Abstract

Recent evidence suggests that the ovine premammillary hypothalamic area (PMH) is an important target for the pineal hormone, melatonin, and its role in seasonal reproduction. In rodents, the PMH is a complex region consisting of several cell groups with differing neurochemical content and anatomical connections. Therefore, to obtain a better understanding of the potential neural targets for melatonin in this area of the sheep brain, we have undertaken a detailed anatomical characterization of the PMH, including its nuclear divisions and the location of neuropeptide/neurotransmitter cells within them. By combining immunocytochemistry for NeuN, a neuronal marker, with Nissl staining in anestrous, ovariectomized, estradiol-treated ewes, we identified three nuclei within the PMH: a caudal continuation of the hypothalamic arcuate nucleus (cARC), the ventral division of the premammillary nucleus (PMv), and the ventral tuberomammillary nucleus (TMv). The cARC contained neurons that were immunoreactive for tyrosine hydroxylase, dynorphin, estrogen receptor α, cocaine- and amphetamine-regulated transcript peptide (CART), and nitric oxide synthase (NOS). The PMv was also characterized by the presence of cells that contained NOS and CART, although the size of these cells was larger than that of their corresponding phenotype in the cARC. By contrast, in the TMv, of the markers examined in the present study, only fibers immunoreactive for orexin were seen. Thus, the ovine PMH is a heterogeneous region comprised of three subdivisions, each with distinct morphological and neurochemical characteristics. This anatomical map of the PMH provides a basis for future studies to determine the functional contribution of each component to the influence of melatonin on seasonal reproduction.

Introduction

Seasonal reproduction is a strategy employed by many species to make use of the selective advantage that results from offspring being born at a time of year when a favorable environment is present [1, 2]. The major environmental cue controlling the timing of seasonal reproductive transitions in sheep is photoperiod [2]. For example, in most breeds of sheep, estrous cycles occur during the short daylengths of autumn and winter (breeding season) and cease during the longer daylengths of spring and summer (anestrus). The influence of daylength on seasonal reproduction in the ewe is mediated by the pineal hormone, melatonin, which provides an endocrine measure of daylength [3, 4]. Melatonin, in turn, acts directly on the brain, ultimately regulating seasonal changes in pulsatile secretion of GnRH from the hypothalamus [5, 6]. A major unanswered question in the study of seasonal reproduction is the identity of the neurons that are responsive to melatonin and that convey this photoperiodic information to the GnRH system and reproductive neuroendocrine axis.

Use of [125I]melatonin for receptor autoradiography has allowed the identification of putative target sites for melatonin within the hypothalamo-hypophyseal system of the sheep. Melatonin binding was initially detected in the anterior and posterior hypothalamus and in the pars tuberalis [7, 8]. Although the pars tuberalis is a target for melatonin in seasonal rhythms of prolactin in rams [9] and, possibly, in ewes [6, 10], it does not appear to mediate the influence of melatonin on the reproductive neuroendocrine axis, because melatonin implanted directly in the pars tuberalis does not appear to modify the secretion of LH [6, 10]. In contrast, melatonin microimplants placed in the premammillary hypothalamic area (PMH) stimulated LH secretion [8]. Effective sites of melatonin microimplants within the PMH closely corresponded to the location of high melatonin-receptor binding within this area. As defined by receptor autoradiography, this binding area is delimited rostrally by the start of the infundibular recess of the third ventricle, dorsally and laterally by the fornix, caudally by the mammillary bodies, and ventrally by the base of the brain [8].

In rodents [1114], the region of the caudal hypothalamus corresponding to the PMH in sheep is complex and consists of several cell groups, including the premammillary nuclei and caudal portions of the arcuate nucleus (ARC). Therefore, to increase our understanding of the potential neural targets for melatonin in this area of the sheep brain, we have undertaken a detailed anatomical characterization of the PMH. In the present study, we used Nissl staining along with immunocytochemistry for NeuN and several neurochemical markers selected based on previous descriptions of their presence in the caudal hypothalamus of sheep and/or other species. These markers included tyrosine hydroxylase (TH) [15], dynorphin (DYN) [16, 17], estrogen receptor α (ERα) [18], orexin (ORX) [19], nitric oxide synthase (NOS) [20, 21], and cocaine- and amphetamine-regulated transcript peptide (CART) [13, 22, 23].

Materials and Methods

Animals

Adult, anestrous, Suffolk-crossbred ewes (n = 9) were maintained outdoors in an open barn until approximately 1 mo before they were killed at West Virginia University. They were moved indoors 2–3 days before ovariectomy, with the time of lights-on and lights-off changed biweekly to simulate natural changes in daylength. Ewes were fed a diet of silage supplemented with grain to maintain weight and had ad libitum access to water. All ewes were ovariectomized using sterile procedures and gas anesthesia (oxygen:nitrous oxide [3:1] + 1–3% halothane as needed). Approximately 2 wk after ovariectomy, an estradiol-filled silicone implant (length, 3 cm; inner diameter, 3.35 mm; outer diameter, 4.65 mm; Sil-med; Tri-anim, Sylmar, CA), which maintains circulating estradiol concentrations of approximately 1–2 pg/ml [24], was placed s.c., and animals were killed 1 wk later. All animal procedures were performed as approved by the West Virginia University Animal Care and Use Committee.

