|Home | About | Journals | Submit | Contact Us | Français|
The development of the inner ear involves complex processes of morphological changes, patterning and cell fate specification that are under strict molecular control. SOX2 and SOX9 are SOX family transcription factors that are involved in the regulation of one or more of these processes. Previous findings have shown early expression of SOX9 in the otic placode and vesicle at E8.5–E9.5. Here we describe in detail, the expression pattern of SOX9 in the developing mouse inner ear beyond the otocyst stage and compare it with that of SOX2 from E9.5 to E18.5 using double fluorescence immunohistochemistry. We found that SOX9 was widely expressed in the otic epithelium, periotic mesenchyme and cartilaginous otic capsule. SOX2 persistently marked the prosensory and sensory epithelia. During the development of the sensory epithelia, SOX2 was initially expressed in all prosensory regions and later in both the supporting and hair cells up to E15.5, when its expression in hair cells gradually diminished. SOX9 expression overlapped with that of SOX2 in the prosensory and sensory region until E14.5 when its expression was restricted to supporting cells. This initial overlap but subsequent differential expression of SOX2 and SOX9 in the sensory epithelia, suggest that SOX2 and SOX9 may have distinct roles in molecular pathways that direct cells towards different cell fates.
The mammalian inner ear is an intricate organ responsible for the perception of sound and balance. The mouse inner ear arises from a thickening of the surface ectoderm called otic placode located adjacent to rhombomeres 5 and 6 of the hindbrain (reviewed in Barald and Kelley, 2004). At E9.0, the otic placode invaginates to form the otic vesicle. Neuroblasts delaminate from the ventral thickening of the otic vesicle and form the otic ganglion which will become the sensory innervation of the inner ear (Rubel and Fritzsch, 2002). The otic vesicle also undergoes a series of morphological changes until it reaches its mature shape by E17 (Morsli et al., 1998).
The inner ear consists of six sensory organs: the three cristae in the semi-circular canals and the maculae in the utricle and saccule are responsible for vestibular function; the organ of Corti is responsible for auditory function. The sensory patches in these organs consist of hair and supporting cells. The development of sensory patches in the inner ear requires complex processes of prosensory cell specification and cell fate determination (reviewed in Fritzsch et al., 2006; Kelley, 2007), in which SOX2 and SOX9 are likely to be involved (see below).
SOX2 and SOX9 are SOX family transcription factors characterized by a high mobility group (HMG) DNA-binding domain. Mutations in human SOX2 results in anophthalmia, a severe eye malformation, and some patients showed sensorineural hearing loss (Fantes et al., 2003; Hagstrom et al., 2005). SOX2 interacts with EYA1 for prosensory specification (Zou et al., 2008). Mouse mutants with no and reduced expression of SOX2 in the developing inner ear failed to establish a prosensory domain and formed an abnormal sensory epithelium with disorganized and fewer hair cells, respectively (Kiernan et al., 2005). The expression pattern of Sox2/SOX2 in the inner ear has been described in chick (Uchikawa et al., 1999; Neves et al., 2007) and mouse (Wood and Episkopou, 1999; Kiernan et al., 2005; Hume et al., 2007). In chick, SOX2 expression is restricted to the supporting cells during differentiation of the sensory epithelia (Neves et al., 2007). This is in contrast to that in the mouse in which SOX2 is expressed in both hair and supporting cells of all the sensory epithelia until early neonatal stage (P2) when its expression in hair cells is lost (Kiernan et al., 2005; Hume et al., 2007; Dabdoub et al., 2008).
Mutations in human SOX9 result in skeletal abnormality and XY sex reversal syndrome, campomelic dysplasia (CD), which has a high lethality rate. Sensorineural hearing loss has been reported in the surviving CD patients, suggesting a role of SOX9 in inner ear development (Savarirayan et al., 2003). Various studies in Xenopus (Saint-Germain et al., 2004; Taylor and LaBonne, 2005), zebrafish (Yan et al., 2005) and mouse (Barrionuevo et al., 2008) have described the expression of SOX9 at the otic placode and/or vesicle stage and they have focused on uncovering the early role of SOX9 in inner ear development. In Xenopus and zebrafish, the otic placodes and vesicles were absent upon SOX9 depletion (Saint-Germain et al., 2004; Yan et al., 2005). In addition, over-expression of SOX9 in Xenopus results in enlarged or ectopic otic vesicles (Taylor and LaBonne, 2005). In mouse, SOX9 was found to be important for otic placode invagination (Barrionuevo et al., 2008). However, the dynamics of SOX9 expression in the different cell types beyond early stages (>E9.5) of inner ear development has not been described.
