Effects of gonadectomy and treatment with gonadal steroids and luteinizing hormone secretion in hypogonadal male and female mice with preoptic area implants.

The ability of male and female hypogonadal (hpg) mice with preoptic area (POA) grafts to show negative feedback was studied. GnRH neurons within POA grafts send axons into the median eminence of the hpg host (HPG/POA), resulting in increased gonadotropin production and gonadal development in the mice that are genetically unable to produce GnRH. The present studies evaluated whether negative feedback, an aspect of normal reproductive function, is present in male and female HPG/POA mice. In normal male mice plasma LH was increased 1 or 5 months after castration and returned to baseline after testosterone propionate treatment. In contrast, no alterations in plasma LH were measured in similarly treated HPG/POA males. HPG/POA female mice were ovariectomized 6 weeks or 3 or 6 months after graft surgery and received sc 17 beta-estradiol (E2) implants 3 months later. Normal mice were studied when 6 weeks old and 8 months old (age-matched to HPG/POA mice ovariectomized 3 months after graft surgery). Further, to determine whether the mice were capable of positive feedback, 1 week after receiving E2 implants, mice in the 6-week and 3-month postgraft surgery groups were challenged with sequential administration of estradiol benzoate and progesterone. The significant increase in plasma LH after ovariectomy or decrease after E2 implant in normal female mice was not present in most HPG/POA female mice. Just 2 of the 24 HPG/POA females studied had increased plasma LH after gonadectomy, and in only 1 of these was plasma LH suppressed by E2 treatment. The ability of an individual HPG/POA mouse to show positive feedback did not predict the ability to show negative feedback, nor did the ability to show negative feedback predict positive feedback capability. Among the mice that failed to respond to ovariectomy with increased LH release were some that had elevated LH in response to steroid challenge or had spontaneously ovulated. On the other hand, neither mouse that had increased LH release after ovariectomy had shown positive feedback to a steroid challenge. Immunocytochemical evaluation revealed GnRH cells within the grafts and GnRH fiber innervation of the host's median eminence, but there was no correlation between numbers of GnRH cells or extent of innervation with the ability to show either negative or positive feedback, nor was the presence of vasoactive intestinal peptide cells within the grafts of predictive value. The failure of negative feedback in most of the HPG/POA mice tested may be due to the failure to establish as yet unidentified but essential afferents to the grafted GnRH cells and/or their axonal processes.

Estradiol-concentrating cells in the brains of hypogonadal female mice and in their intraventricular preoptic area implants.

Estradiol-concentrating cells were evaluated in the brains of hypogonadal female mice and in their intraventricular preoptic area brain grafts using autoradiography for [3H]estradiol. Normal distribution of estradiol-concentrating cells was observed in the brains of the hypogonadal mice with dense collections of these cells in the lateral septum; the medial preoptic area; the medial anterior hypothalamus; the ventromedial, arcuate, and periventricular nuclei of the hypothalamus; and the medial and cortical nuclei of the amygdala. In addition, estradiol-concentrating cells were present in all the transplants, with the estimated number of such cells in the transplants ranging from 390 to 2600. There was no correlation between numbers of estradiol-concentrating cells within the transplants and degree of reproductive recovery in the hypogonadal mice. Gonadotropin-releasing hormone (GnRH) immunocytochemistry of alternate sections revealed GnRH-immunoreactive material within the grafts and immunoreactive fibers exiting the grafts and entering the hosts' median eminence. No specific relationship between GnRH cells and estradiol-concentrating cells was evident within the grafts, nor was there any indication of identity of estrogen-concentrating cells with GnRH cells.

Allografts of CNS tissue possess a blood-brain barrier. I. Grafts of medial preoptic area in hypogonadal mice.

