help button home button Endocrine Society Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Cohen, P. E.
Right arrow Articles by Pollard, J. W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Cohen, P. E.
Right arrow Articles by Pollard, J. W.
Endocrinology Vol. 143, No. 4 1413-1422
Copyright © 2002 by The Endocrine Society

NEUROENDOCRINOLOGY

Colony-Stimulating Factor 1 Regulation of Neuroendocrine Pathways that Control Gonadal Function in Mice

Paula E. Cohen1, Liyin Zhu, Kayoko Nishimura and Jeffrey W. Pollard

Center for the Study of Reproductive Biology and Women’s Health (P.E.C., L.Z., K.N., J.W.P.), Departments of Developmental and Molecular Biology and Obstetrics and Gynecology and Women’s Health (P.E.C., J.W.P.), Albert Einstein College of Medicine, Bronx, New York 10461

Address all correspondence and requests for reprints to: Jeffrey W. Pollard, Ph.D., Department of Developmental and Molecular Biology, Albert Einstein College of Medicine, 1300 Morris Park Avenue, Bronx, New York 10461. E-mail: . pollard{at}aecom.yu.edu


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Colony stimulating factor 1 (CSF-1) is the primary regulator of cells of the mononuclear phagocytic lineage. Consequently mice lacking CSF-1 (Csf1op/Csf1op) have depleted populations of macrophages in many tissues. In addition, both sexes have reduced fertility with females having extended estrus cycles and poor ovulation rates, whereas males have low circulating LH and T. In this study, we show that puberty was significantly delayed in Csf1op/Csf1op females compared with control littermates. Restoration of circulating CSF-1 over the first 2 wk of life accelerated puberty, and this treatment until puberty completely corrected the extended estrous cycles. In a standard LH surge induction protocol, Csf1op/Csf1op females showed diminutive negative and no positive feedback response to E2. These data, together with that from male Csf1op/Csf1op mice that showed normal release of LH with a GnRH agonist, indicate that the hypothalamus is the site of the primary defect causing fertility problems in CSF-1-deficient mice. In the hypothalamus, microglia are the only CSF-1 receptor-bearing cells, and the recruitment of a full complement these cells is slightly delayed in Csf1op/Csf1op mice. These data suggest a role for CSF-1 and its target cells, microglia, in establishing the feedback sensitivity to circulating steroid hormones in the hypothalamus of mice.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
COLONY STIMULATING FACTOR 1 (CSF-1) is a homodimeric polypeptide growth factor whose primary function is to regulate the survival, proliferation, differentiation, and function of cells of the mononuclear phagocytic lineage. This lineage includes mononuclear phagocytic precursors, blood monocytes, tissue macrophages, osteoclasts, and microglia of the brain, all of which possess cell surface receptors for CSF-1 (CSF-1R). The CSF-1R, the product of the c-fms protooncogene, is a member of the type III tyrosine kinase receptor family. CSF-1R is also located on cells of the reproductive system, including oocytes and trophoblastic cells (1, 2), suggesting a role for this growth factor outside the hematopoietic system. CSF-1 is found in most tissues of the body including the brain, where it is synthesized in spatial and temporal restricted patterns as well as in the serum, where it is derived from endothelial cells (3, 4). Recent studies that have used a ß-galactosidase reporter gene driven by the CSF-1 promoter have shown synthesis in many cell types including granulosa and Leydig cells of the male and female reproductive tracts (4). In addition, during pregnancy, CSF-1 is induced to high concentrations in the uterine epithelium (5).

Studies of CSF-1 function within the hematopoietic and other systems have been greatly aided by the identification of the osteopetrotic mouse, as being homozygous for a null mutation for the CSF-1 gene. This mutation, originally designated op but now renamed Csf1op, is a single base pair insertion at position 262 (exon 4) of the CSF-1 gene that results in a frame shift and a premature termination of translation (6, 7, 8). Consequently, the protein product is 63 amino acids in length, considerably shorter than the minimum 150 amino acids required for full biological activity (9) consistent with the complete failure to detect CSF-1 protein or activity in these mice (7). The range of phenotypes exhibited by these mice demonstrates the importance of mononuclear phagocytes in numerous biological systems: the mice are deficient in many populations of macrophages, they are toothless and suffer from osteopetrosis as a result of the paucity of osteoclasts, and their immunological responses to certain pathogens are disrupted (10, 11). In support for a role of CSF-1 in reproduction, the homozygous null mutant mice also suffer from reduced fertility at a number of levels: females fail to undergo regular estrous cycles, have a reduced ovulatory frequency and number (12), and are unable to nurture their young (13), whereas the males exhibit low libido and reduced sperm numbers (14). At the endocrine level, female Csf1op/Csf1op mice fail to exhibit the normal preovulatory surge in circulating E2 concentrations that is characteristic of proestrus (12), whereas male Csf1op/Csf1op mice have a 90% reduction in circulating T concentrations (14).

Further studies in male mice have demonstrated that the reduced basal levels of T are due to a 90% reduction in LH concentration despite the fact that the pituitary of Csf1op/Csf1op males responds well to treatment with the potent GnRH analog, histerilin (14). Interestingly, however, when male Csf1op/Csf1op mice are castrated, serum LH concentrations decline (14), in contrast to the rise in circulating gonadotropin predicted by negative feedback regulation. Similarly, when Csf1op/Csf1op males are treated with exogenous T, serum LH concentrations rise (14), instead of the negative feedback-induced drop that would be predicted. These studies suggest that the feedback sensitivity of the hypothalamic-pituitary system is disrupted in Csf1op/Csf1op mice and that this disruption results in deregulation of gonadal function. The neuroendocrine disruption seen in Csf1op/Csf1op mice appears to be at least partially restricted to the hypothalamic-pituitary-gonadal (HPG) axis because adrenal gland function appears normal in Csf1op/Csf1op males (14).