Perfusion and Tissue Processing

Ewes were heparinized (two i.v. injections of 25,000 U of heparin given 10 min apart) and killed by an overdose of sodium pentobarbital (∼2 g in 7 ml of saline i.v.), and the heads were perfused bilaterally via the carotid arteries with 6 L of 4% paraformaldehyde in 0.1 M phosphate buffer (PB, pH 7.3) with 0.1% sodium nitrite added as a vasodilator and 10 U heparin/ml fixative. After removal of the brains, hypothalami were dissected out, stored overnight in fixative at 4°C, and then infiltrated in 30% sucrose in phosphate buffer at 4°C. Thick (50-μm) coronal sections were cut on a freezing microtome and stored in a cryopreservative solution at −20°C [25] until processed for immunocytochemistry.

NeuN/Nissl Staining

Nuclear boundaries within the PMH were delineated by combining immunocytochemistry for NeuN, a commonly used neuronal marker [26], with Nissl staining of the same sections. A series of every fifth section through the caudal hypothalamus from five animals was used for the combined NeuN/Nissl staining. The NeuN immunocytochemistry was performed on free-floating sections. Before incubation with the primary antibody, sections were washed thoroughly, incubated in 1% hydrogen peroxide for 10 min to eliminate endogenous peroxidase activity, washed again, and incubated in 4% normal donkey serum (Jackson ImmunoResearch Laboratories, West Grove, PA) for 1 h. Sections were then incubated in a mouse monoclonal antibody against NeuN (1:1000; Chemicon, Temecula, CA) for 24 h at room temperature. Following washes, the sections were incubated in biotinylated donkey anti-mouse immunoglobulin (Ig) G (1:400; Jackson ImmunoResearch Laboratories) followed by incubation with an avidin-biotin-horseradish peroxidase conjugate for 1 h (1:400; ABC Elite Kit; Vector Laboratories, Burlingame, CA), and the reaction was visualized using 3,3′-diaminobenzidine (DAB; Sigma-Aldrich, St. Louis, MO) as the chromogen. All incubations and washes were performed at room temperature in 0.1 M PBS (pH 7.4) with 0.4% Triton X-100 (Fisher Scientific, Fair Lawn, NJ) added for incubations. Sections were mounted out of PB and dried overnight. For Nissl staining, slides were immersed in distilled, deionized H20 (5 min), then placed in acetate buffer (pH 3.4) for 30 sec and stained in 0.5% cresyl violet (Sigma-Aldrich) for 15 min. Next, the slides were dipped into acidified 70% ethanol; dehydrated through 70%, 95%, and 100% ethanol (5 min each); cleared in CitriSolv (twice for 10 min each time; Fisher Scientific); and cover- slipped with DPX mountant (BDH Laboratory Supplies, Poole, U.K.).

Immunocytochemistry

Full series of every fifth section through the PMH were processed for single-label immunocytochemical detection of TH (n = 2 ewes), DYN (n = 3 ewes), ERα (n = 3 ewes), CART (n = 2 ewes), NOS (n = 3 ewes), and ORX (n = 2 ewes). All incubations were performed at room temperature, and sections were washed thoroughly in PBS following each incubation step. Sections were incubated in 1% hydrogen peroxide in PBS for 10 min, then in a blocking solution including PBS plus 0.4% Triton X- 100 plus 4% normal donkey serum for 1 h. Overnight primary antibody incubation included either mouse monoclonal anti-TH (1:2000; Chemicon), mouse monoclonal anti-ERα (1:500; DAKO Corporation, Glostrup, Denmark), rabbit polyclonal anti-DYN A (1:20 000; Peninsula Laboratories, San Carlos, CA), rabbit polyclonal anti-ORX A (1:10 000; Abcam, Cambridge, U.K.), rabbit polyclonal anti-CART (1:40 000; Phoenix Pharmaceuticals, Belmont, CA), or rabbit polyclonal anti-nNOS (1:10 000; Chemicon) diluted in blocking solution. Sections were then incubated for 1 h in either biotinylated donkey anti-mouse IgG (1:400, Jackson ImmunoResearch Laboratories) for TH and ERα or biotinylated donkey anti-rabbit IgG (1:400; Jackson ImmunoResearch Laboratories) for DYN, ORX, CART, and NOS, then incubated in an avidin-biotin conjugate (1: 400) for 1 h and, finally, in DAB to visualize the reaction. Each series was mounted on slides, dehydrated in graded ethanol, cleared in CitriSolv (twice for 10 min each time), and cover-slipped using DPX mountant.

The antibodies used for TH [15, 27], DYN [17], ERα [28], and NOS [21] have been previously characterized for use in sheep brain. Controls for TH, ERα, DYN, ORX, CART, and NOS included omission of the primary antibodies, which eliminated all immunostaining. In addition, preabsorption of CART antibody with nanomolar concentrations of purified rat CART peptide (amino acids 55–102; Phoenix Pharmaceuticals) also completely eliminated all immunostaining.

Data Analysis

Sections were viewed under bright-field illumination. Images were captured using a digital camera (Magnafire; Optronics, Goleta, CA) attached to a Leica microscope (Deerfield, IL) and imported into Adobe Photoshop 7.0 (Adobe Systems, San Jose, CA). Images were not altered in any way except for minor adjustments of brightness and contrast.

Results

Nuclear Divisions of the PMH

The boundaries of the PMH, based on melatonin-receptor binding [8], are defined as the area delimited rostrally by the infundibular recess of the third ventricle, dorsally and laterally by the fornix, caudally by the mammillary bodies, and ventrally by the base of the brain. Analysis of NeuN/Nissl-stained sections revealed three nuclei within the PMH (Fig. 1): a caudal continuation of the hypothalamic arcuate nucleus (cARC), the ventral division of the premammillary nucleus (PMv), and the ventral tuberomammillary nucleus (TMv). The cARC is seen at rostral, mid, and caudal levels through the PMH, and it consists of densely packed, small- to medium-sized neurons (diameter, 8–15 μm) along the wall of the mammillary recess of the third ventricle and extends approximately 1.0 mm lateral to the ventricle (Fig. 1, A–D). The PMv has an ovoid shape and is located immediately ventromedial to the fornix at rostral to caudal levels through the PMH. The PMv consists of small- to medium-sized cell bodies (diameter, 8–17 μm) that often had an oval shape (Fig. 1, A–C and E). The TMv consisted of a thin plate of cells along the ventrolateral extent of the PMH, and was seen only at the mid and caudal levels (Fig. 1, B and C). The TMv cells were large (diameter, 10–30 μm) and either triangular or spindle-shaped (Fig. 1F).

