| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Integrative Physiology and the Center for Neuroscience (J.C.G., P.-S.T.), University of Colorado, Boulder, Colorado 80309-0354; and Departments of Internal Medicine and Cell Biology (S.M.M.), University of Virginia, Charlottesville, Virginia 22908
Address all correspondence and requests for reprints to: John C. Gill, Department of Integrative Physiology, UCB 354 Clare Small 113, University of Colorado, Boulder, Colorado 80309-0354. E-mail: john.gill{at}colorado.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The complexity of GnRH neuronal development underscores the importance of signaling molecules that regulate GnRH neuronal differentiation. Although a few factors, such as AP-2
(8), nasal embryonic LHRH factor (9), and 
-aminobutyric acid (10, 11) have been implicated in specific stages of GnRH neuronal development, factors with consistent and potent trophic effects on the developing GnRH system have not been definitively identified. Recent reports indicate a causal relationship between the loss of function mutation in fibroblast growth factor (FGF) receptor 1 and Kallmann syndrome, a pathology characterized by complete or partial loss of GnRH function and anosmia, suggests the possible involvement of FGF signaling in the formation of the GnRH system (12, 13). To investigate this possibility, we examined whether FGF signaling critically regulates GnRH neuronal development. FGFs are molecules that exert a broad spectrum of neurotrophic and differentiative effects on developing neurons (reviewed in Ref.14). FGFs are highly neurotrophic in the immortalized GnRH neuronal cell lines, GT1 cells, and GT1 cells express FGF receptors (FGFRs) (15, 16). However, because of the difficulty associated with studying a highly scattered and scarce population of neurons, these findings, with a few exceptions (17), have not been extended to endogenous GnRH neurons during development.
In this study, we tested the hypothesis that FGF signaling critically regulates endogenous GnRH neurons during development. First, to verify the ability of GnRH neurons to respond directly to FGF signaling, we performed single and dual immunocytochemistry (ICC) to demonstrate the presence of FGFRs in the nasal placodes and developing GnRH neurons. Second, using a variety of in vitro culture systems and a transgenic mouse model expressing the enhanced green fluorescent protein (GFP) in GnRH neurons (18), we tested the differentiative and neurotrophic responses of developing GnRH-GFP neurons to FGF signaling during stages of cell fate specification and neurite extension. These results should contribute substantially to the understanding of how FGF signaling regulates a maturing neuroendocrine system.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Localization of FGFRs in developing nasal placode and GnRH neurons by ICC
Timed-pregnant nontransgenic females or P3 mice were killed, and then embryos (E10.5 and E15.5) or brains from P3 mice were removed and immersion fixed in 4% paraformaldehyde for 4–6 h, followed by cryoprotection in 20% sucrose overnight. These three time points (E10.5, E15.5, and P3) each represent a unique stage of GnRH neuronal development, i.e. fate specification (E10.5), axon targeting (E15.5), and full maturation (P3). GnRH neuronal migration occurs between approximately E11 and E15, and this process was not examined in this report. Nontransgenic mice, rather than GnRH-GFP mice, were used because we believe the initial characterization on the distribution of FGFRs should be performed in the absence of transgenic proteins that might confound our data interpretation.
Frozen sections of embryonic heads or postnatal brains were cut at 13- to 16-µm thicknesses and thaw mounted onto poly-L-lysine-coated slides. Before staining, sections were incubated in a mixture of 10% methanol and 1% H2O2 in PBS containing 0.4% Triton-X100 (PBST) to quench the endogenous peroxidase activity. After PBST washes, sections were incubated in primary antibodies against FGFRs. All FGFR antisera were obtained from Santa Cruz Biotechnologies (Santa Cruz, CA) and included polyclonal antisera against FGFR1 (SC-121; 1:200), FGFR2 (SC-122; 1:200), FGFR3 (SC-123; 1:1,200), and FGFR4 (SC-7520; 1:200). Sections were incubated in PBST containing 4% normal donkey serum and 10% normal horse serum with the appropriate concentrations of antibodies for 4 d at 4 C. After washing in PBST, sections were incubated in biotinylated donkey antirabbit (for FGFRs 2 and 3) or donkey antigoat (for FGFRs 1 and 4) IgG (Jackson ImmunoResearch, West Grove, PA) for 1 h, washed in PBST, and incubated with Vectastain ABC reagent (Vector Laboratories, Burlingame, CA) for 1 h. Sections were subsequently washed and reacted with Alexa 488-labeled tyramide using the tyramide amplification system (TSA; Molecular Probes, Eugene, OR).