Studies of the HPG axis in Csf1op/Csf1op females are complicated by the cyclical nature of female hormonal regulation. However, their estrous cycles are disrupted such that they reach estrus approximately every 13 d compared with the 4- to 5-d cycles observed in wild-type mice, and this is associated with the absence of a proestrous estrogen surge (12). We hypothesize that this failure of estrous cycling results from a primary disruption of the hypothalamic-pituitary axis, as suggested by the evidence in Csf1op/Csf1op males. The aim of the present studies was to address this possibility by examining estrous cycling and the onset of puberty, two parameters that are regulated by hypothalamic endocrine signals, and by measuring feedback responses to exogenous steroid hormones in ovariectomized (OVX) female mice. We demonstrate that, like male Csf1op/Csf1op mice, the females also show a perturbed hypothalamic-pituitary feedback system to gonadal steroids and that this appears to be due to the failure during the first 2 wk of life to establish this system.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Mice
Heterozygote (+/Csf1op) and homozygote mutant (Csf1op/Csf1op) osteopetrotic mice were housed under conditions of controlled temperature (25 C) and light (on from 0800–2000 h daily) and fed ad libitum with either standard mouse chow or a powdered variety of the same. In addition, the diets of Csf1op/Csf1op mice were supplemented with infant baby formula (Enfamil) once daily. Homozygote mutant mice were identified at 10 d of age by the absence of incisors, whereas heterozygotes were distinguished from wild-type littermates by previously described PCR genotyping method (15). When required, CSF-1 injections (12 ng) were administered sc each day in 0.1 ml 0.9% saline, a regimen that restores at least circulating CSF-1 concentrations (16). All surgical procedures were performed under methoxyfluorane (Metofane) anesthesia, and the mice were left under a heating lamp to recover. All animal procedures were conducted under procedures approved by the Albert Einstein College of Medicine animal care committee.

Timing of puberty
The onset of puberty was assessed in two ways: 1) vaginal opening and 2) the completion of the first and second estrus. Mice were checked daily from 3 wk of age to determine whether vaginal opening had occurred. From the day of opening onwards, vaginal cell samples were taken each day, smeared onto a clean glass slide, and stained lightly with hematoxylin and eosin. Mice were judged as being in one of four stages on the basis of the cellular profile of each smear: proestrus (80–100% intact, live epithelial cells), estrus (100% cornified epithelia), metestrus (~50% cornified epithelia and 50% leukocytes), or diestrus (80–100% leukocytes).

LH surge induction in adult mice
Female mice underwent ovariectomy at 12 wk of age and left for 2 wk to recover. They were then given a standard steroid hormone regimen to induce an LH surge, according to previously published methods (17, 18): briefly, on d 1 of the experiment, the females were each given a single SILASTIC brand (Dow Corning, Midland, MI) implant containing E2 in arachis (peanut) oil at a concentration of 1 µg/ml (19), which produced serum E2 concentrations of less than 50 pg/ml and which were designed to suppress the OVX-induced rise in circulating LH concentrations. At 0900 h on d 6, the mice were given an sc injection of 1 µg E2 benzoate (EB) in arachis oil, followed 24 h later (d 7) by 400 µg progesterone in arachis oil. Animals were killed by exsanguination between 1800 h and 2400 h on d 7, and their serum samples were stored at -70 C until assay. LH RIAs were performed as previously described (14) using standards and antisera generously provided by the National Hormone and Pituitary Program.

Hypothalamic localization of microglia
+/Csf1op and Csf1op/Csf1op mice were killed at embryonic day 14 (E14), E16, and 1, 7, and 11 d post partum (pp) and 15 wk of age. Whole brains were fixed in periodate lysine paraformaldehyde glutaraldehyde fixative overnight at 4 C. After paraffin embedding, serial coronal sections (10 µm) were cut and processed for immunohistochemical localization of: 1) GnRH (using a rabbit polyclonal antibody, R1245, kindly provided by Dr. Terry Nett) (E14, E16, d 1, and d 11 pp); 2) the pan-macrophage marker F4/80 (20), using a rat monoclonal antibody obtained from Dr. M. Chang, Northshore University Hospital, Long Island and purified by CellMax after serum-free culture, as described (21) (d 5, 7, and 11 pp and 15 wk); 3) CSF-1R, the c-fms protooncogene, (using a protein A purified anti-CSF-1R antibody (d 11 pp) (21). Immunohistochemical procedures were performed as described before (14).

For quantitative analysis of GnRH-positive cell numbers, the number of cell bodies was counted in alternate serial sections using the same microscopic field in every section. This field was bounded laterally by the anterior commissures and the lateral ventricles and dorsally by the corpus callosum. For each mouse, at least four alternate sections were counted and the mean cell body numbers assessed per section. For quantitation of F4/80+ and CSF-1R cell surface area, a single field on either side of the third ventricle in the hypothalamus was chosen for counting at a magnification of 25x. Images were captured from these two areas and the area of antibody positivity quantified using NIH Image software with the total numbers being added together from the left and right side fields. The percentage area occupied by F4/80+ or CSF-1R-positive cells was then obtained by division with the total number of pixels in the selected area.

Statistical analysis
All data were analyzed using the Prism statistics package. Mann-Whitney, ANOVA (followed by Dunn’s Multiple posttest), and Student’s or Welch’s t tests were used as appropriate.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Onset of puberty is delayed in Csf1op/Csf1op female mice and corrected by CSF-1 treatment
Two measures were made to assess the onset of puberty, the date of vaginal opening and the onset of two regular estrous cycles. In +/Csf1op females, vaginal opening occurred around 4 wk of age (mean of 34.2 d ± 2.7; Fig. 1Go), but in Csf1op/Csf1op females this was significantly delayed until around 8 wk of age (mean of 52.0 d ± 1.7; P < 0.01; Fig. 1Go). The onset of estrous cycling in Csf1op/Csf1op females was also significantly delayed compared with +/Csf1op littermates (Fig. 1Go). In +/Csf1op females, the first estrous cycle was observed at around 7 wk of age (53.6 d ± 1.1) and the second at 8 wk of age (mean of 58.8 d ± 0.5) and thereafter continued with an approximate 5-d periodicity. In Csf1op/Csf1op females (Fig. 1Go), however, the first cycle was not until at least 10 wk of age (mean of 78.3 d ± 7.7 d), with considerable variation seen between individual animals. The second cycle was then observed between 12 and 14 wk of age (mean of 88.1 d ± 7.4). These delayed cycles are significantly different from that seen in +/Csf1op females (P < 0.01).