Both SOX2 (Bylund et al., 2003; Graham et al., 2003; Ferri et al., 2004; Okubo et al., 2006; Taranova et al., 2006; Que et al., 2007; Holmberg et al., 2008) and SOX9 (Huang et al., 1999; Bi et al., 2001; Sahar et al., 2005; Seymour et al., 2008) operate in a dose dependent manner in cell fate specification. The expression of SOX2 and SOX9 has been previously studied in relation to each other in the mouse developing retina, adult pituitary gland and brain (Sottile et al., 2006; Fauquier et al., 2008; Poche et al., 2008). These studies have suggested cell populations expressing either SOX2 or SOX9, or both together may represent different populations of pluripotent progenitors or cells undergoing different cell fate decisions. To facilitate the interpretation of its role in inner ear development, we present for the first time a comprehensive description of the temporal and spatial expression profile of SOX9 in the developing inner ear beyond the otocyst stage in relation to the expression of SOX2.
In E9.5 otocyst, SOX2 was expressed in the proneural region in which neuroblast will delaminate to form the statoacoustic ganglion (Fig. 1A). It was also expressed in the adjacent hindbrain. SOX9 was faintly expressed in the periotic mesenchyme and prominently in the entire otic epithelium (Fig. 1B).
The expression signal of SOX9 was stronger on the medial side while that of SOX2 was stronger on the ventral side of E9.5 otic vesicle (Fig. 1). This variation in SOX2 and SOX9 expression was also the results of a smaller percentage of cells expressing SOX2 in the dorsal side and SOX9 in the lateral side of the otic vesicle, respectively. The expression pattern pattern of SOX2 and SOX9 overlapped most strongly at the ventro-medial side which has been predicted to give rise to the saccule, cochlea and part of the vestibular ganglion in chick (Wu and Oh, 1996) and mice (Farinas et al., 2001). Fig. 1C shows a gradient of expression signal in the otic vesicle which varied from the dorsal–medial side with predominant SOX9 expression signal, to the ventral–medial side with overlapping SOX2 and SOX9 signal, and finally to the lateral side with predominant SOX2 signal.
In the E10.5 otocyst, SOX9 was expressed in the otic epithelium and periotic mesenchyme. Regions with relatively stronger and weaker signal for SOX9 in the otic epithelium were observed (Fig. 2A′–C′). In the most anterior transverse sections of the E10.5 otocyst, SOX2 was expressed and overlapped with SOX9 on the ventro-lateral side of the otocyst (Fig. 2A–A″). In subsequent sections (Fig. 2B–B″), two domains of SOX2 expression were detected in the otocyst – one on the lateral side and one on the ventral side. SOX2 was also detected in the delaminating neuroblasts that surround the vestibular ganglion. In more posterior sections (Fig. 2C–C″), SOX2 was expressed and overlapped with SOX9-expressing cells found on two domains on the medial side of the otocyst. However, SOX9 expression signal was relatively weaker in the ventral–medial SOX2-expressing region than in the rest of the otocyst.
At E12.5, SOX2 expression marked the sensory primordia of the cristae (Fig. 3A, B and E), maculae (Fig. 3D) and cochlear duct (Fig. 3C). SOX9 was expressed in the periotic mesenchyme and the entire otic epithelium. In the lateral and posterior cristae and vestibular ganglion, relatively weak SOX9 expression signal than in the rest of the otic epithelium was detected (Fig. 3A′ and E). This difference was less obvious in the cochlea (Fig. 3C′). In the utricular macula (Fig. 3D), SOX2 was expressed in the entire sensory epithelia while SOX9 expression became restricted to the supporting cells.
By E14.5, the sensory epithelia of the utricle, saccule and cristae have initiated differentiation into hair and supporting cells. At E14.5–E18.5, SOX2 was expressed in both cell types while SOX9 expression became restricted to supporting cells only (Figs. 4–6). SOX9 was also expressed throughout the non-sensory otic epithelium and periotic mesenchyme of the developing inner ear.
In the cochlea, hair cells start to differentiate at E13.5–E14.5. Hair cell cycle exit starts in the apex at E11.5 (Matei et al., 2005) whereas differentiation starts in the mid-basal region of the cochlear duct and progresses in the apical direction (Chen et al., 2002). A more differentiated cochlear sensory epithelium is thus expected at the basal end. The gradient developmental pattern of the organ of Corti allows analysis on the graded developmental changes of the expression pattern of SOX2 and SOX9 in a single cochlea.
At E14.5, an obvious degree of differentiation of the sensory epithelia was noticeable when comparing the basal and apical cochlea (Fig. 4E and F). A clearer demonstration of the relationship between the differentiation gradient along the cochlear duct and the expression pattern of SOX2 and SOX9 can be seen at E16.5 and E18.5 (Figs. 5A–D, 6A–D). In the undifferentiated cochlear duct (Fig. 6A), SOX9 expression signal was found throughout the entire otic epithelium and overlapped with that of SOX2 at the sensory epithelium. As differentiation continued, SOX9 became restricted to the supporting cells including the inner and outer pillar cells, inner phalangeal supporting cells and Deiters’ cells (Fig. 6C). SOX2 maintained its expression in both the hair and supporting cells. At E18.5, SOX9 continued to be excluded from the hair cells but was expressed on the other part of the otic epithelium including the supporting cells, stria vascularis, the epithelial layer of the Reissner’s membrane, interdental cells, spiral limbus, spiral prominence (Fig. 6C and D). In particular, SOX9 expression signal in the interdental cells and spiral limbus was noticeably stronger than that in other parts of the otic epithelium.
