| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Integrative Physiology and Center for Neuroscience, University of Colorado, Boulder, Colorado 80309-0354
Address all correspondence and requests for reprints to: Dr. Wilson C. J. Chung, Department of Integrative Physiology and Center for Neuroscience, University of Colorado, Boulder, Colorado 80309-0354. E-mail: wilson.chung{at}colorado.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Aberrant embryonic development of the GnRH system results in severe clinical consequences. For example, abnormal GnRH system development is a hallmark of Kallmann syndrome (KS), a human neurological disorder characterized by hypogonadotropic hypogonadism associated with anosmia. Loss-of-function gene mutations have been shown to cause KS. These include mutations on anosmin-1, fibroblast growth factor receptor 1 (Fgfr1), prokineticin receptor-2, and prokineticin-2 genes (6, 7, 8, 9).
Loss-of-function Fgfr1 mutations in KS is of specific interest because the targeted expression of a dominant-negative fibroblast growth factor receptor (FGFR) in mouse GnRH neurons resulted in a 30% decrease in the total number of GnRH neurons in the forebrain, late pubertal onset, and premature reproductive senescence (10). Nonetheless, several important mechanistic questions regarding the role of fibroblast growth factor (FGF) signaling during GnRH system development remain unanswered. First, developing GnRH neurons express both FGFR1 and FGFR3 (11). This and the existence of 22 FGF ligands (12) make it difficult to pinpoint the precise FGF ligand/receptor combination(s) mediating the development of the mouse GnRH system. Second, it is unclear which stage of GnRH system development is most severely disrupted by FGF signaling deficiency.
Recently loss-of-function mutations in the Fgf8 gene were found in KS patients (13), suggesting that FGF8 may be a crucial ligand for GnRH system development. Binding studies showed that FGF8 can readily bind to FGFR1 and FGFR3 with similar affinities, and the activation of both receptors induces mitogenic activity (14). These findings underscore the need to identify which specific FGFR mediates FGF8 signaling.
In this study, we investigated whether FGF8 plays a critical role in the development of the GnRH system and, if so, during which specific stage. We also examined whether FGFR1 or FGFR3 is important for the development of the GnRH system in mice. To elucidate these questions, we studied the mouse GnRH system development in transgenic mice that express significantly lower levels of functional FGF8 protein or FGFR1 protein or in mice in which FGFR3 expression is completely abolished. Our data suggest that FGF8 actions, primarily mediated through FGFR1, are critically required for the very early phase of GnRH neuronal development.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Tissue collection
Embryos retrieved on E11.5 and E14.5 were immersion-fixed in 4% paraformaldehyde in 0.1 M phosphate buffer at room temperature for 2 h. Similarly, the heads of PN 0 pups were immersion fixed for 6 h at room temperature. Nonfixed tissue samples from each individual mouse embryo or pup were taken for genotyping purposes. Embryos and PN 0 tissues were stored in either 100% methanol at –20 C or 25% sucrose in phosphate buffer at 4 C before immunocytochemistry (ICC) (11) or in situ hybridization (ISH).
GnRH ICC
GnRH neurons in embryonic and neonatal heads were detected using the rabbit anti-GnRH polyclonal antibody LR5 (a gift from Dr. R. Benoit, Montréal General Hospital, Montréal, Québec, Canada) as described in detail previously (11). Briefly, whole E11.5 embryos or serial parasagittal sections of E14.5 embryos (20 µm) were incubated with LR-5 (1:5000) diluted in 0.1 M phosphate-buffered saline and 0.4% Triton X-100 containing 4% normal donkey serum and 10% normal horse serum for 5 d. Serial coronal sections (20 µm) from PN 0 animals were incubated with LR-5 (1:10,000) for 4 d at 4 C. Then E11.5 whole embryos were incubated in horseradish peroxidase-conjugated donkey antirabbit (Jackson ImmunoResearch Laboratories, West Grove, PA) for 2 d, reacted with 0.05% 3, 3-diaminobenzidine, and embedded in paraffin. Serial ribbons of parasagittal sections (12 µm) were made with a rotary microtome (Leica, Bannockburn, IL), mounted on gelatin-coated glass slides and deparaffinized. Sections from E14.5 embryos and PN 0 pups were incubated with a biotinylated-donkey antirabbit antibody (1:500), followed by avidin and biotin-coupled horseradish peroxidase (ABC kit; Vector Labs, Burlingame, CA), and reacted with DAB. All sections were dehydrated and coverslipped with Permount (Fisher, Pittsburgh, PA).