View larger version (39K):
[in this window]
[in a new window]
  		Figure 1. Onset of puberty in Csf1op/Csf1op female mice. Puberty was assessed by vaginal opening age (left panel) and onset of estrous cycling (right panel) for heterozygotes (open bars), Csf1op/Csf1op (filled bars), and CSF-1-treated (hatched and cross-hatched bars, respectively) female mice. Numbers on each bar represent the mean age (in days) + SD values for at least 6 females.

 
A critical period for the establishment of the HPG axis is over the first 2 wk of life. Therefore, we analyzed the effects of restoration of circulating CSF-1 through the first 2 wk of life on the onset of puberty. When Csf1op/Csf1op females were treated with CSF-1 from birth, vaginal opening was observed approximately 3 wk earlier (35.0 d ± 2.5 compared with 52.0 d; Fig. 1Go), representing a significantly earlier opening time (P < 0.01) that was not significantly different to that of +/Csf1op females. In contrast, CSF-1 treatment of +/Csf1op mice had no effect on vaginal opening times. Similarly, this treatment significantly accelerated the onset of estrus cycling in Csf1op/Csf1op females by 2 wk, although the timing was more irregular than found in wild-type mice. The first and second estrous cycle being significantly shortened by CSF-1 (P = 0.04 and <0.01 for first and second cycles, respectively). Body weights were also measured in this group with no significant differences being found between the CSF-1 treated and untreated mice in either the wild-type or mutant group (data not shown). These data show that CSF-1 treatment of CSF-1 deficient mice through the first 2 wk of life is essentially sufficient to allow puberty to occur on cue, and this is independent of gross effects on body weight.

Cycle lengths following CSF-1 treatment
We have previously shown that Csf1op/Csf1op females exhibit significantly extended estrous cycles compared with their heterozygote littermates (12). However, when Csf1op/Csf1op females are given daily injections of CSF-1 from birth, their cycle frequency improves to near normal, as assessed by daily vaginal smear analysis. These results suggested that CSF-1 was essential either to establish the adult pattern of estrous cycling and/or to maintain estrous cycling in the adult. Thus, to investigate whether CSF-1 was effective in restoring estrous cycle frequency when given only until puberty, this experiment was repeated by injecting pups with CSF-1 from d 2 pp until 6 wk of age, and monitoring cycle frequency in the adult females. The effect of CSF-1 injections throughout life and only up until 6 wk of age was the same, as shown in Fig. 2AGo. In both cases Csf1op/Csf1op females showed a significant improvement in cycle lengths compared with untreated Csf1op/Csf1op females. In the latter case, females reached estrus once every 13–14 d on average, whereas in CSF-1-treated mice, estrus was achieved every 5–6 d (Fig. 2AGo). This improvement was the same for mice treated with CSF-1 throughout life and for those treated only prepubertally. The mean cycle length for these CSF-1-treated Csf1op/Csf1op groups was not statistically different from that of the untreated +/Csf1op group (Fig. 2AGo).



View larger version (26K):
[in this window]
[in a new window]
  		Figure 2. Estrous cycles lengths in Csf1op/Csf1op female mice. Mean estrous cycle length (A) and length of each cycle stage (B) for heterozygotes (open bars), Csf1op/Csf1op (filled bars), and CSF-1-treated Csf1op/Csf1op (throughout life, hatched bars and until puberty, cross-hatched bars, respectively) female mice. Error bars depict SD values. M, Metestrus; D, diestrus; P, proestrus; E, estrus.

 
The length of time spent at each stage of the estrous cycle was also not different between +/Csf1op and CSF-1-treated Csf1op/Csf1op females (Fig. 2BGo). In all three of these treatment groups, about 30–35% of the cycle is devoted to metestrus, 15–20% to diestrus, 20–30% to proestrus, and 20–30% to estrus. By contrast, Csf1op/Csf1op females that remain untreated have extended periods of diestrus, with up to 50% of their time spent at this stage (Fig. 2BGo).

LH surge induction
To investigate the cause of the extended estrous cycles and the flat E2 profile through the cycle, we explored the responsiveness of the hypothalamic-pituitary axis to both positive and negative feedback. The removal of the endogenous source of sex steroids (by ovariectomy) produces a negative feedback effect that results in a compensatory rise in circulating LH concentrations (Fig. 3Go) from baseline levels during the estrous cycle (data not shown). Such an increase in circulating LH was observed in both the +/Csf1op and Csf1op/Csf1op females, although the extent of the increase in Csf1op/Csf1op females was only 15.7% of that seen in +/Csf1op mice (3.1 ± 2.0 and 20.1 ± 3.8 ng/ml, respectively; P < 0.003, Mann-Whitney U test). This increased circulating LH was then suppressed using a low-dose E2 implant containing a solution of 1 µg/ml E2 in arachis oil. Within 24 h of implant placement in both +/Csf1op and Csf1op/Csf1op females, circulating LH concentrations were reduced to within the lower limits of RIA sensitivity (0.330 ± 0.179 and 0.272 ± 0.150 ng/ml, respectively).



View larger version (17K):
[in this window]
[in a new window]
  		Figure 3. Loss of sex steroid hormone feedback in female Csf1op/Csf1op mice. Serum LH concentrations following ovariectomy and LH surge-induction hormonal regimen in +/Csf1op (open bars) and Csf1op/Csf1op (filled bars) females. OVX females were then treated with low-dose estrogen pellets for 6 d, followed by a bolus injection of EB on d 6, supplemented with a progesterone dose on d 7. LH surges were monitored by blood sampling on the evening of d 7 between 1800 h and 2400 h OVX group (first column) received only oil vehicle pellets and injections for 7 d. *, P values are given in the accompanying text.