This study represents the first part of a three-part investigation of blood vessels supplying CNS tissue transplanted within the brains of adult mammalian hosts. The results emphasize that blood vessels in solid CNS grafts contribute a blood-brain barrier to that of the host. Neurosecretory cells in basal forebrain grafts placed intraventricularly on the dorsal surface of the host median eminence, a neurosecretory site containing fenestrated blood vessels, do not stimulate similar blood vessels to inhabit the transplanted tissue. Solid grafts of the medial preoptic area containing neurons that synthesize and secrete gonadotropic hormone-releasing hormone (GnRH) were obtained from AKR mice and placed into the third cerebral ventricle of hypogonadal (HPG) mice genetically incapable of synthesizing GnRH. GnRH neurons in the allografts were confirmed immunohistochemically. Blood vessels supplying the host median eminence and the allograft at 10 days to 3 months post-transplantation were analyzed with peroxidase cytochemistry applied in three ways: to HPG mice injected systemically with native horseradish peroxidase; to HPG mice infused into the aorta with peroxidase subsequent to perfusion fixation; and to HPG mice brains fixed by immersion and incubated for endogenous peroxidase activity in red cells retained within blood vessels. The median eminence of the HPG mouse was innervated by GnRH neurons residing within the graft, and blood vessels traversing the median eminence-allograft interface were seen rarely. The allografts contained no fenestrated endothelia, and no extravasations of blood-borne HRP were related directly to leaky blood vessels supplying the grafted tissue. Endothelial cells throughout the CNS grafts were similar morphologically to blood-brain barrier endothelia; they were nonfenestrated, exhibited interendothelial tight junctional complexes and an endomembrane system of organelles, and they endocytosed blood-borne HRP that eventually was sequestered within dense body lysosomes. The results support the belief that blood vessels supplying CNS tissue transplanted to a host brain manifest endothelial characteristics identical to those of the tissue in normal life and to those of the host CNS.

Thymocyte maturity in male and female hypogonadal mice and the effect of preoptic area brain grafts.

Hypogonadal mice with a genetic deficiency of gonadotropin-releasing hormone (GnRH) have low levels of luteinizing hormone (LH), follicle-stimulating hormone (FSH) and gonadal steroids. In this study we found differences from normal mice in many aspects of thymic development. Thymus weights and cellularity were higher in hypogonadal than in normal male mice but lower in hypogonadal than in normal females. Although all normal mice had higher proportions of mature, single staining thymocytes (CD8+ or CD4+) than seen in hypogonadal mice, there was a sex difference in the basis for this shift. Significantly more double-staining (CD8+, CD4+) thymocytes were seen in hypogonadal males than in normal males while both groups had similar single-staining populations. However, in females, both single-staining CD8+ and CD4+ thymocytes were more numerous in normal than in hypogonadal females while numbers of double-staining cells were similar in the two groups. These studies indicate that a mature thymocyte profile may be arrived at through differential effects of reproductive hormones in males and females. When brain grafts containing GnRH cells were used to correct reproductive deficits in hypogonadal mice, there were higher splenocyte counts in males with grafts, a similar trend in females, and a lower ratio of single staining CD4+ to CD8+ thymocytes in all females with grafts vs. all females without, regardless of whether or not the grafts corrected the reproductive hormone status of the recipients, indicating an effect of the graft surgery on the immune system.

Positive feedback in hypogonadal female mice with preoptic area brain transplants.

When fetal preoptic area (POA) brain grafts that contain gonadotropin-releasing hormone cells are transplanted into the third ventricle of adult female hypogonadal mice, the animals respond with sexual maturation, persistent estrus, and the ability to ovulate reflexively after mating. However, the absence of normal spontaneous ovulatory cyclicity suggests an impairment in positive feedback. We, therefore, studied the effect of administration of progesterone alone or of sequential estradiol benzoate and progesterone on plasma levels of luteinizing hormone (LH) in groups of hypogonadal (HPG) mice in persistent estrus after receiving POA grafts (HPG/POA). Individual differences in responsivity to progesterone were related in part to the length of time in persistent estrus. Approximately 30% of HPG/POA grafts tested 2 months after graft showed increased levels of plasma LH. This was reduced to 10% when animals were tested 5 months after graft. Sequential administration of estradiol benzoate plus progesterone to intact HPG/POA mice was ineffective in elevating LH. The presence of corpora lutea in ovaries verified that only animals with a progesterone induced LH surge ovulated. Other HPG/POA mice were mated, and the occurrence of reflex ovulation was determined. Four of these mice delivered pups: 3 were previous responders to progesterone. One female mated again during the immediate postpartum period and delivered a second litter. Following weaning of all offspring, this animal displayed spontaneous ovarian cyclicity, confirmed by ovarian histology. This is the first proven example of spontaneous ovulation in a mutant mouse with a brain graft. The results show that some HPB/POA mice are capable of positive feedback responses, and rarely, of becoming spontaneous ovulators.

Gonadal steroids influence neurophysin II distribution in the forebrain of normal and mutant mice.

