Efferent and afferent connections of the ventromedial hypothalamic nucleus determined by neural tracer analysis: implications for lordosis regulation in female rats.

G Model NSR-3761; No. of Pages 15

ARTICLE IN PRESS Neuroscience Research xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Neuroscience Research journal homepage: www.elsevier.com/locate/neures

Efferent and afferent connections of the ventromedial hypothalamic nucleus determined by neural tracer analysis: Implications for lordosis regulation in female rats Yuji Shimogawa a , Yasuo Sakuma b , Korehito Yamanouchi a,∗ a Neuroendocrinology, Department of Human Behavior and Environment Sciences, School of Human Sciences, Waseda University, 2-579-15, Mikajima, Tokorozawa, Saitama 359-1192, Japan b University of Tokyo Health Science, 4-11, Ochiai, Tama, Tokyo 206-0033, Japan

a r t i c l e

i n f o

Article history: Received 4 September 2014 Received in revised form 8 October 2014 Accepted 11 October 2014 Available online xxx Keywords: Ventromedial hypothalamic nucleus Neural connection Anterograde neural tracer Retrograde neural tracer Female rat Lordosis

a b s t r a c t Neural connections of the ventromedial hypothalamic nucleus (VMN) to and from forebrain and midbrain structures, which are involved in the neuroendocrine regulation of reproduction, were investigated. A retrograde (fluoro-gold [FG]) or an anterograde neural tracer (phaseolus vulgaris-leucoagglutinin [PHAL]) was injected into the left side of the VMN in ovariectomized rats. Six days after injection with FG or 11 days after injection with PHA-L, brains were fixed and sectioned. After immunohistochemistry, digital images of FG-labeled neural cell bodies (FG-cells) or PHA-L-labeled fibers (PHA-L-fibers) were analyzed. Injection sites of FG and PHA-L were mainly in the ventrolateral VMN. Considerable numbers of FG-cells and PHA-L-fibers were present in the left side of the medial amygdala, ventral lateral septum, preoptic area, bed nucleus of stria terminalis, dorsomedial hypothalamic nucleus, arcuate nucleus, periventricular nucleus of thalamus, and midbrain central gray. The lateral dorsal raphe nuclei contained many PHA-Lfibers but few FG-cells. By contrast, both sides of the median raphe nucleus contained many FG-cells but few PHA-L-fibers. Reciprocal direct neural connection between the right and left side of the VMN were observed. The present results provide an anatomical basis for functional relationships between the VMN and these nuclei. © 2014 Elsevier Ireland Ltd and the Japan Neuroscience Society. All rights reserved.

1. Introduction The ventromedial hypothalamic nucleus (VMN) consists of dorsomedial, central, and ventrolateral subdivisions. It is located just lateral to the arcuate nucleus from the middle to caudal level of the hypothalamus (Paxinos and Watson, 2007). The VMN is surrounded by an area containing dendrites of VMN neurons and afferent projections but is lacking in cells (Millhouse, 1973). Projections of the rat VMN have been reported using autoradiographic method (Saper et al., 1976) or using phaseolus vulgaris-leucoagglutinin (PHA-L) tracing (Canteras et al., 1994). Subdivisions of the VMN exhibit different projection patterns (for review, see Bleier and Byne, 1985). A previous electrophysiological study has shown that neural connections between the VMN and brain regions such as the medial preoptic area (POA), stria terminalis, amygdale (AMG),

∗ Corresponding author. Tel.: +81 4 2947 6727; fax: +81 4 2947 6727. E-mail address: [email protected] (K. Yamanouchi). URL: http://www.f.waseda.jp/hedgehog/ (K. Yamanouchi).

anterior hypothalamic nucleus (AH), midbrain central gray (MCG), and median eminence are important for neuroendocrine functions of rats (Renaud and Martin, 1975). The VMN plays important roles in reproductive behaviors, such as female sexual behavior, male sexual behavior and aggressive behavior. The VMN is a key nucleus in the regulation of female sexual behavior, lordosis (Pfaff et al., 2006). The VMN plays a facilitative role in regulating lordosis because destruction of this nucleus inhibits lordosis; conversely, electrical stimulation facilitates it (Pfaff and Sakuma, 1979a, b). The ventrolateral (vl) VMN shows high expression of estrogen receptor (ER) mRNA (Simerly et al., 1990) and has many ER␣-positive neurons (Yamada et al., 2009). Direct application of estrogen into the VMN facilitates estrous behavior in ovariectomized rats (Barfield and Chen, 1977; Rajendren et al., 1991). Thus, the VMN is an important nucleus for induction of lordosis, i.e., induction of estrous states by estrogen in female rats. The neural systems that regulate lordosis are complicated and involve many neural substrates. The VMN is thought to facilitate lordosis behavior through the function of the MCG, a center for lordosis regulation, because destruction of the MCG prevents the

http://dx.doi.org/10.1016/j.neures.2014.10.016 0168-0102/© 2014 Elsevier Ireland Ltd and the Japan Neuroscience Society. All rights reserved.

Please cite this article in press as: Shimogawa, Y., et al., Efferent and afferent connections of the ventromedial hypothalamic nucleus determined by neural tracer analysis: Implications for lordosis regulation in female rats. Neurosci. Res. (2014), http://dx.doi.org/10.1016/j.neures.2014.10.016


