Differential requirements for FGF3, FGF8 and FGF10 during inner ear development

Abstract

FGF signaling is required during multiple stages of inner ear development in many different vertebrates, where it is involved in induction of the otic placode, in formation and morphogenesis of the otic vesicle as well as for cellular differentiation within the sensory epithelia. In this study we have looked to define the redundant and conserved roles of FGF3, FGF8 and FGF10 during the development of the murine and avian inner ear. In the mouse, hindbrain-derived FGF10 ectopically induces FGF8 and rescues otic vesicle formation in Fgf3 and Fgf10 homozygous double mutants. Conditional inactivation of Fgf8 after induction of the placode does not interfere with otic vesicle formation and morphogenesis but affects cellular differentiation in the inner ear. In contrast, inactivation of Fgf8 during induction of the placode in a homozygous Fgf3 null background leads to a reduced size otic vesicle or the complete absence of otic tissue. This latter phenotype is more severe than the one observed in mutants carrying null mutations for both Fgf3 and Fgf10 that develop microvesicles. However, FGF3 and FGF10 are redundantly required for morphogenesis of the otic vesicle and the formation of semicircular ducts. In the chicken embryo, misexpression of Fgf3 in the hindbrain induces ectopic otic vesicles in vivo. On the other hand, Fgf3 expression in the hindbrain or pharyngeal endoderm is required for formation of the otic vesicle from the otic placode. Together these results provide important insights into how the spatial and temporal expression of various FGFs controls different steps of inner ear formation during vertebrate development.

Genetic rescue of Muenke syndrome model hearing loss reveals prolonged FGF-dependent plasticity in cochlear supporting cell fates

“…Significantly, the hearing loss and supporting cell fate transformation are rescued when one copy of Fgf8 is removed from the Spry2-null background (Shim et al 2005). These data, together with the loss-of-function studies Hayashi et al 2007;Jacques et al 2007;Puligilla et al 2007;Zelarayan et al 2007), support the model of an FGF signaling gradient in the developing cochlear sensory epithelium in which support cell progenitors experiencing the highest levels of FGF8/ FGFR3 signaling differentiate as pillar cells, and those that are farther from the FGF source differentiate as Deiters' cells (Shim et al 2005).…”

“…Normally, FGF8 promotes pillar cell differentiation by activating FGFR3 (Hayashi et al 2007;Jacques et al 2007;Puligilla et al 2007;Zelarayan et al 2007), so increasing FGF signaling by removing Spry2, the negative feedback regulator that is initially expressed broadly in cochlear support cells but eventually becomes restricted to Deiters' cells, causes one Deiters' cell to transform into a pillar cell by P7. This transformation is rescued by removing one copy of Fgf8 (Shim et al 2005).…”

“…Thus, mice lacking Fgfr3 or humans carrying a presumed dominantnegative mutation in FGFR3 are hearing impaired, and the mice show incomplete differentiation of pillar cells Toydemir et al 2006;Hayashi et al 2007;Puligilla et al 2007). Similarly, mice lacking inner ear expression of Fgf8 also show aberrant pillar cell differentiation (Jacques et al 2007;Zelarayan et al 2007). FGFR3c is presumed to be the relevant isoform for pillar cell specification because although Fgfr3c-null mice have not been tested for hearing loss, they show other phenotypes associated with complete loss of Fgfr3, whereas Fgfr3b-null mice have no reported abnormalities (Eswarakumar and Schlessinger 2007).…”

Differential requirements for Fgf3 and Fgf8 during mouse forebrain development

“…To obtain mouse mutants that lacked Fgf8 expression in the telencephalon (Fgf8 TelKO ), we crossed animals carrying a Fgf8 null allele and the Foxg1-Cre transgene (Fgf8 d2,3/ϩ ;Foxg1 Cre/ϩ ) with animals carrying a conditional Fgf8 allele (Fgf8 flox/flox ) to obtain Fgf8 flox/d2,3 ; Foxg1 Cre/ϩ (Fgf8 TelKO ) mutants as previously described (Storm et al, 2003). To obtain Fgf3 Ϫ/Ϫ /Fgf8 flox/d2,3 ; Foxg1 Cre/ϩ (Fgf3 Ϫ/Ϫ /Fgf8 TelKO ) mutants we crossed Fgf3 ϩ/Ϫ /Fgf8 flox/d2,3 ;Foxg1 Cre/ϩ animals with Fgf3 Ϫ/Ϫ /Fgf8 flox/flox mice, as described previously (Zelarayan et al, 2007).…”

A novel method for retinoic acid administration reveals differential and dose‐dependent downregulation of Fgf3 in the developing inner ear and anterior CNS

“…Invagination of this placode to form the otocyst and the activity of numerous growth factor signaling pathways transforms this simple hollow epithelial sphere into a highly sculptured organ containing sensory hair cells and their sensory neurons organized in an array of different end organs. Mutational analyses of fibroblast growth factors FGF3 and 10 have shown these growth factors play critical co-operative roles in inner ear induction in the mouse (Alvarez et al, 2003;Wright and Mansour, 2003;Zelarayan et al, 2007) as well as a role for Fgf3 in development of the vestibulo-acoustic ganglion (Mansour et al, 1993). Consistent with this latter phenotype, Fgf3 is expressed in nascent otic neuroblasts, the precursor of the sensory nerves innervating the hair cells in the ear (Powles et al, 2004) that migrate anteriorly from the antero-ventral otocyst to generate the vestibulo-acoustic ganglion.…”