Fig. 1

Photomicrographs of NeuN immunoreactivity and Nissl staining in coronal sections through the ewe PMH. A–C) Low-power (×5) images of rostral (A), mid (B), and caudal (C) levels of the PMH. Dashed lines indicate the three nuclear subdivisions of PMH: the cARC, the PMv, and the TMv. D–F) Higher-magnification (×20) images of NeuN-ir/Nissl-staining in cARC (D), PMv (E), and TMv (F). fx, Fornix; MR, mammillary recess; SUM, supramammillary nucleus. Bar = 500 μm (A–C) and 50 μm (D–F)

Fig. 1

Photomicrographs of NeuN immunoreactivity and Nissl staining in coronal sections through the ewe PMH. A–C) Low-power (×5) images of rostral (A), mid (B), and caudal (C) levels of the PMH. Dashed lines indicate the three nuclear subdivisions of PMH: the cARC, the PMv, and the TMv. D–F) Higher-magnification (×20) images of NeuN-ir/Nissl-staining in cARC (D), PMv (E), and TMv (F). fx, Fornix; MR, mammillary recess; SUM, supramammillary nucleus. Bar = 500 μm (A–C) and 50 μm (D–F)

Neurochemical Characterization of the PMH

TH-immunoreactive cells

TH-immunoreactive (ir) cells were identified by the presence of dense reaction product that filled their somas and dendrites. The greatest number of TH-ir cells and fibers were observed in the ventrolateral cARC at the rostral and mid levels through the PMH (Fig. 2, A and B). These cells belong to the A12 cell group [15]. A small number of TH-positive fibers were also present in the ventral tuberomammillary nucleus. In addition, a few scattered A10 TH-ir cells were identified within the PMH ventral to the fornix and lateral to the PMv (Fig. 2A).

Fig. 2

Photomicrographs of TH-, DYN-, and ERα-immunoreactivity in the ewe PMH. A and B) TH-ir cells and fibers at low magnification (×5; A) and higher magnification (×20; B) in the cARC. C and D) DYN-ir cells and fibers at low magnification (×5; C) and at higher magnification (×20; D) in the cARC. The arrow in D indicates an example of a DYN-ir fiber in close apposition to a DYN-ir cell body. E and F) ERα-ir cells at low magnification (×5; E) and higher magnification (×20; F) in the cARC. fx, Fornix; MR, mammillary recess. Bar = 200 μm (A, C, and E) and 20 μm (B, D, and F)

Fig. 2

Photomicrographs of TH-, DYN-, and ERα-immunoreactivity in the ewe PMH. A and B) TH-ir cells and fibers at low magnification (×5; A) and higher magnification (×20; B) in the cARC. C and D) DYN-ir cells and fibers at low magnification (×5; C) and at higher magnification (×20; D) in the cARC. The arrow in D indicates an example of a DYN-ir fiber in close apposition to a DYN-ir cell body. E and F) ERα-ir cells at low magnification (×5; E) and higher magnification (×20; F) in the cARC. fx, Fornix; MR, mammillary recess. Bar = 200 μm (A, C, and E) and 20 μm (B, D, and F)

DYN-ir cells

DYN-positive cell bodies were predominantly located in the ventrolateral and central portions of the cARC (Fig. 2, C and D) and were seen at all rostral to caudal levels through the PMH. The DYN-ir cells also extended into the ventromedial portion of the cARC, adjacent to the mammillary recess. In all these portions of the cARC, DYN cells were frequently seen in close apposition to DYN-positive fibers and varicosities (Fig. 2D, arrow), similar to those close contacts between DYN cells and fibers previously described in the rostral ARC [17].

ERα-ir cells

Nuclear ERα staining was present at all rostral to caudal levels through the PMH, predominantly in cells of the cARC (Fig. 2, E and F). Cells possessing ERα immunoreactivity were most numerous in the ventrolateral and central portions of the cARC, overlapping the position of TH and DYN cells (Fig. 2, A and C), but they were also present in all other parts of the cARC as defined by Nissl/NeuN staining (Fig. 1). A few scattered ERα cells were also seen in the lateral PMv ventral to the fornix (Fig. 2E).

CART-ir cells

Most CART-ir cells were densely packed within the ventrolateral portion of cARC throughout its full rostral to caudal extent (Fig. 3A). These cells had a spherical or ovoid shape, and they were relatively small (diameter, 10–15 μm) (Fig. 3B). Whereas dense CART-ir fibers were present in the medial, periventricular portions of the cARC, cells were not seen in this area. In addition to those cells in the cARC, scattered CART-ir cells were also seen in the PMv (Fig. 3, A and C); these cells were larger (diameter, 20–25 μm) than those in the cARC and generally had an ovoid shape (Fig. 3C).