ARTICLE IN PRESS Y. Shimogawa et al. / Neuroscience Research xxx (2014) xxx–xxx

2

lordosis facilitating effects of electrical stimulation of the VMN (Sakuma and Pfaff, 1979). A direct projection from the vlVMN to the MCG has been reported to send facilitative signals for lordosis. The nature of these axons is sexually dimorphic (Sakuma and Pfaff, 1981). The connection between the VMN and the MCG is the most critical neural tract for induction of estrous states. In the telencephalon, the lateral septum (LS) exerts a lordosisinhibiting influence, since destruction of this nucleus (Nance et al., 1974; Kondo et al., 1990) or injection of ibotenic acid into the LS (Tsukahara and Yamanouchi, 2001) facilitates lordosis behavior. The medial and lateral amygdala have facilitative and inhibitory influences, respectively (Masco and Carrer, 1980). The stria terminalis is also involved in lordosis regulation (Takeo et al., 1995). In the diencephalon, in addition to the VMN, many nuclei regulate lordosis behavior. The POA also plays an inhibitory role in regulating lordosis (Powers and Valenstein, 1972; Takeo et al., 1993; Sakuma, 1994). The habenula plays a facilitative role in lordosis regulation (Masco and Carrer, 1980). In the midbrain, in addition to the MCG, serotonergic neurons of the dorsal raphe nucleus (DR) are an important inhibitory regulator of lordosis (Kakeyama and Yamanouchi, 1996). The reticular nucleus in the midbrain is also involved in the control of the MCG (Pfaff, 1980). These regions may influence the function of the VMN directly or indirectly. However, the nuclei that send axons to the vlVMN and efferent pathways from the vlVMN to the nuclei involved in regulation of lordosis are not clear. In addition, recently, the right and left VMN has been found to influence the ER␣ expression in each region, because in ovariectomized rats, lesions in the right or left VMN increase ER␣ expression in the contralateral side (Shimogawa et al., 2014). In the present experiment, to clarify the efferent and afferent neural connections and potential bilateral connections of the VMN that are involved in lordosis regulation, anterograde and retrograde neural tracers were injected into the left side of the VMN, specifically into the ventrolateral region, and histochemical analyses were performed from the telencephalon to the midbrain in ovariectomized rats. 2. Materials and methods 2.1. Animals Seven-week-old female Wistar rats were purchased (Takasugi Experimental Animal, Saitama, Japan) and maintained on a controlled light-dark cycle (14L:10D, lights off at 19:00) with constant temperature (22–24 ◦ C). Food and water were freely accessible. All experiments were conducted according to the regulations for Animal Experimentation at Waseda University (Approval No. 2011A003, 2012-A004, 2013-A037). One week after purchase, all animals were ovariectomized under isoflurane anesthesia to eliminate the influence of sex steroids. A retrograde neural tracer, fluoro-gold (FG: Biotium, Inc. Hayward, CA, USA), or an anterograde neural tracer, phaseolus vulgaris-leucoagglutinin (PHA-L: Vector Laboratories, Inc. Burlingame, CA, USA), was then injected into the left side of the VMN, specifically, the ventrolateral part to clarify neural connections and tracts involved in the regulation of lordosis behavior.

9.8 mm below, and 0.8 mm left of the bregma. An alternating current of 2 ␮A was applied for 8–14 min using a Midgard Precision Current Source (Stoelting-Muromachi Kikai, Tokyo, Japan). Six days after FG injection, animals were deeply anesthetized with pentobarbital sodium solution (50 mg/kg body weight, Somnopentyl, Kyoritsu Seiyaku, Tokyo) and then perfused intracardially with icecold 50 mM phosphate-buffered saline (PBS, pH 7.4) followed by 4% paraformaldehyde–50 mM phosphate buffer. Brains were removed and post-fixed overnight with the same fixative used for perfusion. Brains were further immersed in 30% sucrose–50 mM PBS for 5 days at 4 ◦ C. Serial coronal brain sections (50-␮m thickness) were taken with a cryostat and collected. Sections mounted on slides were dehydrated through a graded series of ethanol, cleared by xylene, and then coverslipped with marinol. Brain sections were observed under a fluorescence microscope (U-LH100HGAPO, Olympus, Tokyo, Japan). The FG injection site and the distribution of FG-labeled neuronal cells were plotted onto the rat brain map of Paxinos and Watson (2007). Sections were observed under stimulation with a UV excitation filter and whitecolored cells were identified as FG cells. Digital images of FG cells were obtained with a Polaroid Digital Microscope Camera PDMCII/OL (Olympus) and stored in a computer. 2.3. Injection of anterograde tracer One week after the ovariectomy, PHA-L (2.5% in sodium phosphate buffer, pH 8.0) was injected iontophoretically into the left VMN through a glass micropipette (tip diameter 10–20 ␮m). Rats were fixed in a stereotaxic instrument in which the incisor bar was set 3.3 mm below the interaural line under isoflurane anesthesia. The tip of the micropipette was lowered to a point 2.5 mm caudal to the bregma, 9.8 mm below the bregma, and 0.8 mm left of the bregma according to a rat brain atlas. An alternating current of 5 ␮A was applied for 18–20 min using a Midgard Precision Current Source (Stoelting-Muromachi Kikai). Eleven days after PHA-L injection, coronal brain sections (50-␮m thickness) were obtained by the process same as above. For PHA-L immunostaining, free-floating sections were incubated with 0.6% H2 O2 –50 mM PBS for 30 min at room temperature (RT) before and after rinsing thrice with 50 mM PBS for 10 min. Sections were then incubated with 5% normal goat serum (NGS, Chemicon, CA, USA)–0.1% Triton X-100–50 mM PBS at RT for 90 min, and then incubated with rabbit anti-PHA-L (1:300, Vector Laboratories, Inc.)–0.1% Triton X-100–50 mM PBS for 72 h at 4 ◦ C. After washing thrice with 50 mM Tris–HCl buffered saline (TBS) for 10 min, sections were reacted with Envision Plus System-HRP (Dako, Glostrup, Denmark) for 30 min at RT. After rinsing thrice with 50 mM TBS for 10 min, sections were reacted with a Metal Enhanced DAB Substrate Kit (34065, Thermo Fisher Scientific, IL, USA) for visualization of PHA-L immunoreactivity. Immunostained sections mounted on slides were dehydrated through a graded series of ethanol, cleared by xylene, and then coverslipped with marinol. These sections were examined by light microscopy and digital images were stored on a computer. The PHA-L injection site and the distribution of anterograde tracer-labeled fibers and cells were traced onto the rat brain atlas of Paxinos and Watson (2007).