After the TSA reaction, E15.5 and P3 sections were washed extensively overnight, blocked for the presence of endogenous biotin (avidin/biotin blocking kit, Vector Laboratories), and processed for the detection of the GnRH decapeptide by incubation with a specific polyclonal antibody against GnRH (LR5, a gift from Dr. Robert Benoit, McGill University Health Center, Montréal, Canada) for 4 d at 4 C. After 4 d, sections were washed, incubated in biotinylated donkey antirabbit IgG for 1 h, washed, and reacted with Cy3-conjugated streptavidin (Jackson ImmunoResearch). Fluorescent signals were visualized with an IX70 inverted microscope (Olympus, Tokyo, Japan) equipped with rhodamine and fluorescein filter sets. Some sections were imaged by a SP2 laser-scanning confocal microscope (Leica, Wetzlar, Germany) at 120 nm thickness to confirm the colocalization of FGFRs in GnRH neurons. Because antisera against FGFRs 2 and 3 and GnRH were all generated in rabbits, we also performed tests to ensure there were no cross-reactivities between the first and second antigen detection. Dual ICC conducted in the above sequence with the omission of LR5 generated no Cy3 signals, indicating the lack of cross-reactivities with the FGFR detection (data not shown).
Embryonic nasal explant culture
Embryos at E10, E10.5, E11, and E11.5 were obtained from timed-pregnant females. These bracketed time points were chosen to ensure the full coverage of the fate specification period. To isolate nasal explants from embryos, a coronal cut was made approximately 0.5 mm from the tip of the nasal prominence using a scalpel blade. The nasal explants containing the bilateral nasal placodes were kept in Earle’s balance salt solution in ambient air on ice until plating. A fraction of nasal explants were immersion fixed in 4% paraformaldehyde to be used as initial or 0 d in vitro (DIV), controls. The remaining explants were each placed into a well in a 12-well plate containing 0.7 ml phenol red-free DME/F12 medium (InVitrogen, Carlsbad, CA) supplemented with 5 U/ml penicillin and 5 µg/ml streptomycin. Some E10.5 nasal explants were treated with 10 µM FGFR-1 antagonist, SU5402 (Calbiochem, San Diego, CA). In one series of experiments, E10.5 explants were treated with 10 µM bromodeoxyuridine (BrDU), a mitotic indicator, for the first 24 h to examine whether GnRH neurons emerged from dividing precursors. All nasal explants were incubated in a humidified atmosphere of 95% air/5%CO2 at 35 C for 1, 2, or 3 d and, except for the removal of BrDU from BrDU-pulsed cultures, did not receive any medium change. After each time point, nasal explants were fixed and processed for whole-mount GnRH ICC to quantify the number of GnRH neurons that had emerged (defined by the appearance of GnRH immunopositive neurons regardless of their locations) from the nasal placodes.
Whole-mount GnRH ICC of nasal explant cultures
After immersion-fixation in 4% paraformaldehyde for 15–30 min, nasal explants (including 0 DIV controls) were washed in PBST and incubated in a mixture of 83% methanol and 5% H2O2 for 1 h to quench the endogenous peroxidase activity. Explants were rehydrated in 70, 50, and 25% methanol and the nonspecific binding blocked by incubation in PBST containing 10% normal horse serum for 2 h. Explants were subsequently incubated with the anti-GnRH antibody LR5 (1:4000) in 4% normal donkey serum and 10% normal horse serum for 6–10 d. After extensive washing, tissues were incubated in a donkey antirabbit IgG conjugated with horseradish peroxidase (Jackson ImmunoResearch), washed, and the signal visualized by the addition of 3,3'-diaminobenzidine.
After the first color reaction, explants were dehydrated in increasing concentrations of ethanol, cleared in Histoclear (National Diagnostics, Atlanta, GA), and embedded in Paraplast (Oxford Labs, St. Louis, MO). Ten-micrometer sections were cut on a rotary microtome and mounted onto poly-L-lysine-coated slides. Sections from explants pulsed with BrDU were further processed for BrDU ICC using a monoclonal anti-BrDU antiserum (1:1000; Roche, Indianapolis, IN) after an antigen retrieval step (1.5 N HCl for 1 h at 37 C). After incubation in the primary antiserum at 4 C overnight, sections were washed, incubated in biotinylated sheep antimouse IgG (Jackson ImmunoResearch), washed, reacted with Vectastain ABC kit, and then processed for color reaction using Vector SG (Vector Laboratories) as a chromagen. Slides were coded to conceal their identities, and the number of neurons positive for GnRH (brown) and, in some cases, for BrDU (gray) were examined within each explant under a light microscope. The number of GnRH neurons was counted by a blind observer. Only GnRH immunopositive neurons with distinct nuclei and well-delineated cellular borders were scored, regardless of their locations in the explant.