Fig. 3

Photomicrographs of CART (A–C) and nNOS (D–F) immunoreactivity within the ewe PMH. Immunolabeling is shown at low magnification (×5; CART [A] and nNOS [D]) and higher magnification (×20) of the cARC (CART [B] and nNOS [E]) and PMv (CART [C] and nNOS [F]). fx, Fornix; MR, mammillary recess. Bar = 200 μm (A and D) and 20 μm (B, C, E, and F)

Fig. 3

Photomicrographs of CART (A–C) and nNOS (D–F) immunoreactivity within the ewe PMH. Immunolabeling is shown at low magnification (×5; CART [A] and nNOS [D]) and higher magnification (×20) of the cARC (CART [B] and nNOS [E]) and PMv (CART [C] and nNOS [F]). fx, Fornix; MR, mammillary recess. Bar = 200 μm (A and D) and 20 μm (B, C, E, and F)

NOS-ir cells

NOS-ir cells were found at all rostral to caudal levels through the PMH in both the cARC and PMv (Fig. 3D). In the cARC, NOS-positive cells were small (diameter, 8–10 μm), lightly stained, and located predominantly in the medial and ventromedial portions of the nucleus (Fig. 3, D and E). By contrast, NOS-ir cells in the PMv were larger (diameter, 10–15 μm) and more scattered throughout the nucleus (Fig. 3, D and F). NOS-positive cells were largely absent from the TMv.

ORX-ir fibers

ORX-ir cells were not seen in the PMH, although they were clearly present in other areas of the sheep hypothalamus, including the perifornical region (not shown). ORX-ir fibers were seen at all levels of the PMH but varied in their density (Fig. 4, A and B). The heaviest ORX fibers were seen in the TMv (Fig. 4B), while a more moderate density of ORX-ir fibers was observed in the PMv and the lateral portion of the cARC. By contrast, very few ORX-positive fibers were seen in the medial, periventricular region of the cARC (Fig. 4A).

Fig. 4

Photomicrographs of ORX immunoreactivity in the ewe PMH. A) Low magnification (×5) of the TMv. B) Higher magnification (×20) of the TMv. fx, Fornix; MR, mammillary recess. Bar = 200 μm (A) and 20 μm (B)

Fig. 4

Photomicrographs of ORX immunoreactivity in the ewe PMH. A) Low magnification (×5) of the TMv. B) Higher magnification (×20) of the TMv. fx, Fornix; MR, mammillary recess. Bar = 200 μm (A) and 20 μm (B)

Discussion

To our knowledge, the present study provides the first detailed description of the PMH in sheep. Based on NeuN immunolabeling and Nissl staining, we identified three nuclei—cARC, PMv, and TMv—within the melatonin-binding region of the ewe hypothalamus defined as the PMH [8]. Differences between these three nuclei in the morphological characteristics of individual cells, including their cell size and shape, provide further evidence that each is a distinct subdivision of the PMH. Using the phenotypic markers examined here, the majority of identified neuronal perikarya were within the cARC, whereas fewer perikarya were identified in the PMv, and only fiber labeling was identified in the TMv (Table 1). The absence of cell body labeling within the TMv must be viewed in light of the caveat that the present study was carried out without previous intracerebral colchicine injections to enhance immunolabeling of cell bodies. However, as discussed below, the distribution of neurochemical markers we observed in each PMH subdivision is largely in agreement with their localization in other species, based both on studies employing colchicine [29] and on those that did not [13, 15, 23, 29, 30]. Most of the phenotypic markers we used have been previously reported within the ARC of ewes [15, 1719, 21], but their location specifically within the cARC, which overlaps with the melatonin-binding region of PMH, has not been previously described. None of the phenotypic markers used here have been previously described within the PMv or TMv, nor has the localization of CART peptide been described previously in the sheep hypothalamus.

Table 1

Summary of the distribution of neurochemical markers within PMH nuclei.

Neuronal pheno-type Regions of PMH 
cARC PMv TMv 
TH Cell bodies and fibers None Few fibers 
DYN Cell bodies and fibers None None 
ERα Cell bodies Few cell bodies None 
CART Cell bodies and fibers Few cell bodies None 
NOS Cell bodies Cell bodies None 
ORX Few fibers Moderate fibers Dense fibers 
Neuronal pheno-type Regions of PMH 
cARC PMv TMv 
TH Cell bodies and fibers None Few fibers 
DYN Cell bodies and fibers None None 
ERα Cell bodies Few cell bodies None 
CART Cell bodies and fibers Few cell bodies None 
NOS Cell bodies Cell bodies None 
ORX Few fibers Moderate fibers Dense fibers 

Consistent with previous studies [15, 17, 21, 27], the cARC of anestrous ewes in the present study contained TH- ir (A12 cell group), ERα-ir, DYN-ir, and NOS-ir cells. The anatomical overlap of ERα-ir cells with TH- and DYN- containing neurons is consistent with previous reports that these cells contain gonadal steroid receptors [17, 27] and that the DYN neurons play an important role in steroid negative feedback in the ewe [17]. Numerous close contacts were also identified between DYN-ir boutons and DYN-ir cell bodies in the cARC of anestrous ewes, which is suggestive of the DYN-DYN synaptic contacts previously identified in the ARC of mid-luteal phase ewes [17]. The function of these reciprocal contacts is not known, but they may provide either local communication among DYN neurons within the ARC or be involved in communication with other DYN-containing brain regions, such as preoptic or anterior hypothalamic areas [17].

Similar neuronal phenotypes identified in cARC are also present within rostral and mid levels of the ARC, suggesting that cARC, although it lies within the melatonin-responsive region of the PMH, is similar neurochemically to the rest of the ARC. Beta-endorphin [31], neurokinin B [32], neuropeptide Y [33], and somatostatin [34] have also been observed in both the cARC and more rostral portions of this nucleus. In addition, heavily labeled CART-ir cell bodies and fibers were present in the cARC. This observation is in contrast to that of a previous study with sheep in which CART mRNA was not detected in the ARC [35]. However, the previous study used a probe specific to murine CART. More recently, CART mRNA has been detected in the ARC of both lactating and nonlactating ewes [36] using ovine probes, which is consistent with the presence of CART in the ARC of all other species studied to date [13, 22, 23, 3740].