2.2. Injection of retrograde tracer 3. Results One week after the ovariectomy, FG (8% solution dissolved in distilled water) was injected iontophoretically into the left VMN through a glass micropipette (tip diameter 40–50 ␮m). Rats were fixed in a stereotaxic instrument in which the incisor bar was set 3.3 mm below the interaural line under isoflurane anesthesia. The tip of the micropipette was lowered to a point 2.5 mm caudal to,

3.1. FG-infusion area: retrograde labeling In all 11 rats, fluorescence was observed in the right side of the VMH between 1.89 and 3.36 mm posterior to the bregma, and the injection site extended around a 1.2-mm area in the




3

Fig. 1. Reciprocal connections between the right and left side of the ventromedial hypothalamic nucleus (VMN) in female rats. (A–C) Representative photomicrograph of neural cells with fluorescence induced by UV excitation in the VMN in an ovariectomized rat that was injected with fluoro-gold (FG) into the ventrolateral (vl) VMN. (A) FG-labeled cells in the contralateral VMN (right side). (B) FG-injection site in the left vlVMN. (C) High-magnification view of the right side of the VMN in (A). Bars indicate 200 ␮m. (D) Schematic illustrations showing left and right connections of the vlVMN from the present results of the FG injection (solid line) and PHA-L injection (dashed line). FG-labeled cells were mainly present in the right side of the vlVMN. PHA-L fibers were observed in the vl and central VMN at a similar level, but with a few PHA-L fibers present in the dorsolateral VMN. (E–H) Representative photomicrographs of the VMN of an ovariectomized rat injected with phaseolus vulgaris-leucoagglutinin (PHA-L) into the left side of the vlVMN. (E) PHA-L-labeled axons in the right VMN (contralateral to the injection). (F) PHA-L injection site in the left vlVMN. (G and H) High-magnification views of the right side of the VMN in (E). Bars indicate 200 ␮m.




4

antero-posterior axis in each rat (Figs. 1B, 2J–K). Strong fluorescence was observed around the injection site (∼2.5 mm posterior to the bregma). In all rats, FG penetrated mainly to the vlVMN and parts of the central VMN; some areas of the shell and limited regions of the lateral hypothalamus were FG positive. A limited region of the dorsomedial VMN was penetrated by FG in 3 rats. In another 3 rats, fluorescence was limited to the vlVMN. The arcuate nucleus was generally not FG positive except 2 rats.

large numbers of FG cells were observed in all anterior and posterior nuclei but not in lateral nuclei of the left side (Figs. 2D–G, 3). On the right side, no FG cells were present in the BNST including the lateral nuclei. Accumbens nucleus (ACCB): a few FG cells were present in left-sided regions of the shell. On the right side, no FG cells were observed (Fig. 2B–D).

3.2. FG-labeled cells (FG cells)

In 4 of 6 rats, the PHA-L injection site was mainly located in the right side of the VMH corresponding to an area 2.04–3.24 mm posterior to the bregma (Paxinos and Watson, 2007) (Fig. 1F). In 2 of 6 rats, dense PHA-L positive areas were present in the rostral VMH (1.54–2.64 mm posterior to the bregma). In all rats, most regions of the vlVMN contained dense PHA-L positive cells and axons. There were no PHA-L positive cells in the arcuate nucleus of all rats. In 3 rats, PHA-L positive cells were noted in the vlVMN and in limited regions of the lateral hypothalamus. In 3 rats, the vlVMN and central VMN and limited regions of the lateral hypothalamus contained PHA-L positive cells and axons.

FG cells were observed in the left side (injection side) of most nuclei or areas examined (Table 1, Figs. 2 and 3). A few FG cells were visualized in the right side. 3.2.1. Diencephalon 3.2.1.1. Hypothalamus. VMN: on the right side (contralateral side of the injection), many FG cells were observed in the ventrolateral and central regions, but a few FG cells were noted in the dorsomedial VMN in most rats (Fig. 1A, 1C, 2I–K). Dorsomedial hypothalamic nucleus (DMN): Many FG cells were observed on the left side in all rats but no FG cells were observed on the right side (Fig. 2J and K). Arcuate nucleus (ARCN): on the left side, many FG cells were observed especially in posterior regions. In the anterior region of the left side, a few FG cells were noted but only in 5 rats (Figs. 1A and B, 2I–K). On the right side, a few FG cells were observed in only the anterior regions. AH: Many FG cells were scattered in this area on the left side in all rats but not on the right side (Fig. 2H and I). Paraventricular nucleus (PVN), supraoptic nucleus (SON): FG cells were noted in the left side but in limited numbers and no FG cells were observed on the right side (Fig. 2G–I). Suprachiasmatic nucleus (SCN): on either side, a few FG cells were noted in all rats (Figs. 2G, 3) (Fig. 3). POA: on either side, many FG cells were observed, an especially large number of FG cells were observed on the left side in all rats (Figs. 2E–G, 3). Anteroventral periventricular nucleus of the POA (AVPV): Many FG cells were observed on both sides in all rats (Figs. 2E, 3). Medial mammillary nucleus (MM), lateral mammillary nucleus (LM), supramammillary nucleus (SM): a few FG cells were noted on the left side but not on the right side (Fig. 2L). 3.2.1.2. Thalamus. The periventricular nucleus (PV) of the thalamus: a few FG cells were observed partially in the left side (Fig. 2H–K). Habenular nuclei (HB): in both the medial and lateral habenular, no FG cells were observed (Fig. 2J–K). 3.2.1.3. Midbrain. Interpeduncular nucleus (IP): no FG cells were present on either side (Fig. 2M and N). MCG: in the rostral region, a few FG cells were observed on the left side (Figs. 2L–P, 3). DR (Figs. 2O and P, 3): lateral areas contained many FG cells on the left side and a few cells were observed on the right side. In contrast, a few or no FG cells were noted in the dorsal or ventral regions on both sides. Median raphe nucleus (MR) (Fig. 2O and P): only a few FG cells were present on both sides in all rats. 3.2.1.4. Telencephalon. In the neocortex, including the cingulate cortex, no FG cells were observed. No FG cells were noted on both sides of the hippocampus. LS: on the left side, many FG cells were observed in the ventral region and were present in the intermediate region close to the ventral region (Figs. 2A–F, 3). On the right side of the ventral LS, a few FG cells were noted. In the dorsal LS, a few FG cells were observed but only on the left side. Medial septum (MS): no FG cells were present. Medial amygdala (MAMG): large numbers of FG cells were observed in the left and many cells in the right side (Figs. 2F–M, 3). LAMG: A few FG cells were present on the left side in 5 rats (Figs. 2F–M, 3). Bed nucleus of stria terminalis (BNST):