The distribution of arginine vasopressin-associated neurophysin (neurophysin II) immunoreactivity was investigated in normal and mutant house mice during development and after various gonadal steroid manipulations. During postnatal development of normal mice dense networks of neurophysin II immunoreactivity in the lateral septal nucleus and lateral habenular nucleus appeared earlier in male than in female mice, with an adult pattern of immunoreactivity being attained by 8 weeks and 12 weeks of age, respectively. The neurophysin II immunoreactivity in the male was denser than that in female mice. After gonadectomy of adult normal mice there was a gradual loss of neurophysin II immunoreactivity in the lateral septum and lateral habenula over a period of 15 weeks. In hypogonadal mice, a mutant in which gonadal development is arrested postnatally due to a deficiency in hypothalamic gonadotrophin releasing hormone, no immunoreactive neurophysin II could be detected in the lateral septum or lateral habenula. A pattern of neurophysin II immunoreactivity similar to that in normal control mice was observed in hypogonadal mice which had been implanted for 4 weeks with silicone elastomer capsules containing testosterone or oestradiol-17 beta, but not 5 alpha-dihydrotestosterone or progesterone. Stimulation of gonadal development and endogenous steroid production in hypogonadal mice by third ventricular grafts of preoptic area tissue from normal neonatal animals also produced a normal pattern of neurophysin II immunoreactivity in the lateral septum and lateral habenula. In the androgen-insensitive testicular feminized mouse immunoreactive neurophysin II was undetectable in the lateral septum and lateral habenula. Treatment of testicular feminized mice with oestradiol-17 beta, but not progesterone, produced a normal pattern of neurophysin II immunoreactivity. The main immunohistological findings were confirmed by radioimmunoassay of tissue extracts which showed that the concentration of arginine vasopressin in lateral septum was far greater in normal males than females and was undetectable in hypogonadal mice; no oxytocin could be detected in the septum of normal or hypogonadal mice. These results show that the expression of neurophysin II immunoreactivity in the lateral septum and lateral habenula of the mouse brain is dependent on the presence of aromatizeable androgens or oestrogens.

Neurotropic effects of estrogen on the neonatal preoptic area grafted into the adult rat brain.

The preoptic area (POA) or cerebral cortex taken from newborn female rats were transplanted into the third ventricle of ovariectomized adult rats. From the day of transplantation, estradiol-17 beta in a silastic capsule was implanted subcutaneously into host animals for 4 weeks. The POA or cerebral cortex transplants were examined at light- and electron-microscopic levels 4 weeks after transplantation. All of the POA or cortical grafts showed an appearance similar to normal neural tissue. Estrogen exposure for 4 weeks via the host induced a significant increase in the volume of the POA grafts. The neuronal population of the POA grafts exposed to estrogen was not significantly different from that of the POA grafts without estrogen treatment. However, the number of axodendritic shaft and spine synapses of the POA grafts exposed to estrogen was significantly greater than that of the POA grafts without estrogen treatment. In contrast, there was no significant difference in the volume of the cortical tissues transplanted into the brain between the control and estrogen-treated groups. These results suggest that estrogen has a stimulatory effect on the development of neuronal substrates in the intraventricular POA graft, increasing its volume and synaptic population.

Quantitative analysis of synaptic input to gonadotropin-releasing hormone neurons in normal mice and hpg mice with preoptic area grafts.

Mutant hypogonadal (hpg) mice with a truncated gene for the precursor to gonadotropin-releasing hormone (GnRH) show certain aspects of recovery of reproductive function after receiving grafts of normal preoptic area into the third ventricle. We have previously shown that GnRH neurons from within the grafts can innervate the appropriate neural-hemal target in the host. To determine if in turn these exogenously derived neurons receive a synaptic input comparable to the GnRH neurons in the normal animal we have now carried out a quantitative ultrastructural analysis to compare the synaptic input to GnRH neurons in the normal preoptic area and in the grafts. In almost all cases GnRH cells or dendrites in normal brains and within the grafts received a synaptic input. In normal animals, input to GnRH dendritic profiles was significantly greater (P less than 0.001) than to the somatic plasma membrane and this trend was also observed within the grafts though the difference was not statistically significant. In addition, no statistically significant difference was found between the input to GnRH structures within the grafts and in normal preoptic area. However, a substantial variability in input among grafted animals was evident which was not observed in normal animals. The sources of variability within the grafts are discussed and we suggest that the deficiencies and differences that exist in regulation of gonadotropin secretion among grafted hpg animals may be reflected in aberrant synaptic input.

Plasma LH rises rapidly following mating in hypogonadal female mice with preoptic area (POA) brain grafts.