 
On d 6 after implant placement, the mice were given a bolus injection of 1 µg EB in arachis oil, followed 24 h later (d 7) by 400 µg progesterone in arachis oil, and serum LH concentrations were monitored at 2-h intervals from 1800 h on d 7 until 2400 h on the same day. At 1800 h, serum LH concentrations were unchanged from basal levels in both +/Csf1op and Csf1op/Csf1op females. By 2000 h, serum LH concentrations in +/Csf1op females were dramatically increased by 121-fold over the basal concentrations seen at 1800 h (44.33 ± 7.16 ng/ml and 0.37 ± 0.20 ng/ml, respectively). In contrast, Csf1op/Csf1op females showed only a modest 3.7-fold increase at 2000 h over basal levels that was not of statistical significance (1.01 ± 0.29 ng/ml and 0.28 ± 0.14 ng/ml, respectively). By 2200 h, the serum LH concentrations in +/Csf1op females were on the decline, although still significantly raised above the baseline concentrations, whereas serum LH concentrations in Csf1op/Csf1op females had returned to baseline levels. By midnight, 2 h later, the serum LH concentrations in +/Csf1op females had also returned to within the baseline range. In all cases, except at 1800 h, the difference in serum LH concentrations between genotypes was statistically significant, even at midnight when basal levels were restored (P < 0.0001, 0.03 and 0.05 for 2000 h, 2200 h, and 2400 h, respectively; Mann-Whitney U test). Thus, in summary, Csf1op/Csf1op females exhibit a small, but significant, increase in circulating LH following ovariectomy but then fail to respond to LH surge induction protocols.

Hypothalamic localization of GnRH neurons in CSF-1-deficient mice
The localization of GnRH neurons within and migration into the hypothalamus of +/Csf1op and Csf1op/Csf1op mice was examined by immunohistochemical methods using an anti-GnRH antibody as a marker. Analysis during development on E14, E16, and d 1 pp showed comparable migration between genotypes of the GnRH neurons from the olfactory placode up into the hypothalamus (data not shown). By d 11 pp, when the GnRH neurons had reached their final location in the thalamus and hypothalamus, there was no significant difference in their numbers in +/Csf1op and Csf1op/Csf1op females, and a slight, but not significant decrease in the number of GnRH-positive cells found in Csf1op/Csf1op males (Table 1Go). These neurons were also found distributed throughout the thalamus and hypothalamus in a similar fashion (data not shown).


View this table:
[in this window]
[in a new window]
  		Table 1. Immunohistochemical analysis of cellular components of the hypothalamus in +/Csf1op and Csf1op/Csf1op mice on d 11 pp

 
Hypothalamic localization of microglia in CSF-1- deficient mice
To determine the distribution of microglia throughout the hypothalamus of male and female +/Csf1op and Csf1op/Csf1op mice, immunohistochemical localization of F4/80 antigen was performed. In the central nervous system, the localization of F4/80 antigen is restricted to microglia because these cells are the only mononuclear phagocytic lineage cells existing in the CNS (22, 23). On d 5 pp in both genotypes, F4/80-positive (F4/80+) cells were abundant on the meninges (pia mater) with a few cells in the brain parenchyma in a perivascular location (data not shown). By d 7 pp F4/80+ cells were observed throughout the brain of both sexes, as demonstrated by other investigators (23, 24). In particular, F4/80+ cells were found throughout the hypothalamus of +/Csf1op females as early as d 7 pp, and persisted through to adulthood (Fig. 4Go, A and C). On d 11 pp when wild-type mice showed adult-like distributions (Fig. 4AGo), the relative densities of microglia found within hypothalamic sections was similar in heterozygote males and females (Table 1Go). In Csf1op/Csf1op females, the density of F4/80 positive cells within the hypothalamus was slightly (~70% of wild-type density) but significantly diminished in prepubertal mice compared with +/Csf1op females at the same age (Fig. 4BGo). Similarly, F4/80+ cells were also significantly diminished within the hypothalamus of Csf1op/Csf1op males compared with heterozygote or wild-type littermates (Table 1Go). However, the distribution of these cells was similar in both genotypes (Fig. 4Go, A and B). In adult female mice however, the F4/80+ cell density was the same between genotypes (Fig. 4Go, C and D; 3.77 ± 0.22 vs. 3.34 ± 0.19, mean % area of hypothalamus, ± SEM, wild-type and mutant female respectively; not significantly different). The distribution and density of microglia were recapitulated in sections of the hypothalamus from male mice stained with an antibody against the CSF-1 receptor (Table 1Go; Fig. 4Go, E and F), where the number of receptor-positive cells within the hypothalamus was significantly reduced by approximately 30% in Csf1op/Csf1op males compared with their heterozygote littermates. Thus, the number of c-fms positive cells was mirrored by the numbers of F4/80+ cells localized within the same region of the hypothalamus, suggesting that most, if not all, of the CSF-1R-bearing cells are microglia (Table 1Go; Fig. 4Go, E and F). These data were confirmed in adult female mice that also showed a pattern of CSF-1R-bearing cells that corresponded to the distribution of microglia (data not shown).



View larger version (129K):
[in this window]
[in a new window]
  		Figure 4. F4/80+ and CSF-1 receptor positive cells in the hypothalamus of mice. Localization of F4/80-positive cells (A–D) and c-fms (E and F) in the hypothalami of +/Csf1op (A, C, E) and Csf1op/Csf1op (B, D, F) females on d 11 pp (A–B) and in the adult (C–F) and in males on d 11 pp (E, F). Positive immunoreactive signal is shown as a brown precipitate against blue counterstaining. Bar, 50 µm.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
We have previously shown that CSF-1 is important for normal male fertility in mice. In Csf1op/Csf1op males, the absence of bioactive CSF-1 results in lower sperm numbers and reduced libido, associated with a 90% reduction in circulating T and LH (14, 15). More significantly, however, these males fail to demonstrate normal hypothalamic-pituitary responses to changes in circulating androgens, suggesting that the primary cause of their infertility is a failure of the feedback control in the hypothalamus. Based on these observations, we previously hypothesized that CSF-1 acts through testicular macrophages to modulate Leydig cell steroidogenesis, but that the main action of CSF-1 in reproductive regulation was to guide, via resident hypothalamic microglia, the development of the intricate neuronal networks that constitute the HPG axis (25).