3.3. PHA-L infusion area: anterograde labeling

3.4. PHA-L positive axons (PHA-L axons) PHA-L axons were observed on both sides of most nuclei or areas examined, but greater numbers were noted on the left side (injection side) than on the right side (Table 1, Figs. 4 and 5). In some nuclei, only the left side contained PHA-L axons. Notably, many PHA-L axons were observed in periventricular areas of the third ventricle (Fig. 4H–K). 3.4.1. Diencephalon 3.4.1.1. Hypothalamus. VMN: on the left side, except in the injection area, dense PHA-L axons were observed in all areas of the VMN (Figs. 1E–H, 4I–K). The areas adjacent to the VMN, especially the lateral hypothalamus, also contained many PHA-L axons. At the level of the injection site, PHA-L axons were present along the third ventricle. PHA-L axons were present in all areas of the right side of the VMN but the density was less than that of the left side. The positive axons were not uniformity distributed in the right VMN. In the ventrolateral VMN, positive axons were more dense than in other subdivisions. As with the right side, PHA-L positive axons were present along the third ventricle. DMN: on the left side, many PHA-L axons were observed (Fig. 4J and K). ARCN: on the left side, dense PHA-L axons were present (Figs. 1E and F, 4I–K). On the right side, the density of PHA-L axons was low. PHA-L axons were also present on both sides of the median eminence. AH (Fig. 4H and I): PHA-L axons were observed on the left side but were not dense in comparison to that of other areas on the left side. PVN: PHA-L axons were present on the left side, although there were not many on the right side (Fig. 4H and I). SCN, SON: A few PHA-L axons were present on both sides (Fig. 4G). POA including the sexually dimorphic nucleus (SDN) and preoptic nucleus (PON): many PHA-L axons were present on the left side (Figs. 4D–G, 5). AVPV: on the left side, many PHA-L axons were observed and a few were present on the right side (Fig. 4E, 5). MM: in the medial MM, PHA-L axons were present on the left side but not on the right side (Fig. 4L). In the lateral MM, a few PHA-L axons were observed on both sides. LM: on both sides, many PHA-L axons were present (Fig. 4L). SM: on both sides, many PHA-L axons were present (Fig. 4L). 3.4.1.2. Thalamus. The periventricular nucleus of the thalamus (PV): the anterior and posterior regions of the PV contained moderate numbers of PHA-L axons on both sides. However, fewer numbers of PHA-L axons were present in the right side than in the right side (Fig. 4H–K). HB: in both the medial and lateral habenular nucleus, limited numbers of PHA-L axons were observed on the left




5

Fig. 2. Distribution of fluoro-gold (FG)-labeled cells from rostral to caudal (A–P) of the rostral forebrain to the middle level of the midbrain in a female rat after direct injection of FG into the left vlVMN. Pale red areas in the vlVMN indicate the FG-injection site (see J and K). Drawings were modified from the rat brain atlas of Paxinos and Watson (2007). Small dots indicate less than 5 FG-labeled cells; medium dots indicate 6–19 FG-labeled cells; large dots indicate more than 20 FG labeled cells. For further abbreviations, refer to Paxinos and Watson (2007).


G Model NSR-3761; No. of Pages 15 6


Fig. 2. (Continued )

side, but no PHA-L axons were observed on the right side (Fig. 4J and K). 3.4.1.3. Midbrain. IP: neither the left nor the right side contained PHA-L axons (Fig. 4M and N). MCG: large numbers of PHA-L axons

were present on the left of the MCG from the superior colliculus to the inferior colliculus (Fig. 4L–P). PHA-L axons were also present along the aqueduct of the right side. DR: on the left side, the dorsal and lateral DR contained PHA-L axons (Fig. 4O and P). By contrast, no PHA-L axons were present on both sides of the ventral DR. MR:




7

Table 1 Evaluation list of fluoro-gold (FG)-cells and phaseolus vulgaris-leucoagglutinin (PHA-L)-fibers. FG

Lateral septal nucleus (LS)

Medial septal nucleus (MS) Accumbens nucleus (ACCB) Amygdaloid nucleus (AMG) Bed nucleus of the stria terminalis (BNST) Preoptic area nucleus (POA) Anteroventral periventricular nucleus (AVPV) Suprashiasmatic nucleus (SCN) Supraoptic nucleus (SON) Anterior hypothalamic area (AH) Paraventricular hypothalamic nucleus (PVN) Ventromedial hypothalamic nucleus (VMN)

Dorsomedial hypothalamic nucleus (DMN) Arcuate hypothalamic nucleus (ARCN) Medial mammillary nucleus (MM) Lateral mammillary nucleus (LM) Supramammillary nucleus (SM) Medial habenular nucleus (MHB) Lateral habenular nucleus (LHB) Paraventricular thalamic nucleus (PV) Interpeduncular nucleus (IP) Midbrain central gray (MCG) Dorsal raphe nucleus (DR)

Dorsal Intermediate Ventral

Medial Lateral

Dorsomedial Central Ventrolateral Anterior Posterior

Dorsal Ventral Lateral

Median raphe nucleus (MR)

PHA-L

Right

Left

Right

Left

− ± − − − ± − − + + ± − − − ± + + − − ± − − − − − − − ± ± ± ± +

− + + − ± ++ ± + ++ + ± ± + ± +

− ± ± − ± ± − + + + ± − ± + ± + + + + ± − + ± − − + − + ± − ± −

− + ++ − ++ ++ ± ++ ++ ++ ± ± ++ ++ ++

a a

+ ± + + ± ± ± ± ± − + ± ± + +

a a

++ + ++ + + + + + ++ − ++ ± − + ±

a Injecting site of FG or PHA-L. Amounts: −, none, ± a few, + many, ++ large.