Hypogonadal female mice, genetically deficient in gonadotropin releasing hormone (GnRH), respond to preoptic area (POA) grafts obtained from normal fetal or neonatal mice with increased gonadotropin levels, ovarian and uterine development and continual vaginal estrus rather than spontaneous ovulatory cyclicity. Previous studies showed that such mice became pregnant following one overnight pairing with a normal male, indicating reflex ovulation. The present study evaluated plasma LH concentrations in relation to mating. Plasma LH levels in the hpg females with POA grafts were significantly elevated 10 min following the male partner's ejaculation, but were no different than baseline at 30, 60 or 120 min following the male's ejaculation. The post-copulatory plasma LH levels of 3.0 +/- 0.6 ng/ml (mean +/- S.E.M.) were considerably lower than the proestrous LH surge seen in the normal females in the colony (16.8 +/- 4.8 ng/ml), but in at least 4 of 10 hpg mice the levels were sufficient to induce ovulation as proved by pregnancy following this single mating. Grafts contained GnRH-reactive cells and fibers that projected to the median eminence of the host brains.

Implantation of fetal preoptic area into the lateral ventricle of adult hypogonadal mutant mice: the pattern of gonadotropin-releasing hormone axonal outgrowth into the host brain.

Transplantation of fetal preoptic area tissue containing gonadotropin-releasing hormone neurons into the third ventricle of male hypogonadal mice resulted in an elevation of pituitary gonadotropin levels and correction of hypogonadism. This reversal of the neuroendocrine deficit was correlated with innervation of the median eminence by gonadotropin-releasing hormone axons. The specificity of fiber outgrowth suggested that local neuromodulatory factors might guide these axons to the nearby median eminence. To test this hypothesis, 14 adult hypogonadal males received unilateral fetal preoptic area grafts into the lateral ventricle, a site distant from the median eminence. After four months, healthy grafts containing numerous gonadotropin-releasing hormone neurons were seen in 9 hosts. However, none of these grafts corrected the hypogonadism of the host and there was no gonadotropin-releasing hormone innervation of the median eminence in any of these animals, thus demonstrating that the presence of gonadotropin-releasing hormone neurons in the ventricular space is itself not sufficient to stimulate the pituitary-gonadal axis. Instead, gonadotropin-releasing hormone axons coursed in the host fimbria, fornix, corpus callosum, and stria terminalis. These fibers could be traced into the anterior hippocampal area, medial and lateral septum, and the anterior hypothalamus. The distribution of these fibers included a number of regions which receive gonadotropin-releasing hormone fiber input in the normal mouse. These findings show that gonadotropin-releasing hormone neurons transplanted into the lateral ventricle can survive and extend processes into the host brain, often projecting to sites of normal gonadotropin-releasing hormone innervation. Their success in contacting these sites suggests that gonadotropin-releasing hormone fiber outgrowth may be influenced by regionally specified trophic and/or guidance factors.

Testosterone-induced enhancement of male medial preoptic tissue transplant volume in female recipients: a possible neuronotrophic action.

To characterize further the action of gonadal hormones on the development of the brain, this study was designed to test the ability of testosterone treatment to modify the volume of male brain tissue transplants in female recipients. One-day-old females, in addition to having either medial preoptic area (MPOA) or caudate nucleus (CN) tissue from neonatal males bilaterally implanted into their own MPOA, also received a subcutaneous injection of either 200 micrograms testosterone propionate (TP) or oil concurrently, and on the following 4 days. All recipients were sacrificed at 30 days of age postnatally. An analysis of male transplant volumes indicated that MPOA transplants in oil-treated recipients were substantially reduced in size (0.06 +/- 0.01 mm3) compared with the initial transplant volume of 0.19 mm3. However, MPOA transplants in recipients treated with TP showed a 79% increase above the initial transplant volume (to 0.34 +/- 0.05 mm3). The resulting 5- to 6-fold difference in MPOA transplant volume between oil and TP treated animals was highly significant. In sharp contrast, the same TP treatment to recipients receiving male CN transplants resulted in no enhancement of transplant volume. Therefore, transplants involving a brain area known to concentrate 3H-labeled testosterone neonatally (i.e. the MPOA) responded to TP treatment with an enhancement of volume which was not observed for transplants consisting of brain tissue not known to concentrate 3H-labeled testosterone (i.e. the CN). The above results suggest that testosterone is a 'neuronotrophic' agent during development that acts specifically on cells within steroid-sensitive brain areas, perhaps to prevent neuronal death within these areas.

Female sexual behavior in hypogonadal mice with GnRH-containing brain grafts.