In concordance with our hypothesis, we also showed that female Csf1op/Csf1op mice have severe fertility defects, manifested by extended estrous cycles that failed to display the preovulatory surge in serum E2, poor ovulation rates, and reduced ovarian sensitivity to exogenous gonadotropin stimulation (12). In females, however, the situation is complicated by the fact that the CSF-1R is localized to the oocyte in the developing follicle and macrophages that are recruited around these growing follicles (12, 26). Thus, the reduced ovulation rates in Csf1op/Csf1op females might reflect insufficient CSF-1 induced events at the level of the ovary instead of, or as well as, CSF-1-functions in the hypothalamus or pituitary. Indeed, studies in this and other laboratories have indicated that the ovary is capable of responding by increased ovulation to locally applied CSF-1 in vivo (12, 27), and that ovulation rates are improved in Csf1op/Csf1op follicular cultures treated with CSF-1 (our unpublished observations). To separate these effects in the current studies, we performed classical LH surge experiments on mice to examine the hypothalamic-pituitary axis independent of the ovary. These studies showed that female Csf1op/Csf1op mice had a significantly reduced negative feedback response to the removal of estrogens by ovariectomy and, a complete absence of positive feedback in response to the surge protocol when compared with wild-type mice.

Thus, both male and female Csf1op/Csf1op mice showed evidence that the sex steroid hormone feedback control in the hypothalamus and pituitary is abnormal. Because in male mice LH could be released to wild-type concentrations by treatment with a GnRH agonist (14), the data support the hypothesis that the primary defect resulting from the absence of CSF-1 is in the hypothalamus. A disruption of this system is also supported by the delayed puberty measured by vaginal opening and onset of estrus cycles seen in Csf1op/Csf1op mice. The requirement for CSF-1 in this process appears to be developmental and not continuous because the aberrant estrus cycles displayed by Csf1op/Csf1op mice could be corrected by treatment to puberty. Furthermore, the delay in puberty can be largely corrected by restoration of circulating CSF-1 concentrations in mice over the first 14 d of life. Together, these data support the hypothesis that CSF-1 acts in the brain during the period of hypothalamic imprinting, to organize the appropriate neuronal circuitry that regulates GnRH responsiveness to sex steroid hormones. However, once established the data argues against a continuous role for CSF-1 in regulating the HPG axis because CSF-1 treatment up to puberty results in the correction of estrus cyclicity in adults now devoid of CSF-1.

While some evidence exists to show that CSF-1Rs are expressed in neuronal cell populations after injury (28), there is no evidence to suggest it is expressed in these cells during development (3). Indeed, our present data suggest that all c-fms-positive cells within the hypothalamus are also F4/80-positive, suggesting that only the microglial cells of the hypothalamus are capable of expressing the CSF-1 receptor upon their cell surface during development. All microglia apparently express the CSF-1R (29), and they are found throughout the central nervous system, comprising approximately 5–15% of all cells in the adult brain (30). These cells are derived from the bone marrow and are members of the mononuclear phagocytic lineage (23, 24, 30). Embryologically, they first appear in the central nervous system of mice at around d 16 (23, 31). By adulthood, the microglia are more ramified and highly dendritic and are present throughout the brain. Microglia that are associated with the hypothalamic regions are notably highly ramified and dendritic (24). The present studies support these earlier observations, showing that microglia are found throughout the hypothalamus of wild-type mice from d 7 pp and reaching wild-type densities by d 11 pp. In contrast, their density is reduced in Csf1op/Csf1op mice during development by approximately 30% compared with wild-type mice, although they reach normal densities by adulthood. These data are consistent with those reported by other groups that have shown regional effects on microglial cell populations due to the absence of CSF-1 in adult mice, with some but not all populations affected (32, 33). It should also be appreciated that the residual microglia will lack all CSF-1 receptor mediated functions.

CSF-1 is also expressed in the brain during development in regional-specific ways, suggesting a direct local effect on microglia (3). Although it is clear from our data that there are also other chemokines/cytokines that are more important for the regulation of microglial densities because these are only slightly reduced in the neonate and are normal in adult Csf1op/Csf1op mice. In cultures of embryonic brains, CSF-1 acts as a trophic factor promoting neuronal viability and enhancing process outgrowth. These effects were only observed in mixed cell cultures and not in cultures containing only neurons (3). An explanation of the current data would be that CSF-1 through its actions on microglia provides trophic cues to GnRH neurons either promoting their viability or guidance during their migration. GnRH neurons arising in the olfactory placode migrate embryonically to become distributed throughout the hypothalamus, particularly the arcuate nucleus, medial, and rostral proptic area (m/rPOA) and medial basal hypothalamus (MBH), as well as in the medial septal region (34). However, analysis of the numbers of GnRH neurons and their migration in Csf1op/Csf1op mice failed to demonstrate any effects of the absence of CSF-1 on either of these parameters, and GnRH neurons appear to be correctly positioned in the hypothalami of these mutant mice. Thus, CSF-1 does not appear to be providing viability or guidance cues to GnRH neurons.

Studies by other groups have demonstrated that the positive and negative sex steroid hormone feedback loops are mediated by different neuronal populations in the hypothalamus. The anteroventral periventricular nucleus and organum vasculosum of the lamina terminalis, both located in the rPOA, are critical for the positive feedback response that results in the preovulatory LH surge (34). Lesion of the anteroventral periventricular result in persistent estrus and abolishes the estrogen-induced LH surge (35, 36, 37, 38), whereas E2 implants into this region result in spontaneous LH surges in rats (39, 40). By contrast, similar implants placed at the level of the MBH, amygdala, or hippocampus fail to elicit LH surges (39, 40). The inhibitory pathways for estrogen regulation of GnRH secretion have been more difficult to define, due in part to variability between species in the negative feedback effects of estrogen. However, the consensus of opinion favors the MBH as the site of negative feedback by estrogen, because deafferenation of this region in rats results in similar levels of EB-induced suppression of LH secretion (41). Besides these spatial differences between the positive and negative feedback loops, it appears that the timing of these events is also different, positive feedback requiring at least 12–24 h of estrogen exposure, and negative feedback occurring within 30 min of estrogen treatment (34, 41).