only a few PHA-L axons were observed on the left side but not on the right side (Fig. 4O and P). 3.4.1.4. Telencephalon. In the neocortex, including the cingulate cortex, no PHA-L axons were observed on both sides. There were no PHA-L axons on the both sides of hippocampus. LS: there were no PHA-L axons on both sides of the dorsal LS (Fig. 4A–F). In the intermediate LS, many PHA-L axons were present on the left side. On the left side of the ventral LS, large numbers of PHA-L axons were present. On the right side of the intermediate and ventral LS, a few PHA-L axons were observed. MS: no PHA-L axons were noted. MAMG: large numbers of PHA-L axons were present on the left side, but only a few PHA-L axons were present on the right side (Fig. 4F–M). LAMG: on the left side, a few PHA-L axons were observed but not on the right side (Fig. 4F–M). BNST: large numbers of PHA-L axons were noted in all anterior, posterior, and lateral nuclei of the left side (Fig. 4D–G). On the right side, many PHA-L axons were also observed. ACCB: many PHA-L axons were present on the left side of the core and shell (Fig. 4B–D). On the right side, there were few PHA-L axons only in shell region. 4. Discussion Results of the FG and PHAL tracing performed in this study confirmed the existence of direct reciprocal neural connections between the right and left side of the vlVMN. Furthermore, the present results show the presence of bilateral projections from the vlVMN to the forebrain and midbrain, albeit the ipsilateral projections are dominant in most structures. Afferents to the vlVMN also originate from both sides of the forebrain and midbrain. However, ipsilateral afferents are dominant, and in some cases, these are the

only ones present. Thus, the VMN has reciprocal connections with the forebrain and midbrain nuclei involved in regulation of lordosis, except for a few nuclei. The vlVMN is an important nucleus for regulating female sexual behavior (see review by Sakuma, 2013). In the vlVMN, ER␣containing neurons are abundant (Simerly et al., 1990) and are controlled by downregulation of estrogen (Yamada et al., 2009). Recently, it has been reported that the right and left VMN influence each other and regulate ER␣ expression in the VMN, because unilateral lesions of the VMN increase the number of ER-immunopositive neurons in the contralateral side of the VMN (Shimogawa et al., 2014). The present results showing reciprocal neural connections between the right and left sides of the VMN provide an anatomical basis for possible bilateral functional connections in regulating ER expression. In particular, the vlVMN sends many axons to the opposite nucleus and receives many afferent fibers from the contralateral side of the nucleus. In the present PHA-L study, one of the pathways of the right and left neural connections of the VMN is thought to ascend along the third ventricle, traverse to the other side on top of the third ventricle, and then descend along the third ventricle. Another tract occurs in the median eminence. These 2 neural tracts that VMN projects axons to the contralateral side have been described as routes of the commissural fiber in a report by Canteras et al. (1994). There is a possible third route consisting of VMN anterior projections. Although only a few PHA-L labeled axons were observed in the right side areas closed to the rostral of the third ventricle, a supraoptic commissure cannot be excluded from possible routes between the right and left side of the VMN. Anterior and lateral projections of the VMN have been reported to play an important role in induction of lordosis, because cuts




Fig. 3. Representative photomicrographs of coronal brain sections of the right side of nuclei important for lordosis regulation in a female rat injected with fluoro-gold (FG) into the vlVMN. Bars indicate 200 ␮m. For abbreviations, see Table 1.

of anterior (Yamanouchi and Arai, 1979) or lateral projections (Malsbury and Daood, 1978; Pfeifle et al., 1980) diminished lordosis response. Furthermore, the rostral VMN neurons and lateral projections are responsible for estrogen-dependent reproductive functions (Akaishi and Sakuma, 1986; Sakuma and Akaishi, 1987).

Reciprocal connections between the VMN and the neocortex were not observed in the present study. The neocortex has not been reported to be involved in the regulation of female sexual behavior. Rostral regions of the hypothalamus and proximal limbic areas, such as the septum, ACCB, and BNST, have reciprocal neural connections with the VMN. Strong reciprocal and ipsilateral




9

Fig. 4. Distribution of phaseolus vulgaris-leucoagglutinin (PHA-L)-labeled fibers from rostral to caudal (A–P) in a female rat after direct injection of PHA-L in the left vlVMN. Pale red areas in J indicate the PHA-L-injection site. Drawings were modified from the rat brain atlas of Paxinos and Watson (2007). For further abbreviations, refer to Paxinos and Watson (2007).





neural connections between the VMN and POA were noted in the present study. Considerable contralateral connections between the POA and VMN were also observed. Krieger et al. (1979) showed projections of the VMN to the forebrain, such as the POA, LS,

BNST, and medial AMG, in an autoradiographic study. The VMN receives POA projections (Pfaff and Conrad, 1978). Since ablation of the POA by electrical lesioning (Powers and Valenstein, 1972) or neurotoxin injection (Hoshina et al., 1994) facilitates lordosis, the




11

Fig. 5. Representative photomicrographs of coronal brain sections of the right and left side of nuclei important for lordosis regulation in a female rat injected with phaseolus vulgaris-leucoagglutinin (PHA-L) into the vlVMN. Bars indicate 100 ␮m. For abbreviations, see Table 1.








