing synaptic rearrangements, PSANCAM might affect the GnRH system via alternative mechanisms. For example, PSA-NCAM might promote rearrangements of cells in the proximity of GnRH neurons, cells that could influence neurosecretion by paracrine mechanisms. Paracrine regulation of a reproductive neuroendocrine transition has been implicated at the time of puberty in rats; in this case, growth factors secreted by astrocytes appear to be critically important for activating GnRH neurons (42, 43). The present study did reveal extensive PSA immunostaining


1323

ically to promote remodeling within the GnRH axis, and they encourage future work to establish the functional significance of this molecule to both the structure and the secretory activity of the GnRH system in adults. Acknowledgments The authors thank Douglas Doop, Gary McCalla, Dr. Thomas Harris, and Dr. Deborah Battaglia for assistance with animal experimentation; Kay Brabec and Dr. A. Kent Christensen for advice and assistance with immunocytochemistry; Martha Brown for performing LH assays; Drs. Gordon Niswender, Leo E. Reichert, Jr., and Al Parlow for providing assay reagents; Dr. Robert Benoit for providing antiserum used for GnRH immunocytochemistry; Dr. Urs Rutishauser for providing endoneuraminidase enzyme and antibody for PSA-NCAM immunocytochemistry; and Chad Foradori for assistance with data analysis.

References

FIG. 5. Summary of the extent of PSA-NCAM immunoreactivity associated with preoptic GnRH neurons in estradiol-treated ovariectomized ewes in the breeding season (BR S, f) or anestrus (AN, 䡺). Data on the left are the results of the first data analysis and depict the percentage of the perimeters of GnRH somas exhibiting PSA-NCAM immunoreactivity, i.e. PSA contact (A), and the percentage of GnRH somas that are contacted at least 90% by PSA-NCAM immunoreactivity (B). Data on the right are the results of the second method of analysis and depict the ratio of PSA-NCAM density to GnRH somal area (C), the density of PSA-NCAM immunoreactivity associated with GnRH perikarya (D), and the density of PSA-NCAM immunoreactivity not directly associated with GnRH perikarya (E). For each type of data analysis, measurements were performed on three consecutive sections of a z-series for each neuron. Results are presented as the mean ⫾ SEM (n ⫽ 4 ewes for both breeding and anestrous seasons). *, P ⱕ 0.02 for the first method of data analysis and P ⱕ 0.05 for the second method. See text for further details of data analysis.

in structures located near GnRH neurons and fibers, but which themselves were not immunoreactive for GnRH. Further research is required to identify these structures and their possible relationship to the secretory activity of GnRH neurons. In summary, our findings demonstrate that a close anatomical association exists between PSA-NCAM and GnRH neurons in the adult ewe. This holds both for the median eminence, where GnRH fibers and terminals are juxtaposed to PSA-NCAM immunoreactivity, and the preoptic area, where GnRH perikarya are extensively contacted by PSANCAM. Further, the expression of PSA-NCAM associated with preoptic GnRH neurons changes seasonally, being greater during the period of enhanced GnRH secretion in the breeding season than during anestrus. These findings provide evidence that PSA-NCAM is well positioned anatom-