Nonradioactive GnRH ISH
Coronal sections (20 µm) through the PN 0 preoptic region of WT and HOMO Fgf8 hypomorphs were processed for ISH using a standard nonradioactive protocol (10). Sections were hybridized with a 270-nt digoxigenin-label riboprobe, which encompasses the entire open reading frame of the GnRH cDNA, from the start to the stop codon (NM_008145), washed extensively, and incubated with an antidigoxigenin antibody conjugated to alkaline phosphatase (Roche, Indianapolis, IN). Following, GnRH mRNA was detected using 5-bromo-4-chloro-3-indoyl-phosphate, 4-toluidine salt/4-nitro blue tetrazolium chloride as chromagens. The distribution of GnRH neurons detected by this probe is identical with that detected by the specific GnRH antibody LR5.
Olf-1 and peripherin ICC
ICC for Olf-1 [a marker for olfactory receptor neurons (ORN)] and peripherin (a marker for intermediate filaments in olfactovomeronasal fibers) were conducted to assess whether Fgf8 hypomorphs suffer disruption in ORN development and the GnRH migratory pathway. Serial parasagittal sections (20 µm) were mounted on gelatin-coated glass slides and incubated with a rabbit anti-Olf-1 polyclonal antibody (for E11.5 embryos, 1:2000, a gift from Dr. R. R. Reed, Johns Hopkins University, School of Medicine, Baltimore, MD) or a rabbit antiperipherin polyclonal antibody (for E14.5 embryos, 1:3000, AB1530. Millipore, Bedford, MA) for 4 d at 4 C, incubated with a Cy3-conjugated donkey antirabbit antibody (1:500), and coverslipped with an antifade medium containing the nuclear stain, 4',6'-diamino-2-phenylindole (5 µM).
Apoptosis
To compare the incidence of apoptosis in the nasal region of WT, HET, and HOMO Fgf8 hypomorphic E11.5 embryos, we stained corresponding parasagittal sections with the histological stain cresyl violet, which reveals the pyknotic cell nuclei of apoptotic cells (18). Comparisons were made in the Olf-1 positive region of the olfactory epithelium.
GnRH neuronal counts
The total number of GnRH neurons was counted in coded serial sections (20 µm) through the whole neonatal head, from the tip of nose to past the median eminence.
Statistical analysis
One-way ANOVA or Student’s t tests were used to test for significant differences (P < 0.05). Student-Newman-Keuls test was used for post hoc testing.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and 3, A and B). The medial-ventral olfactory epithelium in HOMO Fgf8 hypomorphs also showed invagination, albeit to an extent much less than WT and HET Fgf8 hypomorphs (Figs. 1B and 3C).
|
|
, A and B). Similarly, GnRH-IR neurons were found in the preoptic region of WT but not in HOMO Fgf8 hypomorphs on E14.5 (Fig. 1, C and D).
Consistent with the absence of GnRH neurons in HOMO Fgf8 hypomorphic embryos, no GnRH-IR neurons were detected in HOMO Fgf8 hypomorphs on PN 0 (Fig. 2A). The total number of GnRH-IR neurons in the whole head on PN 0 was significantly higher in WT (n = 4; 773 ± 64) than HET (n = 4; 268 ± 63) or HOMO (n = 4; none detected) Fgf8 hypomorphs on PN 0 (P < 0.01). Moreover, HET Fgf8 hypomorphs had less than 50% of the total number of GnRH-IR neurons, compared with WT Fgf8 hypomorphs (Fig. 2A: P < 0.01).
|
, B and C).
Olf-1 neurons and peripherin fibers in Fgf8 hypomorphs
On E11.5, Olf-1-IR sensory neurons were detected in the olfactory epithelium of WT (n = 3), HET (n = 3) and HOMO (n = 3) Fgf8 hypomorphs (Fig. 3, A–C). Specifically, Olf-1-IR cells were found in the dorsal and medial-ventral invagination of olfactory epithelium and doral of WT and HET Fgf8 hypomorphs (Fig. 3, A and B). Although this invagination is greatly reduced in HOMO Fgf8 hypomorphs, Olf-1-IR ORNs remained present in the corresponding medial-ventral olfactory epithelium (Fig. 3C).
Peripherin-IR fibers were detected in the nasal region of WT (n = 3) and HOMO (n = 2) E14.5 Fgf8 hypomorphs. Peripherin-IR fibers from the nasal region target and infiltrate the olfactory bulb in WT Fgf8 hypomorphic embryos (Fig. 3, D, E, F, and H). However, peripherin-IR fibers in HOMO Fgf8 hypomorphic embryos did not enter the brain or extend past the cribriform plate (Fig. 3, G and I).