POA plays an inhibitory role in the regulation of lordosis. Electrical stimulation of the POA inhibits lordosis behavior (Takeo et al., 1993). Furthermore, the POA contains many sexually dimorphic nuclei such as the SDN, PON, and AVPV. The AVPV has dense reciprocal and ipsilateral connections to the VMN. The AVPV contains large numbers of estrogen-concentrating neurons that are thought to be important in the initiation of cyclic secretion of gonadotropins by estrogen (Simerly and Swanson, 1987). In the LS, the ventral region, but not the intermediate and dorsal regions, contained many neurons sending axons to the VMN and axons originating in the VMN in the present study. A review of the connections of the ventral LS suggests that this area sends axons around the VMN but not inside it (Risold and Swanson, 1997). The same anatomical phenomenon observed in the PHA-L experiment was previously observed in our laboratory (Tsukahara and Yamanouchi, 2001). In the ventral LS, many estradiol-concentrating (Krieger et al., 1976) or ER mRNA-expressing neurons are present (Simerly et al., 1990). Since the shell of the VMN consists of dendrites and afferents fibers (Millhouse, 1973), projections from the ventral LS can affect VMN function. Neurons of the intermediate part of the LS that project to the MCG are thought to play an inhibitory role in regulating lordosis behavior (see

13

review, Tsukahara et al., 2014). However, the functional connection between the ventral and intermediate LS is yet to be analyzed. Furthermore, the inhibitory system of the septum and facilitative system of the VMN has been reported to operate in an independent manner (Yamanouchi, 1980). The VMN has reciprocal connections with the all of BNST except the lateral nucleus. The lateral nucleus of the BNST has only a oneway connection from the VMN. The role of the BNST in regulating lordosis remains to be determined. However, the stria terminalis is an important pathway for lordosis (Takeo et al., 1995). The medial AMG, which sends axons to the BNST, controls lordosis in a facilitative fashion, since destruction of the medial AMG decreases lordosis (Masco and Carrer, 1980). The present results show that there are the dense reciprocal ipsilateral connections between the VMN and the medial AMG but not between the VMN and the lateral AMG. The lateral AMG plays an inhibitory role in regulating lordosis (Masco and Carrer, 1980). In the area rostral to the hypothalamus, the ACCB also connected reciprocally to the VMN in the present study but it is unknown whether this region regulates lordosis. The AH, PVN, and SO received many ipsilateral VMN axons but few efferents to the VMN were present. The SCN sent considerable numbers of axons to the VMN but received little or no afferents from the VMN.

Fig. 6. Schematic illustrations showing reciprocal connections between the left VMN and medial ventral part of the lateral septum (LS), bed nucleus stria terminalis (BNST), preoptic area (POA), medial and lateral amygdale (AMG), midbrain central gray (MCG), and dorsal (DR) or medial (MR) raphe nuclei, all of which are involved in lordosis regulation. Red and blue lines with arrow heads indicate afferent and efferent connections, respectively, from the left side of the VMN. The size of the arrow indicates the density of the connection. These diagrams were created from the results shown in Table 1.




The ARCN received ipsilateral projections from the VMN but not many contralateral projections were present. The DMN exhibited many neurons sending axons to the VMN in only the ipsilateral side but received VMN axons bilaterally. In the MM, LM, SUM, there were considerable numbers of afferent fibers from the VMN in both sides but only ipsilateral efferents to the VMN were present. In both the medial and lateral HB, only ipsilateral reciprocal connections to the VMN were observed in the present study. However, no connections were observed between the VMN and IP. The HB has been reported to control lordosis behavior in a facilitative fashion (Modianos et al., 1975) and to send axons to the IP (Herkenham and Nauta, 1979), a region also involved in lordosis regulation (Modianos et al., 1975; Kawakami et al., 1979). Although the role of the PV of the thalamus in regulating reproduction remains unknown, there were marked reciprocal connections to the VMN. The hippocampus exhibited no connections with the VMN. In the midbrain, however, the VMN projected dense axons ipsilaterally and widely to the MCG at all levels. This agrees with previous reports using horseradish peroxidase (Beitz, 1982). There were less contralateral VMN projections to the MCG than ipsilateral connections. Afferent fibers from the rostral MCG to the VMN have been shown previously (Eberhart et al., 1985) although not many were observed in the present study. Routs of ascending fibers of the MCG to the VMN were reported by an electrophysiological experiment (Akaishi et al., 1988). The rostral MCG is an important neural center for signaling the induction of lordosis and modulating VMN function (see review, Sakuma, 2013). ER-containing neural cells in the VMN send axons to the MCG in guinea pigs (Turcotte and Blaustein, 1999). These results present neuroanatomical evidence for a lordosis-facilitating neural tract from the VMN to the MCG. In the midbrain, reciprocal and bilateral neural communications between the VMN and DR were shown herein. This partially agrees with a previous report demonstrating that both the DR and MR send serotonergic axons to both sides of the ventromedial hypothalamus (Kanno et al., 2008). The DR, especially serotonergic neurons of the DR, exerts an inhibitory influence on lordosis behavior in both female and male rats (Kakeyama and Yamanouchi, 1996). Luine et al. (1983) reported that direct application of a serotonergic neurotoxin into the VMN facilitated lordosis. Serotoninergic projections to the VMN are sexually dimorphic (Patisaul et al., 2008). Serotonergic connections from the DR to the VMN are an important inhibitory mediator of lordosis. The MR exhibits considerable numbers of neurons sending axons to the VMN but not many axons from the DR. Thus, neuroanatomical evidence regarding the right and left VMN and other areas in the forebrain and midbrain in the present study provide a basis for functional connections of the VMN to other areas. A summary of the anatomical connections of each nucleus involved in the regulation of reproduction, particularly lordosis behavior, is shown in Fig. 6.

Acknowledgements This study was supported by Grants-in-Aid for Scientific Research: Kiban keisei (2010) from MEXT of Japan to KY, (B) (23310043) and (C) (24590307) (23590285) to YS, from the Japan Society for the Promotion of Science (JSPS) to YS, and by the National Institute for Environmental Studies (14309) to YS.

References Akaishi, T., Sakuma, Y., 1986. Projections of oestrogen-sensitive neurones from the ventromedial hypothalamic nucleus of the female rat. J. Physiol. (Lond.) 372, 207–220.