Primary cultures of dissociated GnRH neurons
GnRH neurons from E15.5 and P3 brains were dissociated to investigate the role of FGF signaling in neurite extension. These two ages were chosen because they represent time points during (E15.5) and after (P3) the period of heightened axon targeting; thus, neurons from these two ages might exhibit differential abilities to extend neurites in culture. Brains from the entire litter of embryonic (E15.5) or postnatal (P3) GnRH-GFP mice were removed and placed in ice-chilled Earle’s balance salt solution supplemented with 5 U/ml penicillin, 5 µg/ml streptomycin, and 2% B27 (InVitrogen). The medial preoptic areas were isolated, minced with a scalpel blade, and enzymatically dissociated with the Papain dissociation system (Worthington, Lakewood, NJ) according to the manufacturer’s instruction. Dissociated cells were plated in 12-well culture plates coated with poly-D, L-lysine (Sigma, St. Louis, MO; 5 µg/ml) and mouse laminin (Sigma; 5 µg/ml) at a density of 5 x 105 cells per well. Cells were allowed to attach for 30 min before changing to a serum-free culture medium consisting of phenol red-free OptiMEM-I (InVitrogen) supplemented with insulin (10 µg/ml), transferrin (0.55 µg/ml), selenium (7 ng/ml), penicillin (5 U/ml), and streptomycin (5 µg/ml). In some wells, 10 µM SU5402 or various doses of recombinant human FGF2 (Promega Inc., Madison, WI) were added. Cells were maintained in a humidified atmosphere of 95% air/5% CO2 at 35 C.
At 1 DIV, culture wells were scanned under an inverted epifluorescence microscope. GnRH neurons were identified by the presence of bright GFP signals and photographed using a SPOT (Diagnostic Industries, Inc., Sterling Heights, MI) digital camera. Images were coded by a naive participant to conceal the identity of the treatments. Image brightness was enhanced with Photoshop software (Adobe Systems, Inc., San Jose, CA), the number of neurites recorded, and the length of each neurite in GnRH-GFP neurons analyzed using the NIH Image (6.0) software (National Institutes of Health, Bethesda, MD). For each treatment, 30–36 GnRH neurons, obtained from three separate litters, were measured.
Statistical analysis
Differences between groups were analyzed by Student’s t test or one-way ANOVA on square root-transformed data, followed by the Student-Newmann-Kuel’s post hoc test. When data failed the homogeneity tests, analysis was performed using the Mann-Whitney U test or the Kruskal-Wallis test followed by Dunn’s post hoc test (see Fig. 4A
). Differences were considered significant when P < 0.05.
|
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
FGFRs 1, 2, and 3 were detected in the medial region of the E10.5 nasal placode epithelium, in which GnRH neurons first emerge (Fig. 1), but substantial differences exist in the pattern of immunostaining among these three FGFRs. FGFR1 was localized predominantly in the nasal epithelium, although weaker immunoreactivity was also observed in the forebrain neuroectoderm (Fig. 1B). FGFR2 immunoreactivity was ubiquitously present in the nasal epithelium, mesenchyme, and forebrain neuroectoderm (Fig. 1C). FGFR3 was localized exclusively in the nasal epithelium (Fig. 1D), with virtually no immunoreactivity in the mesenchyme. FGFR4 immunoreactivity was not detectable in E10.5 nasal placode sections (data not shown). Overall, the presence of three FGFRs in the presumptive birthplace of GnRH neurons reflects the ability of FGF signaling to directly influence precursor cells that give rise to GnRH neurons.
|
). Only FGFRs 1 and 3 were clearly present in a fraction of E15.5 and P3 forebrain GnRH neurons. However, the cellular distribution of FGFR1 staining in GnRH neurons differed substantially from that of FGFR3. For FGFR1, intense patches of staining polarized to one side of the neuron were observed (Fig. 2, A and B). Interestingly, in the majority (
> 80%) of FGFR1-positive GnRH neurons, this polarized staining occurred on the side of the cell with a visible process. FGFR2 was not detected in E15.5 and P3 GnRH neurons (Fig. 2, C and D). FGFR3, like FGFR1, was localized in a fraction of E15.5 and P3 GnRH neurons (Fig. 2, E and F). However, FGFR3 was localized exclusively in the nuclei of GnRH neurons (Fig. 2, E and F), a phenomenon previously described in other cells types in a number of reports (see Discussion).
|
, G and H).