The PMv nuclear division of the PMH contained cells that were immunoreactive for CART and NOS. The presence of CART cells within the PMv is consistent with previous reports in rats [13, 29], Siberian hamsters [30], rhesus monkeys [23], and humans [13]. Additionally, NOS has been reported in the PMv of rats [41] and Syrian hamsters [42]. A few DYN-ir cells were observed in the PMv of rats [16], whereas they were not observed in the PMv of sheep in the present study. However, DYN-ir neurons identified in the PMv of rats were very sparse and only observed following colchicine treatment. Furthermore, a more recent study failed to identify pro-DYN mRNA in the PMv of rats [13], suggesting that the presence of DYN cells may not be a reliable marker of this nucleus. Likewise, a few TH- ir cells have also been identified within the PMv of rats [43]. In the ewes of the present study, a few TH-ir cells of the A10 dopaminergic cell group [15] were present in the region of the PMH lateral to the PMv, but not within any well-defined nuclear boundaries.

None of the phenotypic markers examined in the present study identified the cells within the TMv, although dense ORX fibers were present in this nucleus, which is consistent with previous observations of ORX fibers in the posterior hypothalamus of ewes [19]. In the ewe, ORX-ir cells [19] and prepro-ORX mRNA [44] have been identified in the dorsomedial hypothalamus, lateral hypothalamic area, zona incerta, perifornical area, and posterior hypothalamic area. Thus, neurons in any of these sites could be the source of the ORX fibers observed in the ovine PMH. Recent evidence [45] suggests that histamine neurons may be present in the ovine PMH regions identified in the present study, although we did not examine histamine neurons. These histamine cells are part of a continuum with more caudal histamine cell groups present at the level of the mammillary bodies [45], similar to rats [46, 47] and humans [46, 48].

Recent evidence has pointed to the PMH as an important site for the influence of melatonin on seasonal reproduction. Melatonin microimplants into the PMH were effective at increasing LH concentrations in ewes housed under long daylengths [8], with effective sites of the implants closely corresponding to the location of high melatonin-receptor binding within this area. In addition, preliminary work using Fos as a marker of neuronal activation has indicated that cells in the PMH region exhibit a diurnal difference in Fos expression, with the number of Fos-expressing cells being higher during the day than during the night [49]. The nighttime decrease in Fos expression in the PMH, but not in other brain regions, such as the suprachiasmatic nucleus, was eliminated by pinealectomy [50], suggesting that the nighttime decrease in the PMH was caused by an inhibitory influence of high melatonin levels at that time. The precise location within the PMH of Fos-expressing neurons that show diurnal differences has not been examined. However, it is worth noting that [125I]melatonin binding in the PMH is reportedly higher in more lateral regions [8], where the PMv and TMv are located, than in the periventricular region, where the cARC is located. This suggests that the melatonin-responsive cells of the PMH are probably more concentrated within the PMv and/or TMv than in the cARC. Definitive studies regarding the location of melatonin-responsive cells of the PMH await cellular localization of the melatonin-receptor subtypes located in this brain region.

The neural substrates mediating seasonal reproduction in the ewe have been defined by anatomical, pharmacological, and lesion studies (for review, see [51]) and is comprised of a circuit that includes ERα-containing cells in the preoptic area, dopaminergic cells of the retrochiasmatic A15 cell group, and GnRH cell bodies located in a continuum that extends from the preoptic area to the mediobasal hypothalamus. Melatonin-responsive neurons in the three subdivisions of the PMH identified in the present study could potentially provide afferent input to any of these sites and, ultimately, influence the GnRH system. Of the neurochemicals we examined, CART-ir fibers have been reported to form close contacts with a majority of GnRH perikarya in the Siberian hamster [30], but whether these CART fibers arise from cells in the hamster ARC is not known. Similarly, histamine terminals have been observed to contact GnRH neurons in rats and humans [46], but it is not clear from which cell group of histaminergic neurons they arise. In rats, the PMv sends projections to periventricular areas of the anterior hypothalamus, including regions that contain ERα cells; however, input is notably lacking in regions containing GnRH neurons, such as medial preoptic area [52]. Future studies involving discrete injections of tract tracers into the three divisions of the PMH are necessary to determine which cells in this region may contribute to the control of seasonal changes in the activity of the ovine GnRH system.

Finally, it is worth considering the possibility that given the anatomical complexity of the PMH and the differences among its nuclear divisions, not all nuclei of the PMH are functionally involved in the regulation of seasonal reproduction. Alternatively, each of these three nuclei may have different roles in seasonal reproduction and respond to different signals that regulate the timing of seasonal reproductive transitions. For example, in addition to melatonin, thyroid hormones also serve as critical signals regulating seasonality in sheep and other species. Thyroid hormones are required for the seasonal reproductive transition to anestrus in ewes [53, 54]. Thyroid hormone receptor α-containing neurons are present in the PMH as well as in other regions of the ovine hypothalamus [55], but the precise localization of those neurons in the PMH has yet to be determined. In a recent study [56], thyroxine microimplants in the PMH of thyroidectomized ewes were sufficient to allow the transition to anestrus. Furthermore, thyroid hormones may be required for proper running of the circannual rhythm generator [57]. Thus, individual PMH nuclei and/or neurons may play more than one role in controlling seasonal breeding in ewes and even contain a component of the circannual rhythm mechanism.