Numerous stimulatory and inhibitory neuropeptides have been reported to regulate GnRH neuronal activity, both at the rPOA and the MBH (34). However, only a subset of these effectors has been shown to be involved in positive feedback, including N-methyl-D-aspartate, {gamma}-aminobutyric acid (GABA), NPY, neurotensin, noradrenaline, and ß-endorphin (ßEND). Of these, only GABA is thought to participate in both positive and negative feedback, whereas ßEND is known to be specific for positive feedback responses to estrogen (reviewed in Ref. 34). Both of these neuropeptides are inhibitory to GnRH neurons, and thus would have to be negatively regulated by estrogen to produce a positive feedback response. Indeed, ßEND neurons, which are found exclusively in the arcuate nucleus of the hypothalamus are responsive to estrogen and send projections to the dendrites and soma of the GnRH neurons in the vicinity (42, 43). Similarly, GABA terminals are also found to synapse on GnRH neurons in the rPOA (44, 45, 46), and these GnRH neurons also contain GABAA receptors (47, 48, 49, 50). Most importantly, both ßEND and GABA are down-regulated before the GnRH surge (45, 46, 51). Interestingly, previous studies in this laboratory have demonstrated that Csf1op/Csf1op mice exhibit altered/dulled intracortical responses to the GABAA antagonist, bicuculline methiodide, suggesting that the GABAergic circuitry is altered in these mice (3). This may be part of the explanation of the hypothalamic defects in these CSF-1 mutant mice.

In this context, during development, microglia are thought to be important for remodeling and arborization of neuronal networks. They are often associated with dying neurons, which they are able to phagocytose (52), and they are also able to nip off and engulf neurosecretory termini and thus can alter the arborization patterns of neurosecretory neurons (52). Like their macrophage counterparts in peripheral tissues, microglia can secrete a variety of cytokines, including nerve growth factor, IL-1, IL-6, TNF{alpha} as well as angiogenic factors (24). Such cytokines could be important in the development of neural networks. Recent reports have also suggested that microglia contain ER immunoreactivity and ERß transcripts are readily detectable by RT-PCR in microglia cells in vitro (53). Naftolin and colleagues have suggested that microglia respond to estrogen by remodeling inhibitory synapses from arcuate neurons that would then permit the GnRH surge to progress (54) and, in the context of our studies, maybe important for the establishment of this steroid-hormone responsiveness.

In summary, our data has shown important roles for CSF-1 in the establishment of the sex steroid hormone feedback regulatory system in the hypothalamus during the first 2 wk of life. In the absence of CSF-1 the acquisition of a full complement of microglia into the hypothalamus is delayed and of course, CSF-1 regulated microglial functions will be absent in the residual cells. Consequently, the data (summarized in Fig. 5Go) strongly implicates CSF-1-regulated microglia through their actions in establishing appropriate neuronal circuitry are essential for setting up the hypothalamic feedback system in male and female mice, and for setting the feedback sensitivity appropriately in both sexes.



View larger version (25K):
[in this window]
[in a new window]
  		Figure 5. A model for the action of CSF-1 on the development of sex-steroid feedback in the responses in the hypothalamus. The figure indicates the action of CSF-1-regulated microglia on the establishment of neuronal circuitry between GABA/opioid intermediary, excitory amino acid and GnRH neurons. The absence of CSF-1, caused by the Csf1op mutation (op), at critical stages of development leads to this circuitry being perturbed as a consequence of lost microglial functions important for the establishment of appropriate neuronal circuitry through their remodeling capacity. This results in a compromised sex steroid hormone (E) feedback system in the hypothalamus of both male and female mice.

 


   Footnotes
 
This research was supported by NIH Grants RO1-HD-30820 and 35627.

1 Current address: Department of Molecular Genetics, Albert Einstein College of Medicine, 1300 Morris Park Avenue, Bronx, New York 10461. Back

Abbreviations: ßEND, ß-Endorphin; CSF-1, colony-stimulating factor-1; CSF-1R, cell surface receptor for CSF-1; E, embryonic day; EB, E2 benzoate; GABA, {gamma}-aminobutyric acid; HPG, hypothalamic-pituitary-gonadal; MBH, medial basal hypothalamus; m/rPOA, medial, and rostral proptic area; OVX, ovariectomized; pp, post partum.

Received October 5, 2001.