SU5402 inhibits the emergence of GnRH neurons in nasal explant cultures
A nasal explant culture was established to examine GnRH neuronal fate specification under defined conditions in the presence or absence of a FGFR antagonist, SU5402 (20). To ensure GnRH neuronal fate specification in this culture system mimics the in vivo event as previously described (1, 21), we first examined whether GnRH neurons could emerge from nasal explants during a developmental window comparable with that observed in vivo. Explants from four stages were tested: those isolated from embryos before (E10), at (E10.5 and E11), and after (E11.5) the optimal age of in vivo GnRH neuronal fate specification (21).
After 3 DIV, nasal explants only from E10.5 and E11.0 embryos showed significant increases in the number of GnRH neurons, compared with 0 DIV (d 0) explants (Fig. 3, A–C, E, and F). Virtually no GnRH neurons emerged from explants isolated from E10 embryos (Fig. 3A), and no further increase in GnRH neurons was found in E11.5 explants (Fig. 3, A, D, and G), which already contained a substantial number of GnRH neurons before culture. In E10.5 explants pulsed with BrDU, none of the GnRH neurons present after 3 DIV was positive for BrDU staining, indicating that these GnRH neurons emerged from postmitotic precursors (data not shown). We did not examine BrDU incorporation in E11 and E11.5 embryos because peak GnRH neurogenesis was previously reported to occur between E10 and E11 (21). In explants isolated from E10.5 and E11, GnRH neurons were present predominantly in the medial portion of the nasal epithelium after 3 DIV (Fig. 3, E and F), consistent with the location in which GnRH neurons were first detected in vivo. However, GnRH neurons were also found in the nasal mesenchyme and neuroectoderm of the cultured explants, indicating the occurrence of GnRH neuronal migration. Axons of the olfactory nerve were not normally observed projecting from the explants (data not shown) as previously described (22). The lack of axon growth could be due to the suspended nature of our culture system.
|
). GnRH neurons continued to emerge over the culture period, resulting in a significant increase in GnRH neuron number at 2 and 3 DIV, compared with 0 and 1 DIV (Fig. 4A; P < 0.05 between 0 and 2 DIV; P < 0.001 between 0 and 3 DIV). Because the increase was statistically more robust at 3 DIV, we further tested whether this increase could be suppressed by the addition of a FGFR antagonist, SU5402. SU5402 significantly suppressed the number of GnRH neurons after 3 DIV to the number observed in the 0 DIV controls (Fig. 4B).
FGF2 stimulates and SU5402 inhibits neurite outgrowth in primary cultures of dissociated GnRH-GFP neurons
To investigate whether FGF signaling regulates the extension of GnRH neuronal processes, we established primary cultures of dissociated GnRH neurons to study GnRH neurite outgrowth in the presence or absence of exogenous FGF2 and SU5402. For this purpose, we used the GnRH-GFP mice with the expression GFP targeted to GnRH neurons (18). The use of GnRH-GFP neurons allowed the imaging of live GnRH neurons in real time and reduced damage to neurites from washing and incubation during the lengthy ICC procedure.
Bright-field and epifluorescence images of a representative P3 GnRH-GFP neuron showed that the full extent of neurites was clearly visible by the GFP signal (Fig. 5, A and B). Thus, monitoring the extent of GFP provided an accurate representation of neurite length. GnRH neurons from two stages, E15.5 (axon targeting stage) and P3 (posttargeting stage), were cultured, and their neurite outgrowth in response to increasing concentrations of FGF2 examined. FGF2 was chosen because it is one of the more promiscuous FGF ligands shown to activate all FGFRs (reviewed in Ref.19) and its ability to stimulate neurite outgrowth in immortalized GnRH neuronal cell lines, GT1 cells, documented (15, 16). For both E15.5 and P3 GnRH-GFP neurons, 50 ng/ml FGF2 exerted a significant stimulatory effect, with the lower doses being ineffective (Fig. 5, C and D). Interestingly, neurites in E15.5 GnRH-GFP neurons (Fig. 5C) were approximately 2–3 times longer than P3 GnRH-GFP neurons (Fig. 5D) receiving the same treatment. A significant difference (P = 0.03 by Student’s t test) in neurite lengths was observed between E15.5 and P3 neurons in FGF (50 ng/ml)-treated cultures. When the number of GnRH-GFP neurites was analyzed as a measure of branching, only P3, not E15.5, GnRH-GFP neurons responded to FGF2 (50 ng/ml) with a significant increase (Fig. 5, E and F).