Apoptosis in Fgf8 hypomorphs
Consistent with previous studies (19, 20, 21), we detected apoptotic cell bodies (i.e. pyknotic cell bodies) in the E11.5 olfactory epithelium (Fig. 4A) of the nasal region. However, the number of apoptotic cell bodies in the Olf-1 positive region of the medial-ventral olfactory epithelium, as visualized by cresyl violet, was very low (less than three per section). There were no gross differences in the incidence of apoptosis among WT (n = 3), HET (n = 3), and HOMO (n = 3) Fgf8 hypomorphs (Fig. 4, B–D).
|
, A, B, and E; P < 0.01). Interestingly, three of the four HOMO animals had only four to eight GnRH-IR neurons per head, whereas one HOMO animal had 365 GnRH neurons, indicating incomplete penetrance of this phenotype. There was no significant difference between the total number of GnRH-IR neurons in WT and HET Fgfr1 hypomorphs (Fig. 5E). GnRH ISH showed the presence of GnRH mRNA containing preoptic area neurons in WT but not HOMO Fgfr1 hypomorphs (Fig. 5, C and D).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Fgf8 hypomorphy completely eliminated the presence of GnRH neurons, whereas Fgfr1 hypomorphy severely reduced the total number of GnRH neurons. This discrepancy may be due to the variability in the knockdown of Fgfr1 mRNA in Fgfr1 hypomorphs, which ranges between 66 and 80%. Alternatively, other FGFRs may partially compensate for the decreased levels of FGFR1. Indeed, FGF8 signaling can also be facilitated by FGFR3. Nonetheless, our data indicate that FGF8 signaling is mainly facilitated through FGFR1. This does not rule out a role for other FGF signaling components.
Neuronal migration is an important aspect of GnRH system development. In humans, defective GnRH neuronal migration has commonly been accepted as a cause of hypogonadotropic hypogonadism in KS (22). Previous data showed that migrating GnRH neurons closely follow the scaffold of peripherin-IR olfactovomeronasal fibers to reach the forebrain (23). The abnormal organization of peripherin-IR fibers in the HOMO Fgf8 hypomorphs is consistent with an earlier report (24). Peripherin-IR fibers do not enter the brain of HOMO Fgf8 hypomorphs, whereas an extensive plexus of peripherin-IR fibers was detected in the olfactory bulbs and anterior forebrain of WT Fgf8 hypomorphs. Based on these observations, one might conclude that migratory defects underlie the absence of GnRH neurons in HOMO Fgf8 hypomorphs. However, GnRH neurons were already absent on E11.5. In other words, no GnRH neurons were present to enter migration. Therefore, we hypothesize that defects in early GnRH development, not migration, caused the absence of GnRH neurons in HOMO Fgf8 hypomorphs.
Previous studies showed that HOMO Fgf8 hypomorphs exhibited malformation of the nasal region (15). Therefore, the absence of GnRH-IR neurons from E11.5 onward may be the result of a general FGF-dependent dysgenesis of the nasal region. Although the homozygous state of Fgf8 hypomorphy eliminated all GnRH neurons, the presence of the olfactory epithelium, as indicated by the presence of Olf-1 positive neurons, persisted. Furthermore, the heterozygous state of Fgf8 hypomorphy, which obliterated more than 50% of all GnRH neurons, continued to have normal morphogenesis of the nasal region and olfactory bulbs (19). Together these data strongly suggest that deficits in FGF8 function affect some olfactory epithelium-derived cell populations (i.e. GnRH neurons) more than others. Of interest, we showed that a ventral recess of the olfactory epithelium from which GnRH neurons emerge is greatly reduced in HOMO Fgf8 hypomorphs on E11.5. This recess is highly FGF8 sensitive, and its malformation is likely the cause for the failure of GnRH neurons to emerge. This recess has also been shown to give rise to the vomeronasal organ (VNO) (25). Our preliminary data showing that the VNO is missing in the HOMO Fgf8 hypomorphs (data not shown) are consistent with the notion that this recess fails to develop and give rise to differentiated derivatives. Our data also suggest the presence of discrete developmental zones in the olfactory placode (26), with FGF8 signaling deficiency more likely to impact the ventral zone that gives rise to the VNO and GnRH neurons.