Akaishi, T., Jiang, Z.-Y., Sakuma, Y., 1988. Ascending fiber projections from the midbrain central gray to the ventromedial hypothalamus in the rat. Exp. Neurol. 99, 247–258. Barfield, R.J., Chen, J.J., 1977. Activation of estrous behavior in ovariectomized rats by intracerebral implants of estradiol benzoate. Endocrinology 101, 1716–1725. Beitz, A.J., 1982. The organization of afferent projection to the midbrain periaqueductal gray of the rat. Neuroscience 7, 133–159. Bleier, R., Byne, W., 1985. Septum and hypothalamus. In: Paxinos, G. (Ed.), The Rat Nervous System. Vol. 1. Forebrain and Midbrain. Academic Press, Australia, pp. 87–118. Canteras, N.S., Simerly, R.B., Swanson, L.W., 1994. Organization of projections from the ventromedial nucleus of the hypothalamus: a phaseolus vulgarisleucoagglutinin study in the rat. J. Comp. Neurol. 348, 41–79. Eberhart, J.A., Morrell, J.I., Krieger, M.S., Pfaff, D.W., 1985. An autoradiographic study of projections ascending from the midbrain central gray, and from the region lateral to it, in the rat. J. Comp. Neurol. 241, 285–310. Hoshina, Y., Takeo, T., Nakano, K., Sato, T., Sakuma, Y., 1994. Axon-sparing lesion of the preoptic area enhances receptivity and diminishes proceptivity among components of female rat sexual behavior. Behav. Brain Res. 61, 197–204. Herkenham, M., Nauta, W.J., 1979. Efferent connections of the habenular nuclei in the rat. J. Comp. Neurol. 187, 19–47. Kakeyama, M., Yamanouchi, K., 1996. Inhibitory effect of baclofen on lordosis in female and male rats with dorsal raphe nucleus lesion or septal cut. Neuroendocrinology 63, 290–296. Kanno, K., Shima, S., Ishida, Y., Yamanouchi, K., 2008. Ipsilateral and contralateral serotonergic projections from dorsal and median raphe nuclei to the forebrain in rats: immunofluorescence quantitative analysis. Neurosci. Res. 61, 207–218. Kawakami, M., Akema, T., Ando, A., 1979. Electrophysiological studies on the neural networks among estrogen and progesterone effective brain areas on lordosis behavior of the rat. Brain Res. 169, 287–301. Kondo, Y., Shinoda, A., Yamanouchi, K., Arai, Y., 1990. Role of septum and preoptic area in regulating masculine and feminine sexual behavior in male rats. Horm. Behav. 24, 421–434. Krieger, M.S., Morrell, J.I., Pfaff, D.W., 1976. Autoradiographic localization of estradiol-concentrating cells in the female hamster brain. Neuroendocrinology 22, 193–205. Krieger, M.S., Conrad, L.C., Pfaff, D.W., 1979. An autoradiographic study of the efferent connections of the ventromedial nucleus of the hypothalamus. J. Comp. Neurol. 183, 785–815. Luine, V.N., Frankfurt, M., Rainbow, T.C., Biegon, A., Azmitia, E., 1983. Intrahypothalamic 5,7-dihydroxytryptamine facilitates feminine sexual behavior and decreases [3 H] imipramine binding and 5-HT uptake. Brain Res. 264, 344–348. Malsbury, C.W., Daood, J.T., 1978. Sexual receptivity: critical importance of supraoptic connections of the ventromedial hypothalamus. Brain Res. 159, 451–457. Masco, D.H., Carrer, H.F., 1980. Sexual receptivity in female rats after lesion or stimulation in different amygdaloid nuclei. Physiol. Behav. 24, 1073–1080. Millhouse, O.E., 1973. Certain ventromedial hypothalamic nucleus. Brain Res. 55, 89–105. Modianos, D.T., Hitt, J.C., Popolow, H.B., 1975. Habenular lesions and feminine sexual behavior of ovariectomized rats: diminished responsiveness to the synergistic effects of estrogen and progesterone. J. Comp. Physiol. Psychol. 89, 231–237. Nance, D.M., Shryne, J., Gorski, R.A., 1974. Septal lesions: effects on lordosis behavior and pattern of gonadotropin release. Horm. Behav. 5, 73–81. Patisaul, H.B., Fortino, A.E., Polston, E.K., 2008. Sex differences in serotonergic but not gamma-aminobutyric acidergic (GABA) projections to the rat ventromedial nucleus of the hypothalamus. Endocrinology 149, 397–408. Paxinos, G., Watson, C., 2007. The Rat Brain in Stereotaxic Coordinates, 6th ed. Academic Press, New York. Pfaff, D.W., Conrad, L.C.A., 1978. Hypothalamic neuroanatomy: steroid hormone binding and patterns of axonal projections. Int. Rev. Cytol. 54, 245–264. Pfaff, D.W., Sakuma, Y., 1979a. Deficit in the lordosis reflex of female rats caused by lesions in the ventromedial nucleus of the hypothalamus. J. Physiol. (Lond.) 288, 203–310. Pfaff, D.W., Sakuma, Y., 1979b. Facilitation of the lordosis reflex of female rats from the ventromedial nucleus of the hypothalamus. J. Physiol. (Lond.) 288, 189–202. Pfaff, D.W., 1980. Estrogens and Brain Function: Neural Analysis of a Hormonecontrolled Mammalian Reproductive Behavior. Springer-Verlag, New York. Pfaff, D.W., Sakuma, Y., Kow, L.-M., Lee, A.W.L., Easton, A., 2006. Hormonal, neural, and genomic mechanisms for female reproductive behaviors, motivation and arousal. In: Neill, J.D. (Ed.), Knobil and Neill’s Physiology of Reproduction, vol. 34, 3rd ed. Elsevier, New York, pp. 1825–1919. Pfeifle, J.K., Shivers, M., Edwards, D.A., 1980. Parasagittal hypothalamic knife cuts and sexual receptivity in the female rat. Physiol. Behav. 24, 145–150. Powers, J.B., Valenstein, E.S., 1972. Sexual receptivity: facilitation by medial preoptic lesions in female rats. Science 175, 1003–1005. Rajendren, G., Dudley, C.A., Moss, R.L., 1991. Role of ventromedial nucleus of hypothalamus in the male-induced enhancement of lordosis in female rats. Physiol. Behav. 50, 705–710. Renaud, L.P., Martin, J.B., 1975. Electrophysiological studies of connections of hypothalamic ventromedial nucleus neurons in the rat: evidence for a role in neuroendocrine regulation. Brain Res. 93, 145–151. Risold, P.Y., Swanson, L.H., 1997. Chemoarchitecture of the rat lateral septal nucleus. Brain Res. Brain Res. Rev. 24, 91–113. Sakuma, Y., Pfaff, D.W., 1979. Facilitation of female reproductive behavior from mesencephalic central gray in the rat. Am. J. Physiol. 237, R278–R284.