The differentiation of the spiral ganglion progresses in a basal to apical manner. A temporal change of expression of SOX2 and SOX9 was observed in this process. SOX9 was not detected in the less differentiated spiral ganglion (Figs. 4E–F and and5E),5E), but in more differentiated spiral ganglion cells (Figs. 5F–H and 6E–F). Likewise, SOX2 immunostaining was initially faint and diffuse but was eventually up-regulated in many cells of the spiral ganglion. At E16.5, only SOX2 was detected at the apical end of the spiral ganglion (Fig. 5E). SOX9 was detected at the basal end of the spiral ganglion and was co-expressed with SOX2 (Fig. 5F–H). Expression of SOX2 and SOX9 was more prominent and in more spiral ganglion cells at E18.5 (Fig. 6E–F). TUJ1 is a neuron-specific marker. In the developing chick inner ear, SOX2 was not co-expressed with TUJ1 in the ganglion (Neves et al., 2007). Similarly, expression of SOX9 did not overlap with that of TUJ1, suggesting these cells are not neurons (Fig. 6G–H). SOX2 protein can be very stable, persisting in the absence of transcript and can be localized in the cytoplasm (Avilion et al., 2003). The developmental pattern of SOX2 protein expression reported here is consistent with the recent description of localization of Sox2 mRNAs in the inner ear by whole mount in situ hybridization (Nichols et al., 2008), suggesting concordance between transcript and protein stability in the developing inner ear.
The expression profiles of SOX2 and SOX9 relative to each other from E9.5 to E18.5 mouse inner ear development are summarized in Fig. 7A and B. This study of the expression pattern of SOX9 in relation to SOX2 facilitates the understanding of its roles in inner ear development. A compartment-boundary model of inner ear development has been proposed in which the otocyst is subdivided into lineage-restricted compartments based on regional gene expression (Fekete, 1996; Brigande et al., 2000; Fekete and Wu, 2002). Within these compartments, unknown molecular mechanisms govern the specification of different inner ear structures and their corresponding sensory patches. Fig. 7A summarizes the expression of SOX2 and SOX9 in an E9.5 otic vesicle and lists the inner ear structures that have been proposed by Fekete and Wu (2002) to arise from these compartments, demonstrating that our data are consistent with this model. Using SOX2 as a marker for the prosensory and sensory epithelium, the initial overlap but subsequent differential expression of SOX2 and SOX9 in the sensory epithelium suggests that SOX2 and SOX9 may have distinct roles in molecular pathways that direct cells towards different cell fates.
Pregnancies of F1 (C57BL/6N × CBA/Ca) mice were timed from the day of the vaginal plug which was designated as embryonic day 0.5 (E0.5). Animal care and sacrifice were conducted according to methods authorized by licences from the Department of Health of the Government of the Hong Kong Special Administrative Region.
Staged mouse embryos were fixed and processed using standard procedures (Hogan et al., 1994). Paraffin embedded mouse embryos were cut into 5–6 μm sections. Fluorescence immunohistochemistry was carried out using the following antibody dilutions with 10% heat-inactivated donkey serum (SIGMA D9663), 0.5% Triton X-100, in PBS as diluent: 1:1000 for rabbit anti-SOX9 (Chemicon AB5535), 1:600 for goat anti-SOX2 (Neuromics GT15098), 1:500 for mouse anti-TUJ1 (COVANCE MMS-435P), 1:1000 for Cy™3-conjugated donkey anti-rabbit IgG (Jackson 711-166-152), 1:500 for Alexa Fluor® 488-conjugated donkey anti-goat IgG (Molecular Probes A-11055), 1:500 for Cy™3-conjugated donkey anti-mouse IgG (Jackson 715-166-150), 1:500 for Alexa Fluor® 647 donkey anti-rabbit IgG (Molecular Probes A-31573). Sections were incubated in primary antibody at 4 °C overnight and in secondary antibody in dark at room temperature for 2 h. Nucleus staining was performed at room temperature for 15 min with 1:2000 TOTO®3 iodide (Invitrogen T3604) in PBS. Specimens were mounted in VECTASHIELD mounting media (Vector Labs). Images were captured using ZEISS LSM510 META confocal microscope and exported using LSM Image Browser (version 184.108.40.206).
This work was supported by the Research Grants Council of Hong Kong (HKU7222/97M, HKU2/02C, and HKU4/05C).