Hypogonadal female mice respond to GnRH-containing fetal preoptic area (POA) implants in the third ventricle with vaginal opening and persistent vaginal estrus, ovarian, and uterine development and increased gonadotropin secretion. When these females are mated with normal males, reflex ovulation results in pregnancy. In the present study, POA implants derived from neonatal pups, whether male or female, were also capable of supporting reproductive development in the hypogonadal female mice. Evaluation of female sexual behavior in the mice with grafts showed that these mice responded to normal males with comparable levels of lordosis as are seen in normal female mice on the proestrous days of their cycles.

Effects of castration or testosterone implants upon pituitary function in hypogonadal mice bearing normal foetal preoptic area grafts.

Grafts of normal mouse preoptic area (POA) tissue into the third ventricle of gonadotrophin-releasing hormone (GnRH)-deficient hypogonadal (hpg) mice resulted in an elevation of pituitary GnRH receptors, an increased synthesis of luteinizing hormone (LH) and follicle-stimulating hormone (FSH) by the pituitary gland, an elevation of gonadal LH receptors and in the stimulation of steroidogenesis and spermatogenesis in the testis. In normal mice both castration or the subcutaneous implantation of testosterone capsules for 10 days reduced GnRH receptors, pituitary LH and FSH content, and the latter treatment also caused a 50% reduction in testicular LH receptors. In hpg mice bearing POA grafts testosterone implants failed to affect any of the above parameters, and castration failed to affect pituitary gonadotrophin hormone content, although there was a slight reduction in pituitary GnRH receptors after castration. These experiments suggest that neither the pituitary gonadotroph, nor the GnRH neurone represent major sites for the direct negative feedback of testosterone upon gonadotrophic hormone secretion in male mice.

Modulation of aromatase activity by testosterone in transplants of fetal rat hypothalamus-preoptic area.

Conversion of androgens to estrogens by neural aromatase appears to be a prerequisite for a variety of effects of androgens on brain function, including sexual differentiation. Activity of aromatase is modulated by its substrate testosterone (T) in adult hypothalamus-preoptic area (HPOA), resulting in significantly higher levels in the male. Perinatal sex differences in activity have also been observed in hypothalamus, POA and/or amygdala. However, it is not known if higher levels in the perinatal male occur in response to circulating androgens, nor whether early exposure to gonadal steroids is necessary to establish either basal levels or the androgen sensitivity of aromatase activity in the adult brain. In order to investigate the influence of early steroid exposure on the development of neural aromatase activity, embryonic day (E)17 fetal HPOA was transplanted onto the choroidal pia overlying the superior colliculus of adult ovariectomized-adrenalectomized (OVX-ADX) Holtzman female hosts. In the first experiment, the effect of androgen exposure on aromatase activity in mature HPOA transplants was determined. Hosts received T-filled silastic capsules or underwent sham surgery 7 weeks after transplantation and were sacrificed 7 days later. Aromatase activity was determined in vitro using the stereospecific production of 3H2O from [1 beta-3H]androstenedione as an index of estrogen formation. Aromatase activity was significantly greater in T-treated HPOA versus controls (P less than 0.005). Activity was not affected by the sex of the donor fetus. In the second experiment, the effect of androgen exposure during the first 6 days following transplantation of E17 HPOA (corresponding to the last gestational week) was determined.(ABSTRACT TRUNCATED AT 250 WORDS)

The effects of the age of intracerebroventricular grafts of normal preoptic area tissue upon pituitary and gonadal function in hypogonadal (HPG) mice.

Hypogonadal mice are deficient in the hypothalamic gonadotrophic hormone releasing hormone and as a consequence postnatal testicular development does not occur. Grafting preoptic area tissue from normal mice directly into the hypogonadal third ventricle dramatically reverses the hypogonadism; however, the age of the grafted preoptic area tissue is crucial to the survival and function of the graft. Grafting embryonic tissue (E16-18) resulted in 69% of the hypogonadal mice increasing testis weight some sevenfold within 30 days (5.6 to 35 mg). Postnatal day 1 (P1) tissue grafts elicited a similar rise in testis weight in 77% of recipients, whereas P5 tissue was only successful in 22% of cases. In this experimental group, however, testis weight also increased sevenfold compared with hypogonadal untreated mice. Stimulation of testicular growth in the E16-18 and P1 experimental groups was accompanied by an increase in pituitary gonadotrophic hormone content. P10 tissue did not stimulate testis growth nor was pituitary gonadotrophic hormone control elevated and the majority of grafts failed to survive over the 30 day period of the experiment. The present study has shown that the age of grafted tissue is critical in the restoration of physiological function in hypogonadal mice, and that gonadotrophic hormone releasing hormone neurons from E16-18, P1 and P5 preoptic area grafts that survive the 30 day period of the experiment and whose axons reach the median eminence portal vessels are equipotent in stimulating pituitary gonadotrophin synthesis and secretion.