Sakuma, Y., Pfaff, D.W., 1981. Electrophysiological determination of projections from ventromedial hypothalamus to midbrain central gray: difference between female and male rats. Brain Res. 225, 184–188. Sakuma, Y., Akaishi, T., 1987. Cell size, projection path, and localization of estrogensensitive neurons in the rat ventromedial hypothalamus. J. Neurophysiol. 57, 1148–1159. Sakuma, Y., 1994. Estrogen-induced changes in the neural impulse flow from the female rat preoptic region. Horm. Behav. 28, 438–444. Sakuma, Y., 2013. Mechanisms of behaviors related to reproduction. In: Pfaff, D.W. (Ed.), Neuroscience in the 21th Century. LLC: Springer Science+Business, pp. 1783–1794. Saper, C.B., Swanson, L.H., Cowan, V.M., 1976. The efferent connections of the ventromedial nucleus of the hypothalamus of the rat. J. Comp. Neurol. 169, 409–442. Shimogawa, Y., Maekawa, F., Yamanouchi, K., 2014. Unilateral lesion increases oestrogen receptor ␣ expression in the intact side of the ventromedial hypothalamic nucleus in ovariectomised rats. J. Neuroendocrinol. 26, 258–266. Simerly, R.B., Swanson, L.W., 1987. The distribution of neurotransmitter-specific cells and fibers in the anteroventral periventricular nucleus: implication for the control of gonadotropin secretion in the rat. Brain Res. 400, 11–34. Simerly, R.B., Chang, C., Muramatsu, M., Swanson, L.W., 1990. Distribution of androgen and estrogen receptor mRNA-containing cells in the rat brain: an in situ hybridization study. J. Comp. Neurol. 294, 76–95.

15

Takeo, T., Chiba, Y., Sakuma, Y., 1993. Suppression of lordosis reflex of female rats by efferents of the medial preoptic area. Physiol. Behav. 53, 831–838. Takeo, T., Kudo, M., Sakuma, Y., 1995. Stria terminalis conveys a facilitatory estrogen effect on female rat lordosis reflex. Neurosci. Lett. 184, 79–81. Tsukahara, S., Yamanouchi, K., 2001. Neurohistological and behavioral evidence for lordosis inhibiting tract from lateral septum to periaqueductal gray in male rats. J. Comp. Neurol. 431, 293–610. Tsukahara, S., Kanaya, M., Yamanouchi, K., 2014. Neuroanatomy and sex difference of lordosis-inhibiting system in the lateral septum. Front. Neurosci. 8, 299, 1–8. Turcotte, J.C., Blaustein, J.D., 1999. Projection of the estrogen receptorimmunoreactive ventrolateral hypothalamus to other estrogen receptorimmunoreactive sites in female guinea pig brain. Neuroendocrinology 69, 63–76. Yamada, S., Noguchi, D., Ito, H., Yamanouchi, K., 2009. Sex and regional differences in decrease of oestrogen receptor ␣-immunoreactive cells by oestrogen in rat hypothalamus and midbrain. Neurosci. Lett. 463, 135–139. Yamanouchi, K., Arai, Y., 1979. Effects of hypothalamic deafferentation on hormonal facilitation of lordosis in ovariectomized rats. Endocrinol. Jpn. 26, 307–312. Yamanouchi, K., 1980. Inhibitory and facilitatory neural mechanisms involved in the regulation of lordosis behavior in female rats: effects of dual cuts in the preoptic area and hypothalamus. Physiol. Behav. 25, 721–725.


The afferent and efferent connections of the ventromedial thalamic nucleus in the rat.

Facilitation of the lordosis reflex of female rats from the ventromedial nucleus of the hypothalamus.

Role of the efferent and afferent vagus nerve in the development of ventromedial hypothalamic (VMH) obesity.

Connections between the pontine central gray and the ventromedial hypothalamus are essential for lordosis in female rats.

Deficit in the lordosis reflex of female rats caused by lesions in the ventromedial nucleus of the hypothalamus.

Estradiol regulates markers of synaptic plasticity in the hypothalamic ventromedial nucleus and amygdala of female rats.

Endogenous opioid-immunoreactive neurons of the ventromedial hypothalamic nucleus concentrate estrogen in male and female rats.

Afferent and efferent connections of the oculomotor region of the fastigial nucleus in the macaque monkey.

Electrophysiological studies of connections of hypothalamic ventromedial nucleus neurons in the rat: evidence for a role in neuroendocrine regulation.

Afferent and efferent connections of the laterodorsal tegmental nucleus in the rat.

Afferent and efferent connections of C1 cells with spinal cord or hypothalamic projections in mice.

The regulation of oxytocin receptor binding in the ventromedial hypothalamic nucleus by testosterone and its metabolites.

Afferent and efferent fiber connections of the carp torus longitudinalis.

The regulation of oxytocin receptor binding in the ventromedial hypothalamic nucleus by gonadal steroids.

Afferent and efferent connections of the bullfrog medial pallium.

Afferent connections of the rat's supraoptic nucleus.

Efferent connections of the caudate nucleus in the Virginia opossum.

The efferent connections of the suprachiasmatic nucleus of the hypothalamus.

Neuroanatomy of central cardiovascular control. Nucleus tractus solitarii: afferent and efferent neuronal connections in relation to the baroreceptor reflex arc.

Hypothalamic ventromedial nucleus and hormonal regulation of estrous rhythm in the rat.

The efferent connections of the feline nucleus cuneatus.

Postnatal development of the hypothalamic ventromedial nucleus: neurons and synapses.

Estrogenic regulation of histamine receptor subtype H1 expression in the ventromedial nucleus of the hypothalamus in female rats.

Direct efferent and afferent connections of the hippocampus with the neocortex in the marmoset monkey.