In summary, the ovine PMH is a heterogeneous and complex region consisting of three nuclear subdivisions, each characterized by distinct neurochemical characteristics. The distribution of neuronal divisions and phenotypes described in the present study will serve as a roadmap for future work to determine the functional contribution of each PMH component to the influence of melatonin on seasonal reproduction.

Acknowledgments

The authors thank Maureen Fitzgerald for excellent technical assistance and Jon Williams and Dr. Chad Foradori for assistance with the NOS and DYN immunocytochemistry, respectively.

References

1.
Bronson
FH
,
Heideman
PD
.
Seasonal reproduction in mammals.
In:
Neill
JD
(ed.),
The Physiology of Reproduction
 ,
vol. 2
,
2nd ed
.
New York
:
Raven Press
;
1994
:
541
584
2.
Goodman
RL
.
Seasonal breeding in mammals.
In:
Neill
JD
(ed.),
The Encyclopedia of Reproduction
 .
New York
:
Academic Press
;
1998
:
341
351
3.
Bittman
EL
,
Karsch
FJ
.
Nightly duration of pineal melatonin secretion determines the reproductive response to inhibitory day length in the ewe.
Biol Reprod
 
1984
30
:
585
-
593
4.
Bittman
EL
,
Kaynard
AH
,
Olster
DH
,
Robinson
JE
,
Yellon
SM
,
Karsch
FJ
.
Pineal melatonin mediates photoperiodic control of pulsatile luteinizing hormone secretion in the ewe.
Neuroendocrinology
 
1985
40
:
409
-
418
5.
Malpaux
B
,
Daveau
A
,
Maurice
F
,
Gayrard
V
,
Thiery
JC
.
Short-day effects of melatonin on luteinizing hormone secretion in the ewe: evidence for central sites of action in the mediobasal hypothalamus.
Biol Reprod
 
1993
48
:
752
-
760
6.
Malpaux
B
,
Daveau
A
,
Maurice
F
,
Locatelli
A
,
Thiery
JC
.
Evidence that melatonin-binding sites in the pars tuberalis do not mediate the photoperiodic actions of melatonin on LH and prolactin secretion in ewes.
J Reprod Fertil
 
1994
101
:
625
-
632
7.
Bittman
EL
,
Weaver
DR
.
The distribution of melatonin-binding sites in neuroendocrine tissues of the ewe.
Biol Reprod
 
1990
43
:
986
-
993
8.
Malpaux
B
,
Daveau
A
,
Maurice-Mandon
F
,
Duarte
G
,
Chemineau
P
.
Evidence that melatonin acts in the premammillary hypothalamic area to control reproduction in the ewe: presence of binding sites and stimulation of luteinizing hormone secretion by in situ microimplant delivery.
Endocrinology
 
1998
139
:
1508
-
1516
9.
Lincoln
G
.
Melatonin modulation of prolactin and gonadotrophin secretion. Systems ancient and modern.
Adv Exp Med Biol
 
1999
460
:
137
-
153
10.
Malpaux
B
,
Skinner
DC
,
Maurice
F
.
The ovine pars tuberalis does not appear to be targeted by melatonin to modulate luteinizing hormone secretion, but may be important for prolactin release.
J Neuroendocrinol
 
1995
7
:
199
-
206
11.
Airaksinen
MS
,
Alanen
S
,
Szabat
E
,
Visser
TJ
,
Panula
P
.
Multiple neurotransmitters in the tuberomammillary nucleus: comparison of rat, mouse, and guinea pig.
J Comp Neurol
 
1992
323
:
103
-
116
12.
Ericson
H
,
Watanabe
T
,
Kohler
C
.
Morphological analysis of the tuberomammillary nucleus in the rat brain: delineation of subgroups with antibody against l-histidine decarboxylase as a marker.
J Comp Neurol
 
1987
263
:
1
-
24
13.
Elias
CF
,
Lee
CE
,
Kelly
JF
,
Ahima
RS
,
Kuhar
M
,
Saper
CB
,
Elmquist
JK
.
Characterization of CART neurons in the rat and human hypothalamus.
J Comp Neurol
 
2001
432
:
1
-
19
14.
Swanson
L
.
Brain Maps: Structure of the Rat Brain.
Amsterdam, The Netherlands
:
Elsevier Science Publishers B.V.
;
1992
15.
Tillet
Y
,
Thibault
J
.
Catecholamine-containing neurons in the sheep brainstem and diencephalon: immunohistochemical study with tyrosine hydroxylase (TH) and dopamine-β-hydroxylase (DBH) antibodies.
J Comp Neurol
 
1989
290
:
69
-
104
16.
Fallon
JH
,
Leslie
FM
.
Distribution of dynorphin and enkephalin peptides in the rat brain.
J Comp Neurol
 
1986
249
:
293
-
336
17.
Foradori
CD
,
Coolen
LM
,
Fitzgerald
ME
,
Skinner
DC
,
Goodman
RL
,
Lehman
MN
.
Colocalization of progesterone receptors in parvicellular dynorphin neurons of the ovine preoptic area and hypothalamus.
Endocrinology
 
2002
143
:
4366
-
4374
18.
Lehman
MN
,
Ebling
FJ
,
Moenter
SM
,
Karsch
FJ
.
Distribution of estrogen receptor-immunoreactive cells in the sheep brain.
Endocrinology
 
1993
133
:
876
-
886
19.
Iqbal
J
,
Pompolo
S
,
Sakurai
T
,
Clarke
IJ
.
Evidence that orexin-containing neurones provide direct input to gonadotropin-releasing hormone neurones in the ovine hypothalamus.
J Neuroendocrinol
 