|
, A and B. In response to 50 ng/ml FGF2, neurite outgrowth was increased to 68.2 ± 11.5 µm (P < 0.05, compared with control), but this FGF2-induced increase was blocked by SU5402 (10 µM) to 22.9 ± 9.5 µm, a length that is not significantly different from control cultures (24.9 ± 6.5 µm), indicating the specificity of SU5402 (n = 19–20; Kruskal-Wallis test followed by Dunn’s test).
|
, A and B, and 6), with little neurite ramification. Neurites from GnRH-GFP neurons appeared to target nearby or distant unidentified cell types in the culture (Figs. 5, A, B, and 6, C and D). The addition of FGF2 visibly increased the length of neurites and sometimes resulted in a highly tortuous pathway of neurite growth (Fig. 6D). |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To demonstrate GnRH neurons possess the ability to respond directly to FGF signaling, we examined the presence of FGFR proteins in E15.5 and P3 GnRH neurons. The presence of FGFRs 1 and 3 in E15.5 and P3 GnRH neurons is highly consistent with a previous report that GT1 cells expressed only FGFRs 1 and 3 but not FGFR2 (15). Interestingly, a similar dual ICC experiment conducted previously on the juvenile rat (17) detected extremely weak FGFR1 staining in GnRH neurons that could not be interpreted as a positive signal by the authors (17). The conclusion was FGFR1, if present at all, was expressed at a level too low for the reliable detection by the existing techniques at that time. The use of an additional amplification step (TSA) in the present study circumvented the sensitivity problem and, for the first time, revealed the definitive presence of FGFRs 1 and 3 in endogenous GnRH neurons.
It is interesting to note that the cellular distribution of FGFRs 1 and 3 is not entirely consistent with a pattern known for the transmembrane tyrosine kinase receptors (Fig. 2). FGFR1 was detected in the cytoplasm of a fraction of GnRH neurons in a polarized fashion. FGFR3, on the other hand, was found exclusively in the nuclei of GnRH neurons. The significance of these atypical cellular distributions is unclear. However, a recent study demonstrated that FGFRs 1, 2, and 3, unlike most transmembrane receptors, possess transmembrane domains with short stretches of hydrophobic amino acids that are continuously interrupted by polar amino acids (23). These polar interruptions rendered FGFRs, especially FGFR1, susceptible to dissociation from the membranes of the endoplasmic reticulum. Subsequently, these receptors may be readily released into the cytoplasm and ultimately translocated into the nucleus (23). In this regard, extensive cytoplasmic and nuclear localization of FGFRs 1 and 3 have been observed in a variety of cell types (23, 24, 25, 26, 27, 28, 29). The functional significance of cytoplasmic and nuclear FGFRs is underscored by the recent discovery that these intracellular FGFRs may actively transduce signals from the intracellular FGF ligands, leading to significant physiological changes (30).
Our ICC results that the medial nasal placode of E10.5 embryos was positive for FGFRs 1, 2, and 3 is the first demonstration that FGFR proteins are present in the region in which GnRH neurons first emerge and thus could directly mediate the process of GnRH neuronal fate specification. These results are not surprising because patterning and morphogenesis of the nasal placode and the outgrowth of olfactory axons require FGF signaling (31). In further support of our data, in situ hybridization revealed that the neuroepithelium of the nasal placode expressed mRNAs for not only FGFRs 1 and 2 but also up to seven forms of FGFs (32, 33, 34, 35). Thus, the period during which GnRH neuronal fate is specified coincided with not only the presence of FGFRs but also the production of diverse FGF ligands required to activate FGFRs.