FGF8 is a potent cell survival factor. Indeed, the elimination of FGF8 signaling in Fgf8 conditional KOs resulted in abnormally high levels of apoptosis in the olfactory epithelium between E10.5 and E12.5 (19). On E11.5, we showed that the incidence of apoptosis in the olfactory epithelium was comparable among WT, HET, and HOMO Fgf8 hypomorphs. This discrepancy may be due to the less severe FGF8 deficiency in hypomorphs, compared with the conditional KOs (19). Alternatively, the difference in apoptotic incidence may be due to the nonlinear dosage effect of Fgf8. Previous data showed that the complete elimination or overexpression of FGF8 protein causes higher levels of apoptosis, whereas a decrease in functional FGF8 protein expression results in cell survival (27). This U-shaped effect could cause the incidence of apoptosis to differ dramatically between these two transgenic mice. Lastly, our data do not completely rule out the possibility that Fgf8 hypomorphy can cause increased levels of apoptosis in the early or presumptive olfactory placode, in which Fgf8 mRNA expression was first detected from E8.5 onward (28). Therefore, an intriguing consequence of Fgf8 hypomorphy may be the premature elimination of progenitor cells that are fated to become GnRH neurons before E11.5. Alternatively, Fgf8 hypomorphy may have hampered progenitor cells from becoming GnRH neurons. Unfortunately, the lack of a specific marker for GnRH progenitor cells prevents further studies along this line.
Although both Fgfr1 and Fgfr3 are expressed in developing GnRH neurons (11) and bind FGF8 equally well (14), the total number of GnRH neurons decreased in only HOMO Fgfr1 hypomorphs but not HOMO Fgfr3 KO mice. These data strongly suggest FGF8’s effects on the emergence of GnRH neurons from the olfactory placode are mediated through FGFR1. FGF8 signaling may be direct on GnRH neurons because GnRH neurons express FGFR1 during embryonic development (11).
The targeting of olfactovomeronasal fibers has been hypothesized to be required for the morphogenesis of the olfactory bulbs (29). In Fgf8 hypomorphs, the inability of peripherin-IR fibers to enter and contact the most anterior region of the brain may have caused the failure of olfactory bulb morphogenesis (15). Currently the mechanisms underlying the inability of peripherin-IR fibers to enter the embryonic HOMO Fgf8 hypomorphic brain are unknown. The elongation of olfactory fibers has been shown to be accompanied by migrating olfactory ensheathing cells, which are glia that originate from the olfactory epithelium (30, 31). These olfactory ensheathing cells express high levels of FGFR1 and have been shown to be permissive for the elongation/growth of olfactory axons (32, 33). Together with our results, we hypothesize that the abnormal organization of the peripherin-IR fibers may, in part, be due to decreased FGF8 signaling on FGFR1-expressing olfactory ensheathing cells, possibly impairing their functions, and hence disrupting the growth of peripherin-IR fibers. Alternatively, FGF8 deficiency may inhibit the emergence or number of olfactory ensheathing cells from the olfactory epithelium, resulting in the observed disorganization of peripherin-IR fibers.
Although genetic analysis showed that KS is causally linked to loss-of-function mutations in Fgfr1, and more recently in Fgf8 (13), the cellular effects of these mutations on human GnRH system development are unknown. Interestingly, these loss-of-function mutations are heterozygous in nature. Therefore, it has been recently proposed that the severity and pathogenesis of KS may depend on the compounding effects of heterozygous loss-of-function mutations on multiple causative genes (34). However, our experimental results clearly indicate that the heterozygous state of Fgf8 hypomorphy alone is sufficient to cause significant disruptions GnRH neuronal development. Furthermore, our studies suggest that Fgfr1 and Fgf8 mutations in KS may compromise human GnRH system function by abrogating the initial emergence of GnRH neurons from the olfactory epithelium. Finally, the abnormal organization of peripherin-IR olfactory fibers may explain the occurrence of anosmia in KS.
|
Footnotes |
|---|
Disclosure Statement: The authors have nothing to disclose.
First Published Online June 19, 2008
Abbreviations: E, Embryonic day; FGF, fibroblast growth factor; FGFR, FGF receptor; HET, heterozygous; HOMO, homozygous; ICC, immunocytochemistry; IR, immunoreactive; ISH, in situ hybridization; KO, knockout; KS, Kallmann syndrome; ORN, olfactory receptor neuron; PN, postnatal day; VNO, vomeronasal organ; WT, wild type.
Received November 28, 2007.
Accepted for publication June 9, 2008.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  T. Raivio, Y. Sidis, L. Plummer, H. Chen, J. Ma, A. Mukherjee, E. Jacobson-Dickman, R. Quinton, G. Van Vliet, H. Lavoie, et al. Impaired Fibroblast Growth Factor Receptor 1 Signaling as a Cause of Normosmic Idiopathic Hypogonadotropic Hypogonadism J. Clin. Endocrinol. Metab., November 1, 2009; 94(11): 4380 - 4390. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Hu, S. E. Guimond, P. Travers, S. Cadman, E. Hohenester, J. E. Turnbull, S.-H. Kim, and P.-M. Bouloux Novel Mechanisms of Fibroblast Growth Factor Receptor 1 Regulation by Extracellular Matrix Protein Anosmin-1 J. Biol. Chem., October 23, 2009; 284(43): 29905 - 29920. [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 |