2001
13
:
1033
-
1041
20.
Bhat
GK
,
Mahesh
VB
,
Lamar
CA
,
Ping
L
,
Aguan
K
,
Brann
DW
.
Histochemical localization of nitric oxide neurons in the hypothalamus: association with gonadotropin-releasing hormone neurons and colocalization with N-methyl-d-aspartate receptors.
Neuroendocrinology
 
1995
62
:
187
-
197
21.
Dufourny
L
,
Skinner
DC
.
Influence of estradiol on NADPH diaphorase/neuronal nitric oxide synthase activity and colocalization with progesterone or type II glucocorticoid receptors in ovine hypothalamus.
Biol Reprod
 
2002
67
:
829
-
836
22.
Koylu
EO
,
Couceyro
PR
,
Lambert
PD
,
Kuhar
MJ
.
Cocaine- and amphetamine-regulated transcript peptide immunohistochemical localization in the rat brain.
J Comp Neurol
 
1998
391
:
115
-
132
23.
Dall Vechia
S
,
Lambert
PD
,
Couceyro
PC
,
Kuhar
MJ
,
Smith
Y
.
CART peptide immunoreactivity in the hypothalamus and pituitary in monkeys: analysis of ultrastructural features and synaptic connections in the paraventricular nucleus.
J Comp Neurol
 
2000
416
:
291
-
308
24.
Legan
SJ
,
Karsch
FJ
,
Foster
DL
.
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
 
1977
101
:
818
-
824
25.
Watson RE
Jr
,
Wiegand
SJ
,
Clough
RW
,
Hoffman
GE
.
Use of cryoprotectant to maintain long-term peptide immunoreactivity and tissue morphology.
Peptides
 
1986
7
:
155
-
159
26.
Mullen
RJ
,
Buck
CR
,
Smith
AM
.
NeuN, a neuronal specific nuclear protein in vertebrates.
Development
 
1992
116
:
201
-
211
27.
Lehman
MN
,
Karsch
FJ
.
Do gonadotropin-releasing hormone, tyrosine hydroxylase-, and β-endorphin-immunoreactive neurons contain estrogen receptors? A double-label immunocytochemical study in the Suffolk ewe.
Endocrinology
 
1993
133
:
887
-
895
28.
Scott
CJ
,
Rawson
JA
,
Pereira
AM
,
Clarke
IJ
.
The distribution of estrogen receptors in the brainstem of female sheep.
Neurosci Lett
 
1998
241
:
29
-
32
29.
Koylu
EO
,
Couceyro
PR
,
Lambert
PD
,
Ling
NC
,
DeSouza
EB
,
Kuhar
MJ
.
Immunohistochemical localization of novel CART peptides in rat hypothalamus, pituitary and adrenal gland.
J Neuroendocrinol
 
1997
9
:
823
-
833
30.
Leslie
RA
,
Sanders
SJ
,
Anderson
SI
,
Schuhler
S
,
Horan
TL
,
Ebling
FJ
.
Appositions between cocaine and amphetamine-related transcript- and gonadotropin releasing hormone-immunoreactive neurons in the hypothalamus of the Siberian hamster.
Neurosci Lett
 
2001
314
:
111
-
114
31.
Skinner
DC
,
Herbison
AE
.
Effects of photoperiod on estrogen receptor, tyrosine hydroxylase, neuropeptide Y, and β-endorphin immunoreactivity in the ewe hypothalamus.
Endocrinology
 
1997
138
:
2585
-
2595
32.
Goubillon
ML
,
Forsdike
RA
,
Robinson
JE
,
Ciofi
P
,
Caraty
A
,
Herbison
AE
.
Identification of neurokinin B-expressing neurons as an highly estrogen-receptive, sexually dimorphic cell group in the ovine arcuate nucleus.
Endocrinology
 
2000
141
:
4218
-
4225
33.
Chaillou
E
,
Baumont
R
,
Chilliard
Y
,
Tillet
Y
.
Several subpopulations of neuropeptide Y-containing neurons exist in the infundibular nucleus of sheep: an immunohistochemical study of animals on different diets.
J Comp Neurol
 
2002
444
:
129
-
143
34.
Scanlan
N
,
Dufourny
L
,
Skinner
DC
.
Somatostatin-14 neurons in the ovine hypothalamus: colocalization with estrogen receptor α and somatostatin-28 (1–12) immunoreactivity, and activation in response to estradiol.
Biol Reprod
 
2003
69
:
1318
-
1324
35.
Henry
BA
,
Rao
A
,
Ikenasio
BA
,
Mountjoy
KG
,
Tilbrook
AJ
,
Clarke
IJ
.
Differential expression of cocaine- and amphetamine-regulated transcript and agouti-related protein in chronically food-restricted sheep.
Brain Res
 
2001
918
:
40
-
50
36.
Sorensen
A
,
Adam
CL
,
Findlay
PA
,
Marie
M
,
Thomas
L
,
Travers
MT
,
Vernon
RG
.
Leptin secretion and hypothalamic neuropeptide and receptor gene expression in sheep.
Am J Physiol Regul Integr Comp Physiol
 
2002
282
:
R1227
-
R1235
37.
Vrang
N
,
Larsen
PJ
,
Clausen
JT
,
Kristensen
P
.
Neurochemical characterization of hypothalamic cocaine-amphetamine-regulated transcript neurons.
J Neurosci
 
1999
19
:
RC51
-
8
38.
Hurd
YL
,
Fagergren
P
.
Human cocaine- and amphetamine-regulated transcript (CART) mRNA is highly expressed in limbic- and sensory- related brain regions.
J Comp Neurol
 