GnRH neurons were first detected in the mouse at E10.5 by the expression of GnRH mRNA and the production of the GnRH peptide (1, 21, 36). Currently there is no reliable way to identify GnRH precursors before the production of GnRH mRNA or peptide (4). Because there is little temporal difference between the transcription and translation of GnRH (21), we rely exclusively on the presence of the GnRH peptide to define the fate specification of a GnRH neuron. Specifically, this process refers to the acquisition of the ability of GnRH precursors to synthesize GnRH within their presumed endogenous birthplace. This is a terminology also employed previously (4). There is evidence in nonmammalian vertebrates that GnRH precursors might have a rigid developmental programming whose course cannot be altered and therefore could be considered committed rather than specified (37). However, because our study cannot test this possibility and all the GnRH precursors in our explant cultures were allowed to develop within what presumed to be their natural birthplace, we chose to use the term fate specification. Using a simple explant culture that supports the specification of GnRH neuronal fate, we observed robust increases in the number of GnRH neurons over 3 d in explants isolated from E10.5 and E11.0, but not E10.0 and E11.5, embryos. These two time points (E10.5 and E11.0) coincided with the period during which the majority of GnRH neurons emerged in vivo (1, 21, 36). Interestingly, this increase in our culture system was not due to neurogenesis per se because explants treated with BrDU for the first 24 h failed to incorporate this mitotic marker. More likely, these newborn GnRH neurons emerged from postmitotic GnRH precursors during the final stage of fate specification, a stage that endowed the precursors with the ability to produce GnRH. This observation suggests the exciting possibility that differentiated GnRH neurons, like other peripheral neurons originating in the nasal placode (reviewed in Refs.38 , 39), might be derived from a lineage of continually maturing postmitotic precursor cells. Thus, at least in our culture system, the specification of GnRH neurons did not appear to occur concomitantly with the termination of cell division, as implicated by the previous in vivo study (21).
To examine the contribution of FGF signaling to the initial emergence of GnRH neurons, nasal placode explants were treated with a specific FGFR antagonist (SU5402) to block the actions of endogenous FGFs. SU5402 has been shown to block the tyrosine kinase activity of FGFRs 1, 2, and 3 (20, 40, 41). The ability of SU5402 to significantly inhibit the emergence of GnRH neurons from nasal explants is indicative of the presence of high levels of endogenous FGFs, further supporting a critical role of FGF signaling during this period of GnRH neuronal development. SU5402 could inhibit the emergence of GnRH neurons via several possible mechanisms. First, FGF signaling could be truly required for the final stage of GnRH neuronal fate specification, thus blocking FGFR activity arrested the differentiation process and prevented the expression of GnRH. In this regard, SU5402 has been shown to block chick epiblast cells from acquiring a neuronal fate (42). A second possibility is that SU5402 might decrease the survival of GnRH precursors or newborn GnRH neurons, leading to an overall reduction in the total number of GnRH neurons in the culture. Lastly, SU5402 might exert nonspecific cytotoxic effects on the cultured explants. We do not believe the latter is likely because SU5402 treatment for 3 d did not reduce GnRH neuron number below that of the initial control (0 DIV), as would be expected if there was a general cytotoxic effect. Furthermore, SU5402 has been used previously at higher concentrations in nasal placode tissue without inducing cytotoxicity (34). Overall, these results suggest FGF signaling is critical in completing the final stage of GnRH neuronal fate specification and/or supporting the survival of GnRH precursors or newborn GnRH neurons.
One might hypothesize that SU5402 could inhibit the appearance of GnRH neurons by merely suppressing the production of GnRH peptide to levels below detection by ICC in an already differentiated neuron. Whereas this remains a possibility, evidence from GT1 cells, immortalized GnRH neuronal cell lines, argues against it. Wetsel et al. (43) showed that in GT1 cells, FGF2 treatment actually suppressed the level of GnRH prohormone and promoted the formation of C-terminally extended intermediate GnRH products instead of the mature GnRH peptide. According to these data, the inhibition of FGF signaling by SU5402 should elevate the level of the mature GnRH peptide, resulting in the detection of more GnRH neurons by our ICC. Our results that SU5402 inhibited GnRH neuronal emergence clearly failed to support this possibility.
The regulatory role of FGF signaling was not restricted to the narrow window of GnRH neuronal fate specification. FGF2 also stimulated the extension of neurites in primary cultures of dispersed GnRH-GFP neurons, an action indicative of its ability to promote intrinsic axon outgrowth required for axon targeting. That SU5402 inhibited neurite outgrowth in P3 GnRH-GFP neurons cultured under basal condition further argues for a physiological role of endogenously produced FGFs. Although E15.5 and P3 GnRH neurons showed similar responsiveness to FGF2, an interesting difference was observed, namely the mean length of the neurites in E15.5 GnRH-GFP neurons was approximately 2–3 times longer than postnatal GnRH-GFP. This difference could reflect a greater intrinsic ability of E15.5 GnRH neurons to retain and sprout neurites. For instance, embryonic GnRH-GFP neurons may be able to retain their cellular processes through enzymatic dissociation better than postnatal neurons, resulting in longer starting neurite length and thus longer neurites after 24 h of culture. Another possibility is that E15.5 GnRH-GFP neurons possess a greater intrinsic ability to extend neurites and underwent accelerated neurite outgrowth after plating. Regardless, the relative increases in neurite length in response to FGF2 treatment (2- to 3-fold) were similar for both the E15.5 and the P3 GnRH-GFP neurons. This observation suggests that this particular response to FGF2 is retained at a postnatal age when axon extension would seem less biologically relevant. However, one should note that the ability of GnRH axons to extend axons was observed well into adulthood. In fact, Prevot et al. (44) demonstrated GnRH axons could dynamically extend and retract during the rat estrous cycle, enabling direct contacts between the GnRH axon terminals and the external zone of the median eminence only during the proestrus. Similarly, the ability of transplanted postnatal GnRH neurons to extend axons to the median eminence (45) has been demonstrated. Therefore, in addition to promoting axon targeting during development, FGF signaling could be critical in guiding some of the plastic changes in the GnRH axon terminals after birth.