1. Schwanzel-Fukuda M, Pfaff DW 1989 Origin of luteinizing hormone-releasing hormone neurons. Nature 338:161–164 2. Wray S, Grant P, Gainer H 1989 Evidence that cells expressing luteinizing hormone-releasing hormone mRNA in the mouse are derived from progenitor cells in the olfactory placode. Proc Natl Acad Sci USA 86:8132– 8136 3. Theodosis DT, Poulain DA 1992 Neuronal-glial and synaptic remodelling in the adult hypothalamus in response to physiological stimuli. In: Chadwick DJ, Marsh J (eds) Functional Anatomy of the Neuroendocrine Hypothalamus. Ciba Foundation Symposium 168, Wiley & Sons, Chichester, pp 209 –232 4. Montagnese C, Poulain DA, Vincent J-D, Theodosis DT 1988 Synaptic and neuronal-glial plasticity in the adult oxytocinergic system in response to physiological stimuli. Brain Res Bull 20:681– 692 5. Pa´rducz A, Perez J, Garcia-Segura LM 1993 Estradiol induces plasticity of GABAergic synapses in the hypothalamus. Neuroscience 53:395– 401 6. Naftolin F, Leranth C, Perez J, Garcia-Segura LM 1993 Estrogen induces synaptic plasticity in adult primate neurons. Neuroendocrinology 57:935–939 7. Olmos G, Naftolin F, Perez J, Tranque PA, Garcia-Segura LM 1989 Synaptic remodeling in the rat arcuate nucleus during the estrous cycle. Neuroscience 32:663– 667 8. Horvath TL, Garcia-Segura LM, Naftolin F 1997 Lack of gonadotropin-positive feedback in the male rat is associated with lack of estrogen-induced synaptic plasticity in the arcuate nucleus. Neuroendocrinology 65:136 –140 9. King JC, Rubin BS 1995 Dynamic alterations in luteinizing hormone-releasing hormone (LHRH) neuronal cell bodies and terminals of adult rats. Cell Mol Neurobiol 15:89 –106 10. Witkin JW, Ferin M, Popilskis SJ, Silverman A-J 1991 Effects of gonadal steroids on the ultrastructure of GnRH neurons in the rhesus monkey: synaptic input and glial apposition. Endocrinology 129:1083–1092 11. Perera AD, Plant TM 1997 Ultrastructural studies of neuronal correlates of the pubertal reaugmentation of hypothalamic gonadotropin-releasing hormone (GnRH) release in the rhesus monkey (Macaca mulatta). J Comp Neurol 385:71– 82 12. Witkin J 1987 Aging changes in synaptology of luteinizing hormone-releasing hormone neurons in male rat preoptic area. Neuroscience 22:1003–1013 13. Xiong JJ, Karsch FJ, Lehman MN 1997 Evidence for seasonal plasticity in the gonadotropin-releasing hormone (GnRH) system of the ewe: changes in synaptic inputs onto GnRH neurons. Endocrinology 138:1240 –1250 14. Theodosis DT, Rougon G, Poulain D 1991 Retention of embryonic features by an adult neuronal system capable of plasticity: polysialylated neural cell adhesion molecule in the hypothalamo-neurohypophysial system. Proc Natl Acad Sci USA 88:5494 –5498 15. Theododis DT, El Majdoubi M, Pierre K, Poulain DA 1998 Factors governing activity-dependent structural plasticity of the hypothalamoneurohypophysial system. Cell Mol Neurobiol 18:285–298 16. Theodosis DT, Bonhomme R, Vitiello S, Rougon G, Poulain DA 1999 Cell surface expression of polysialic acid on NCAM is a prerequisite for activitydependent morphological neuronal and glial plasticity. J Neurosci 19:10228 –10236 17. Schwanzel-Fukuda MS, Abraham S, Crossin KL, Edelman GM, Pfaff DW 1992 Immunocytochemical demonstration of neural cell adhesion molecule (NCAM) along the migration route of luteinizing hormone-releasing hormone (LHRH) neurons in mice. J Comp Neurol 321:1–18 18. Schwanzel-Fukuda M, Reinhard GR, Abraham S, Crossin KL, Edelman GM, Pfaff DW 1994 Antibody to neural cell adhesion molecule can disrupt the migration of luteinizing hormone-releasing hormone neurons into the mouse brain. J Comp Neurol 342:174 –185 19. Yoshida K, Rutishauser U, Crandall JE, Schwarting GA 1999 Polysialic acid facilitates migration of luteinizing hormone-releasing hormone neurons on vomeronasal axons. J Neurosci 19:794 – 801 20. Murakami S, Seki T, Rutishauser U, Arai Y 2000 Enzymatic removal of

1324

21.

22.

23. 24. 25. 26. 27.

28.

29.

30. 31.