Accepted for publication December 19, 2001.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Arceci RJ, Pampfer S, Pollard JW 1992 Expression of CSF-1/c-fms and SF/c-kit mRNA during preimplantation mouse development. Dev Biol 151:1–8[CrossRef][Medline]
  2. Arceci RJ, Shanahan F, Stanley ER, Pollard JW 1989 Temporal expression and location of colony-stimulating factor 1 (CSF-1) and its receptor in the female reproductive tract are consistent with CSF-1-regulated placental development. Proc Natl Acad Sci USA 86:8818–8822[Abstract/Free Full Text]
  3. Michaelson MD, Bieri PL, Mehler MF, Xu H, Arezzo JC, Pollard JW, Kessler JA 1996 CSF-1 deficiency in mice results in abnormal brain development. Development 122:1–12[Abstract]
  4. Ryan GR, Dai XM, Dominguez MG, Tong W, Chuan F, Chisholm O, Russell RG, Pollard JW, Stanley ER 2001 Rescue of the colony-stimulating factor 1 (CSF-1)-nullizygous mouse (Csf1op/Csf1op) phenotype with a CSF-1 transgene and identification of sites of local CSF-1 synthesis. Blood 98:74–84[Abstract/Free Full Text]
  5. Pollard JW, Bartocci A, Arceci RJ, Orlofsky A, Ladner MB, Stanley ER 1987 Apparent role of the macrophage growth factor, CSF-1, in placental development. Nature 330:484–486[CrossRef][Medline]
  6. Yoshida H, Hayashi S-i, Kunisada T, Ogawa M, Nishikawa S-I, Okamura H, Sudo T, Shultz LD 1990 The murine mutation osteopetrosis in the coding region of the macrophage colony stimulating factor gene. Nature 345:442–444[CrossRef][Medline]
  7. Wiktor-Jedrzejczak W, Bartocci A, Ferrante Jr AW, Ahmed-Ansari A, Sell KW, Pollard JW, Stanley ER 1990 Total absence of colony-stimulating factor 1 in the macrophage-deficient osteopetrotic (op/op) mouse [published erratum appears in Proc Natl Acad Sci USA 1991 Jul 1;88(13):5937]. Proc Natl Acad Sci USA 87:4828–4832[Abstract/Free Full Text]
  8. Pollard JW, Hunt JS, Wiktor-Jedrzejczak W, Stanley ER 1991 A pregnancy defect in the osteopetrotic (op/op) mouse demonstrates the requirement for CSF-1 in female fertility. Dev Biol 148:273–283[CrossRef][Medline]
  9. Kawasaki ES, Ladner MB 1990 Molecular biology of macrophage colony stimulating factor. In: Dexter TM, Garland JM, Testa NG, eds. Colony stimulating factors: molecular and cellular biology. New York: Marcel Dekker, Inc.; 155–176
  10. Pollard JW, Stanley ER 1996 Pleiotropic roles for CSF-1 in development defined by the mouse mutation osteopetrotic (op). Adv Dev Biochem 4:153–193
  11. Guleria I, Pollard JW 2000 The trophoblast is a component of the innate immune system during pregnancy. Nat Med 6:589–593[CrossRef][Medline]
  12. Cohen PE, Zhu L, Pollard JW 1997 The absence of CSF-1 in osteopetrotic (csfmop/csfmop) mice disrupts estrous cycles and ovulation. Biol Reprod 56:110–118[Abstract]
  13. Pollard JW, Hennighausen L 1994 Colony stimulating factor 1 is required for mammary gland development during pregnancy. Proc Natl Acad Sci USA 91:9312–9316[Abstract/Free Full Text]
  14. Cohen PE, Hardy MP, Pollard JW 1997 Colony-stimulating factor-1 plays a major role in the development of reproductive function in male mice. Mol Endocrinol 11:1636–1650[Abstract/Free Full Text]
  15. Cohen PE, Chisholm O, Arceci RJ, Stanley ER, Pollard JW 1996 Absence of colony stimulating factor-1 in osteopetrotic (csfmop/csfmop) mice results in male fertility defects. Biol Reprod 55:310–317[Abstract]
  16. Wiktor-Jedrzejczak W, Urbanowska E, Aukerman SL, Pollard JW, Stanley ER, Ralph P, Ansari AA, Sell KW, Szperl M 1991 Correction by CSF-1 of defects in the osteopetrotic op/op mouse suggests local, developmental and humoral requirements for this growth factor. Exp Hematol 19:1049–1054[Medline]
  17. Bronson FH 1981 The regulation of luteinizing hormone secretion by estrogen: relationships among negative feedback, surge potential, and male stimulation in juvenile, peripubertal, and adult female mice. Endocrinology 108:506–516[Abstract]
  18. Attardi B 1984 Progesterone modulation of the luteinizing hormone surge: regulation of hypothalamic and pituitary progestin receptors. Endocrinology 115:2113–2122[Abstract]
  19. Cohen PE, Milligan SR 1993 Silastic implants for delivery of oestradiol to mice. J Reprod Fertil 99:219–223[Abstract]
  20. Hume DA, Gordon S 1983 The mononuclear phagocyte system of the mouse defined by immunohistochemical localization of antigen F4/80: identification of resident macrophages in renal medullary and cortical interstitium and the juxtaglomerular complex. J Exp Med 157:1704–1709[Abstract/Free Full Text]
  21. Gouon-Evans V, Rothenberg ME, Pollard JW 2000 Postnatal mammary gland development requires macrophages and eosinophils. Development 127:2269–2282[Abstract]
  22. Hume DA, Robinson AP, Macpherson G, Gordon S 1983 The mononuclear phagocyte system of the mouse defined by immunohistochemical localization of antigen F4/80. J Exp Med 158:1522–1536[Abstract/Free Full Text]
  23. Perry VH, Gordon S 1988 Macrophages and microglia in the nervous system. Trends Neurosci 11:273–277[CrossRef][Medline]
  24. Perry VH, Gordon S 1991 Macrophages and the nervous system. Int Rev Cytol 125:203–244[Medline]
  25. Cohen PE, Nishimura K, Zhu L, Pollard JW 1999 Macrophages: important accessory cells for reproductive function. J Leukoc Biol 66:765–772[Abstract]
  26. Arceci RJ, Pampfer S, Pollard JW 1992 Role and expression of colony-stimulating factor-1 and steel factor receptors and their ligands during pregnancy in the mouse. Reprod Fertil Dev 4:619–632
  27. Araki M, Fukumatsu Y, Katabuchi H, Schultz LD, Takahashi K, Okamura H 1996 Follicular development and ovulation in macrophage colony-stimulating factor-deficient mice homozygous for the osteopetrosis (op) mutation. Biol Reprod 54:478–484[Abstract]
  28. Wang Y, Berezovska O, Fedoroff S 1999 Expression of colony stimulating factor-1 receptor (CSF-1R) by CNS neurons in mice. J Neurosci Res 57:616–632[CrossRef][Medline]
  29. Brosnan CF, Shafit-Zagardo B, Aquino DA, Berman NEJ 1993 Expression of monocyte/macrophage growth factors and receptors in the central nervous system. Adv Neurol 59:349–361[Medline]
  30. Lawson LJ, Perry VH, Dri P, Gordon S 1990 Heterogeneity in the distributing and morphology of microglia in the normal adult mouse brain. Neuroscience 39:151–170[CrossRef][Medline]
  31. Dalmau I, Finsen B, Tonder N, Zimmer J, Gonzalez B, Castellano B 1997 Development of microglia in the prenatal rat hippocampus. J Comp Neurol 377:70–84[CrossRef][Medline]
  32. Wegiel J, Wisniewski HM, Dziewiatkowski J, Tarnawski M, Kozielski R, Trenkner E, Wiktor-Jedrzejczak W 1998 Reduced number and altered morphology of microglial cells in colony stimulating factor-1-deficient osteopetrotic op/op mice. Brain Res 804:135–139[CrossRef][Medline]
  33. Blevins G, Fedoroff S 1995 Microglia in colony-stimulating factor 1-deficient op/op mice. J Neurosci Res 40:535–544[CrossRef][Medline]
  34. Herbison AE 1998 Multimodal influence of estrogen upon gonadotropin-releasing hormone neurons. Endocr Rev 19:302–330[Abstract/Free Full Text]
  35. Ronnekleiv OK, Kelly MJ 1986 Luteinizing hormone-releasing hormone neuronal system during the estrous cycle of the female rat: effects of surgically induced persistent estrus. Neuroendocrinology 43:564–576[Medline]
  36. Ronnekleiv OK, Kelly MJ 1988 Plasma prolactin and luteinizing hormone profiles during the estrous cycle of the female rat: effects of surgically induced persistent estrus. Neuroendocrinology 47:133–141[Medline]
  37. Wiegand SJ, Terasawa E, Bridson WE, Goy RW 1980 Effects of discrete lesions of preoptic and suprachiasmatic structures in the female rat. Alterations in the feedback regulation of gonadotropin secretion. Neuroendocrinology 31:147–157[Medline]
  38. Wiegand SJ, Terasawa E 1982 Discrete lesions reveal functional heterogeneity of suprachiasmatic structures in regulation of gonadotropin secretion in the female rat. Neuroendocrinology 34:395–404[CrossRef][Medline]
  39. Goodman RL 1978 The site of the positive feedback action of estradiol in the rat. Endocrinology 102:151–159[Medline]
  40. Goodman RL 1978 A quantitative analysis of the physiological role of estradiol and progesterone in the control of tonic and surge secretion of luteinizing hormone in the rat. Endocrinology 102:142–150[Medline]
  41. Blake CA, Norman RL, Sawyer CH 1974 Localization of inhibitory actions of estrogen and nicotine on release of luteinizing hormone in rats. Neuroendocrinology 16:22–35[Medline]
  42. Morrell JI, McGinty JF, Pfaff DW 1985 A subset of ß-endorphin- or dynorphin-containing neurons in the medial basal hypothalamus accumulates estradiol. Neuroendocrinology 41:417–426[Medline]
  43. Jirikowski GF, Merchenthaler I, Rieger GE, Stumpf WE 1986 Estradiol target sites immunoreactive for ß-endorphin in the arcuate nucleus of rat and mouse hypothalamus. Neurosci Lett 65:121–126[CrossRef][Medline]
  44. Jarry H, Leonhardt S, Wuttke W 1991 {gamma}-Aminobutyric acid neurons in the preoptic/anterior hypothalamic area synchronize the phasic activity of the gonadotropin-releasing hormone pulse generator in ovariectomized rats. Neuroendocrinology 53:261–267[Medline]
  45. Herbison AE, Chapman C, Dyer RG 1991 Role of medial preoptic GABA neurones in regulating luteinising hormone secretion in the ovariectomised rat. Exp Brain Res 87:345–352[Medline]
  46. Herbison AE, Dyer RG 1991 Effect on luteinizing hormone secretion of GABA receptor modulation in the medial preoptic area at the time of proestrous luteinizing hormone surge. Neuroendocrinology 53:317–320[Medline]
  47. Jung H, Shannon EM, Fritschy JM, Ojeda SR 1998 Several GABAA receptor subunits are expressed in LHRH neurons of juvenile female rats. Brain Res 780:218–229[CrossRef][Medline]
  48. Barraclough CA 1992 Neural control of the synthesis and release of luteinizing hormone-releasing hormone. Ciba Found Symp 168:233–246[Medline]
  49. Barraclough CA 1994 Neurotransmitter regulation of luteinizing hormone-releasing hormone neuronal function. Acta Biol Hung 45:189–206[Medline]
  50. Bergen HT, Hejtmancik JF, Pfaff DW 1991 Effects of {gamma}-aminobutyric acid receptor agonists and antagonist on LHRH-synthesizing neurons as detected by immunocytochemistry and in situ hybridization. Exp Brain Res 87:46–56[Medline]
  51. Herbison AE, Heavens RP, Dyer RG 1990 Endogenous release of {gamma}-aminobutyric acid from the medial preoptic area measured by microdialysis in the anaesthetised rat. J Neurochem 55:1617–1623[CrossRef][Medline]
  52. Pow DV, Perry VH, Morris JF, Gordon S 1989 Microglia in the neurohypophysis associate with and endocytose terminal portions of neurosecretory neurons. Neuroscience 33:567–578[CrossRef][Medline]
  53. Mor G, Nilsen J, Horvath T, Bechmann I, Brown S, Garcia-Segura LM, Naftolin F 1999 Estrogen and microglia: a regulatory system that affects the brain. J Neurobiol 40:484–496[CrossRef][Medline]
  54. Naftolin F, Leranth C, Horvath TL, Garcia-Segura LM 1996 Potential neuronal mechanisms of estrogen actions in synaptogenesis and synaptic plasticity. Cell Mol Neurobiol 16:213–223[CrossRef][Medline]