In the GnRH cell line, GT1, FGFs remained one of the most potent neurotrophic factors capable of promoting neurite outgrowth and cellular survival (15, 16, 17, 46). Although caveats have been raised regarding the transformed nature of GT1 cells, our results on the presence of FGFRs and neurite outgrowth in endogenous GnRH neurons are strikingly similar to those obtained from GT1 cells, which are believed to possess many properties of postmigratory embryonic GnRH neurons (47). Our current results also corroborated the previous two reports on the ability of FGF to promote axon outgrowth in endogenous GnRH neurons (17, 48). In one (48), it was proposed that FGFs might also serve as a potential chemoattractant during GnRH axon targeting, suggesting the complex nature in which FGF signaling might act to regulate a single stage of development. We acknowledge that our dissociated cell culture system does not closely mimic the in vivo targeting conditions because of the absence of a target. However, GnRH axon targeting requires both the intrinsic extension of axons and the directional guidance of these extending axons by a target. Gibson et al. (48) previously demonstrated the involvement of FGF signaling in the latter. Rather than repeating this observation, our study was designed to focus on FGF contribution to the first process in the absence of chemoattractants from a target.
We have not tested whether inhibitors of other growth factors altered the development of the GnRH system. Because the primary goal of this study was to test the role of FGF signaling, only the inhibitor of FGF signaling was tested. However, we acknowledge that other factors might also contribute to GnRH neuronal development, and this possibility should be an important avenue for exploration in the future. The potential involvement of other factors raises questions regarding the specificity of SU5402, i.e. whether SU5402 could oppose the actions of factors other than FGFs in our culture systems, thus confounding our data interpretation. Whereas we cannot rule out this possibility, we do not believe this is likely for two reasons. First, previous reports have shown that SU5402 has only very weak antagonistic effects on the receptor for platelet-derived growth factor and no effects on receptors for insulin, epidermal growth factor, nerve growth factor, ciliary neurotrophic factor, and glial cell line-derived neurotrophic factor (20, 49). The only other effects of SU5402 reported were on the receptors for vascular endothelial growth factor (20, 50). These reports demonstrated the substantial specificity of SU5402 for FGFRs, making SU5402 the most widely used and accepted inhibitor of FGF signaling (for examples, see Refs.51, 52, 53, 54). Second, in our neurite outgrowth studies on P3 GnRH-GFP neurons, SU5402 specifically blocked the effects of added FGF2 on neurite outgrowth but not to the level below the control. This result suggests that other factors, if present, are not inhibited by SU5402 or the contribution of other factors is minimal in this system.
In this study, we have demonstrated, through a series of in vitro experiments, a critical role of the FGF signaling in the development of GnRH neurons. In addition to enhancing neurite extension at the later phase of development, FGF signaling also appeared critical at the early phase of development when newborn GnRH neurons first emerged. The demonstration of FGFR proteins in GnRH neurons and the developing nasal placodes strongly support the ability of at least a fraction of these neurons to respond directly to FGFs. Our results and a report linking the loss of FGFR1 function to the autosomal dominant Kallmann syndrome (12) should provide a strong impetus for further exploring the involvement of FGF signaling in the development of the GnRH system.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Abbreviations: BrDU, Bromodeoxyuridine; DIV, day in vitro; E, embryonic day; FGF, fibroblast growth factor; FGFR, FGF receptor; GFP, green fluorescent protein; ICC, immunocytochemistry; P, postnatal day; PBST, PBS containing Triton X-100; TSA, tyramide amplification system.
Received February 18, 2004.