Neural grafts and the restoration of pituitary and gonadal function in hypogonadal (HPG) mice.

The hypogonadal mouse is a mutant deficient in the hypothalamic gonadotrophic hormone releasing hormone (GnRH) with a consequent extreme depletion in pituitary luteinising hormone (LH) and follicle stimulating hormone (FSH) and a failure of post-natal gonadal growth. Grafting late foetal-early neonatal preoptic area (POA) tissue into the third ventricle of adult hpg mice resulted in increased GnRH receptors in the pituitary, increased synthesis and secretion of LH and FSH, and full stimulation of spermatogenesis in the male and folliculogenesis in the females. Although the POA grafts in the female stimulated ovarian and uterine growth and vaginal opening, so far there has been no evidence of ovarian cyclicity and the females did not ovulate spontaneously. However the increased ovarian steroidogenesis stimulated full female mating behaviour and a high percentage of the females became pregnant. The evidence suggests that these females must have ovulated reflexly upon mating, raising the question as to where in the graft the mating response was transduced. Over 90% of GnRH positive axons ended up innervating the median eminence, suggesting that even in adult mice this region of the brain retains its trophic capacity. In male hpg mice treated with testosterone there was no evidence of a negative feedback upon the graft induced elevation of pituitary LH and FSH content, suggesting that a site of androgen negative feedback may be at a CNS site removed from the GnRH neurone.

The fate of allogeneic and xenogeneic neuronal tissue transplanted into the third ventricle of rodents.

Neural grafts from day 17-19 fetal rats or mice survived well when transplanted into syngeneic, or immunodeficient hosts, thus demonstrating that there are no non-immunological barriers to cross-species transplantation of neuronal tissue in rats and mice. However, intraventricular grafts from rat to mouse, or vice versa, in immunocompetent animals were rejected in less than 30 days. By this time all graft tissue had been destroyed and scavenged, presumably by the macrophages seen infiltrating the grafts within 10 days of grafting. Rat allografts from major histocompatibility complex disparate donors disparate donors survived well as did grafts between rats differing only at minor histocompatibility loci. However, allografts from donors that differed from recipients at both major and minor histocompatibility complex loci had a variable survival time. When neural tissue was grafted into immunologically primed recipients, it was rejected as was similar tissue grafted beneath the kidney capsule of an allogeneic host. Concomitant grafting of allogeneic tissue under the kidney capsule and into the third ventricle was followed by rejection in both sites. A striking observation in these studies was the induction of Class I major histocompatibility complex antigens on grafted neuronal tissue. High levels of antigen expression were correlated with a vigorous host response and poor graft survival but lower levels were not indicative of impending graft destruction. Whilst the brain can be regarded as an immunologically privileged site, the privilege is not absolute and caution needs to be exercised in the interpretation of results from allogeneic or xenogeneic grafts.

Ultrastructure of gonadotropin-releasing hormone neuronal structures derived from normal fetal preoptic area and transplanted into hypogonadal mutant (hpg) mice.

The hpg mutant mouse lacks the neurohormone gonadotropin-releasing hormone (GnRH) and hence has a reproductive deficit. This deficit can be corrected by placement of normal fetal preoptic area into the third ventricle (see Krieger et al., 1985). We have now used ultrastructural immunocytochemistry to investigate the morphology of GnRH neurons in such intraventricular grafts, the routes that their axons take as they exit into the host, and the neurosecretory terminations that they make in the host median eminence. The GnRH cells in the transplant were similar in morphology to that reported for such cells in the preoptic area of other rodents. There was a large central nucleus, frequently indented and containing 1 or 2 nucleoli. The thin rim of cytoplasm was filled with rough endoplasmic reticulum, Golgi stacks, and mitochondria. Both dendritic and axonal profiles were identified, and a modest synaptic input to the former was found. Between the host and the implant a complex multilayered ependymal zone developed, and it was through this region that GnRH axons exited into the host arcuate nucleus and median eminence, usually surrounded by ependymal or glial elements. Within the median eminence, GnRH terminals were in close association with fenestrated blood vessels forming a normal neurosecretory terminus.