This article has been cited by other articles:


Home page
 Hum Reprod UpdateHome page
W. V. Ingman and R. L. Jones
Cytokine knockouts in reproduction: the use of gene ablation to dissect roles of cytokines in reproductive biology
Hum. Reprod. Update, March 1, 2008; 14(2): 179 - 192.
[Abstract] [Full Text] [PDF]

Home page
 EndocrinologyHome page
W. V. Ingman, R. L. Robker, K. Woittiez, and S. A. Robertson
Null Mutation in Transforming Growth Factor {beta}1 Disrupts Ovarian Function and Causes Oocyte Incompetence and Early Embryo Arrest
Endocrinology, February 1, 2006; 147(2): 835 - 845.
[Abstract] [Full Text] [PDF]

Home page
 Hum Reprod UpdateHome page
R. Wu, K. H. Van der Hoek, N. K. Ryan, R. J. Norman, and R. L. Robker
Macrophage contributions to ovarian function
Hum. Reprod. Update, March 1, 2004; 10(2): 119 - 133.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Cohen, P. E.
Right arrow Articles by Pollard, J. W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Cohen, P. E.
Right arrow Articles by Pollard, J. W.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Endocrinology Endocrine Reviews J. Clin. End. & Metab.
Molecular Endocrinology Recent Prog. Horm. Res. All Endocrine Journals