polysialic acid from neural cell adhesion molecule perturbs the migration route of luteinizing hormone-releasing hormone neurons in the developing chick forebrain. J Comp Neurol 420:171–181 Bonfanti L, Olive S, Poulain DA, Theodosis DT 1992 Mapping of the distribution of polysialylated neural cell adhesion molecule throughout the central nervous system of the adult rat: an immunohistochemical study. Neuroscience 49:419 – 436 Alonso G, Prieto M, Legrand A, Chauvet N 1997 PSA-NCAM and B-50/ GAP-43 are coexpressed by specific neuronal systems of the adult mediobasal hypothalamus that exhibit remarkable capacities for morphological plasticity. J Comp Neurol 384:181–199 Perera AD, Lagenaur CF, Plant TM 1993 Postnatal expression of polysialic acid-neural cell adhesion molecule in the hypothalamus of the male rhesus monkey (Macaca Mulatta). Endocrinology 133:2729 –2735 Lee W, Watanabe M, Glass JD 1995 Photoperiod affects the expression of neural cell adhesion molecule and polysialic acid in the hypothalamus of the Siberian hamster. Brain Res 690:64 –72 Lehman MN, Goodman RL, Karsch FJ, Jackson GL, Berriman SJ, Jansen HT 1997 The GnRH system of seasonal breeders: anatomy and plasticity. Brain Res Bull 44:445– 457 Karsch FJ, Dierschke DJ, Weick RF, Yamaji T, Hotchkiss J, Knobil E 1973 Positive and negative feedback control by estrogen of luteinizing hormone secretion in the rhesus monkey. Endocrinology 92:799 – 804 Legan SJ, Karsch FJ, Foster DL 1977 The endocrine control of seasonal reproductive function in the ewe: a marked change in response to the negative feedback action of estradiol on luteinizing hormone secretion. Endocrinology 101:818 – 824 Karsch FJ, Dahl GE, Evans NP, Manning JM, Mayfield KP, Moenter SM, Foster DL 1993 Seasonal changes in gonadotropin-releasing hormone secretion in the ewe: alteration in response to the negative feedback action of estradiol. Biol Reprod 49:1377–1383 Hauger RL, Karsch FJ, Foster DL 1977 A new concept for control of the estrous cycle of the ewe based on the temporal relationships between luteinizing hormone, estradiol and progesterone in peripheral serum and evidence that progesterone inhibits tonic LH secretion. Endocrinology 101:807– 817 Niswender GD, Midgley Jr AR, Reichert Jr LE 1968 Radioimmunologic studies with murine, ovine, and porcine luteinizing hormone. In: Rosenberg E (ed) Gonadotropins. GERON-X, Los Altos, pp 299 –306 Niswender GD, Reichert Jr LE, Midgley Jr AR, Nalbandov AV 1969 Radio-

32.

33.

34.

35.

36.

37.

38. 39.

40.

41.

42. 43.

Endo • 2001 Vol. 142 • No. 3

immunoassay for bovine and ovine luteinizing hormone. Endocrinology 84:1166 –1173 Watson Jr RE, Weigand SJ, Clough RW, Hoffman GE 1986 Use of a cryoprotecant to maintain long-term peptide immunoreactivity and tissue morphology. Peptides 7:155–159 Hunyady B, Krempels K, Harta G, Mezey EJ 1996 Immunohistochemical signal amplification by catalyzed reporter deposition and its application to double immunostaining. J Histochem Cytochem 44:1353–1362 Rutishauser U, Watanabe M, Silver J, Troy FA, Vimr ER 1985 Specific alteration of NCAM-mediated cell adhesion by an endoneuraminidase. J Cell Biol 101:1842–1849 Lehman MN, Robinson JE, Karsch FJ, Silverman A-J 1986 Immunocytochemical localization of luteinizing hormone-releasing hormone (LHRH) pathways in the sheep brain during anestrus and the mid-luteal phase of the estrous cycle. J Comp Neurol 244:19 –35 Rutishauser U, Landmesser L 1996 Polysialic acid in the vertebrate nervous system: a promoter of plasticity in cell-cell interactions. Trends Neurosci 19:422– 427 Muller D, Wang C, Skibo G, Toni N, Cremer H, Calaora V, Rougon G, Kiss JZ 1996 PSA-NCAM is required for activity-induced synaptic plasticity. Neuron 17:413– 422 Fields RD, Itoh K 1996 Neural cell adhesion molecules in activity-dependent development and synaptic plasticity. Trends Neurosci 19:473– 480 Lehman MN, Karsch FJ, Robinson JE, Silverman A-J 1988 Ultrastructure and synaptic organization of luteinizing hormone-releasing hormone (LHRH) neurons in the anestrous ewe. J Comp Neurol 273:447– 458 Seki T, Rutishauser U 1998 Removal of polysialic acid-neural cell adhesion molecule induces aberrant mossy fiber innervation and ectopic synaptogenesis in the hippocampus. J Neurosci 18:3757–3766 Doherty P, Cohen J, Walsh FS 1990 Neurite outgrowth in response to transfected N-CAM changes during development and is modulated by polysialic acid. Neuron 5:209 –219 Ojeda SR, Ma YJ 1995 The role of growth factors in the neuroendocrine control of female sexual development. Curr Opin Endocrinol Diabetes 2:148 –156 Rage F, Hill DF, Sena-Esteves M, Breakefield XO, Coffey RJ, Costa ME, McCann SM, Ojeda SR 1997 Targeting transforming growth factor ␣ expression to discrete loci of the neuroendocrine brain induces female sexual precocity. Proc Natl Acad Sci USA 94:2735–2740