Accepted for publication April 20, 2004.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
- aminobutyric acid(A) receptor manipulation on migrating gonadotropin-releasing hormone neurons through the entire migratory route in vivo and in vitro. Endocrinology 141:1254–1262, and insulin-like growth factor-I on a hypothalamic gonadotropin-releasing hormone neuronal cell line. J Neurosci Res 49:739–749[CrossRef][Medline]
This article has been cited by other articles:
 |
|
 |
|
  W. C. J. Chung, S. S. Moyle, and P.-S. Tsai Fibroblast Growth Factor 8 Signaling through Fibroblast Growth Factor Receptor 1 Is Required for the Emergence of Gonadotropin-Releasing Hormone Neurons Endocrinology, October 1, 2008; 149(10): 4997 - 5003. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. C. Gill, B. Wadas, P. Chen, W. Portillo, A. Reyna, E. Jorgensen, S. Mani, G. A. Schwarting, S. M. Moenter, S. Tobet, et al. The Gonadotropin-Releasing Hormone (GnRH) Neuronal Population Is Normal in Size and Distribution in GnRH-Deficient and GnRH Receptor-Mutant Hypogonadal Mice Endocrinology, September 1, 2008; 149(9): 4596 - 4604. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Xu, Y. Qin, R. H. Reindollar, S. P. T. Tho, P. G. McDonough, and L. C. Layman A Mutation in the Fibroblast Growth Factor Receptor 1 Gene Causes Fully Penetrant Normosmic Isolated Hypogonadotropic Hypogonadism J. Clin. Endocrinol. Metab., March 1, 2007; 92(3): 1155 - 1158. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Giacobini, A. Messina, S. Wray, C. Giampietro, T. Crepaldi, P. Carmeliet, and A. Fasolo Hepatocyte Growth Factor Acts as a Motogen and Guidance Signal for Gonadotropin Hormone-Releasing Hormone-1 Neuronal Migration J. Neurosci., January 10, 2007; 27(2): 431 - 445. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Pitteloud, J. S. Acierno Jr., A. Meysing, A. V. Eliseenkova, J. Ma, O. A. Ibrahimi, D. L. Metzger, F. J. Hayes, A. A. Dwyer, V. A. Hughes, et al. Mutations in fibroblast growth factor receptor 1 cause both Kallmann syndrome and normosmic idiopathic hypogonadotropic hypogonadism PNAS, April 18, 2006; 103(16): 6281 - 6286. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. R. Ojeda, A. Lomniczi, C. Mastronardi, S. Heger, C. Roth, A.-S. Parent, V. Matagne, and A. E. Mungenast Minireview: The Neuroendocrine Regulation of Puberty: Is the Time Ripe for a Systems Biology Approach? Endocrinology, March 1, 2006; 147(3): 1166 - 1174. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. A. Tobet and G. A. Schwarting Minireview: Recent Progress in Gonadotropin-Releasing Hormone Neuronal Migration Endocrinology, March 1, 2006; 147(3): 1159 - 1165. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. C. Gill and P.-S. Tsai Expression of a Dominant Negative FGF Receptor in Developing GNRH1 Neurons Disrupts Axon Outgrowth and Targeting to the Median Eminence Biol Reprod, March 1, 2006; 74(3): 463 - 472. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. A. Christian, J. L. Mobley, and S. M. Moenter Diurnal and estradiol-dependent changes in gonadotropin-releasing hormone neuron firing activity PNAS, October 25, 2005; 102(43): 15682 - 15687. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Pitteloud, J. S. Acierno Jr., A. U. Meysing, A. A. Dwyer, F. J. Hayes, and W. F. Crowley Jr. Reversible Kallmann Syndrome, Delayed Puberty, and Isolated Anosmia Occurring in a Single Family with a Mutation in the Fibroblast Growth Factor Receptor 1 Gene J. Clin. Endocrinol. Metab., March 1, 2005; 90(3): 1317 - 1322. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. P. Bless, H. J. Walker, K. W. Yu, J. G. Knoll, S. M. Moenter, G. A. Schwarting, and S. A. Tobet Live View of Gonadotropin-Releasing Hormone Containing Neuron Migration Endocrinology, January 1, 2005; 146(1): 463 - 468. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Gonzalez-Martinez, S.-H. Kim, Y. Hu, S. Guimond, J. Schofield, P. Winyard, G. B. Vannelli, J. Turnbull, and P.-M. Bouloux Anosmin-1 Modulates Fibroblast Growth Factor Receptor 1 Signaling in Human Gonadotropin-Releasing Hormone Olfactory Neuroblasts through a Heparan Sulfate-Dependent Mechanism J. Neurosci., November 17, 2004; 24(46): 10384 - 10392. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