2000
425
:
583
-
598
39.
Adam
CL
,
Moar
KM
,
Logie
TJ
,
Ross
AW
,
Barrett
P
,
Morgan
PJ
,
Mercer
JG
.
Photoperiod regulates growth, puberty and hypothalamic neuropeptide and receptor gene expression in female Siberian hamsters.
Endocrinology
 
2000
141
:
4349
-
4356
40.
Johansen
JE
,
Broberger
C
,
Lavebratt
C
,
Johansson
C
,
Kuhar
MJ
,
Hokfelt
T
,
Schalling
M
.
Hypothalamic CART and serum leptin levels are reduced in the anorectic (anx/anx) mouse.
Mol Brain Res
 
2000
84
:
97
-
105
41.
Yokosuka
M
,
Hayashi
S
.
Colocalization of neuronal nitric oxide synthase and androgen receptor immunoreactivity in the premammillary nucleus in rats.
Neurosci Res
 
1996
26
:
309
-
314
42.
Hadeishi
Y
,
Wood
RI
.
Nitric oxide synthase in mating behavior circuitry of male Syrian hamster brain.
J Neurobiol
 
1996
30
:
480
-
492
43.
Gonzalo-Ruiz
A
,
Alonso
A
,
Sanz
JM
,
Llinas
RR
.
A dopaminergic projection to the rat mammillary nuclei demonstrated by retrograde transport of wheat germ agglutinin-horseradish peroxidase and tyrosine hydroxylase immunohistochemistry.
J Comp Neurol
 
1992
321
:
300
-
311
44.
Iqbal
J
,
Henry
BA
,
Pompolo
S
,
Rao
A
,
Clarke
IJ
.
Long-term alteration in bodyweight and food restriction does not affect the gene expression of either prepro-orexin or prodynorphin in the sheep.
Neuroscience
 
2003
118
:
217
-
226
45.
Tillet
Y
,
Batailler
M
,
Panula
P
.
Histaminergic neurons in the sheep diencephalon.
J Comp Neurol
 
1998
400
:
317
-
333
46.
Fekete
CS
,
Strutton
PH
,
Cagampang
FR
,
Hrabovszky
E
,
Kallo
I
,
Shughrue
PJ
,
Dobo
E
,
Mihaly
E
,
Baranyi
L
,
Okada
H
,
Panula
P
,
Merchenthaler
I
,
Coen
CW
,
Liposits
ZS
.
Estrogen receptor immunoreactivity is present in the majority of central histaminergic neurons: evidence for a new neuroendocrine pathway associated with luteinizing hormone-releasing hormone-synthesizing neurons in rats and humans.
Endocrinology
 
1999
140
:
4335
-
4341
47.
Panula
P
,
Yang
HY
,
Costa
E
.
Histamine-containing neurons in the rat hypothalamus.
Proc Natl Acad Sci U S A
 
1984
81
:
2572
-
2576
48.
Panula
P
,
Airaksinen
MS
,
Pirvola
U
,
Kotilainen
E
.
A histamine-containing neuronal system in human brain.
Neuroscience
 
1990
34
:
127
-
132
49.
Daveau
A
,
Chemineau
P
,
Malpaux
B
.
Day-night rhythm in c-Fos expression in the premammillary hypothalamic area of the ewe.
Society for Neuroscience Abstracts
 
1999
25
:
870
50.
Malpaux
B
,
Daveau
A
,
Chemineau
P
.
Day-night changes in c-fos expression in the pre-mammillary hypothalamic area of the ewe are pineal dependent.
Eur J Neurosci
 
2000
12
:
421
51.
Lehman
MN
,
Coolen
LM
,
Goodman
RL
,
Viguie
C
,
Billings
HJ
,
Karsch
FJ
.
Seasonal plasticity in the brain: the use of large animal models for neuroanatomical research.
Reprod Suppl
 
2002
59
:
149
-
165
52.
Canteras
NS
,
Simerly
RB
,
Swanson
LW
.
Projections of the ventral premammillary nucleus.
J Comp Neurol
 
1992
324
:
195
-
212
53.
Moenter
SM
,
Woodfill
CJ
,
Karsch
FJ
.
Role of the thyroid gland in seasonal reproduction: thyroidectomy blocks seasonal suppression of reproductive neuroendocrine activity in ewes.
Endocrinology
 
1991
128
:
1337
-
1344
54.
Webster
JR
,
Moenter
SM
,
Woodfill
CJ
,
Karsch
FJ
.
Role of the thyroid gland in seasonal reproduction. II. Thyroxine allows a season-specific suppression of gonadotropin secretion in sheep.
Endocrinology
 
1991
129
:
176
-
183
55.
Jansen
HT
,
Lubbers
LS
,
Macchia
E
,
DeGroot
LJ
,
Lehman
MN
.
Thyroid hormone receptor (α) distribution in hamster and sheep brain: colocalization in gonadotropin-releasing hormone and other identified neurons.
Endocrinology
 
1997
138
:
5039
-
5047
56.
Anderson
GM
,
Hardy
SL
,
Valent
M
,
Billings
HJ
,
Connors
JM
,
Goodman
RL
.
Evidence that thyroid hormones act in the ventromedial preoptic area and the premammillary region of the brain to allow the termination of the breeding season in the ewe.
Endocrinology
 
2003
144
:
2892
-
2901
57.
Billings
HJ
,
Viguie
C
,
Karsch
FJ
,
Goodman
RL
,
Connors
JM
,
Anderson
GM
.
Temporal requirements of thyroid hormones for seasonal changes in LH secretion.
Endocrinology
 
2002
143
:
2618
-
2625

Author notes

1
Supported by NIH R01 HD17864 and USDA 2001-02264 to R.L.G. Preliminary reports have appeared in Society for Neuroscience, 2002, Abstract 572.6.