SEARCH

SEARCH BY CITATION

Keywords:

  • differentiation;
  • cellular patterning;
  • gradient

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. ACKNOWLEDGMENTS
  8. REFERENCES

Background: Insulin-like growth factor (IGF) signaling plays important roles in growth and cellular differentiation in the cochlear sensory epithelium. However, the roles of IGF binding proteins (IGFBPs), a family of IGF modulators, remain to be elucidated in this system. To begin to examine the role of IGFBPs, we used reverse transcription polymerase chain reaction (RT-PCR) and in situ hybridization to determine the temporal and spatial patterns of expression for Igfbps within the developing mouse cochlea. Results: RT-PCR analysis indicates that Igfpb2–5 are expressed in the cochlea between embryonic day (E) 13.5 and postnatal day (P) 0. In addition, the expression of each Igfbp significantly increased between E13.5 and P0. In situ hybridization indicates that Igfbp2, 3, 4, and 5 have distinct and complementary expression patterns in the developing cochlea. Moreover, expression patterns of Igfbp3 and 5 demonstrate contrasting gradients along the basal-to-apical axis of the cochlea. Conclusions: Igfbp2–5 are expressed in distinct and complementary patterns during cochlear development. These data suggest that IGFBPs may act to precisely regulate activation of IGF signaling in the developing cochlea in a cell type-specific manner, contributing to cellular patterning and differentiation in the cochlea. Developmental Dynamics 242:1210–1221, 2013. © 2013 Wiley Periodicals, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. ACKNOWLEDGMENTS
  8. REFERENCES

The mammalian cochlea acts as the primary organ for auditory perception. The sensory epithelium of the cochlea (the organ of Corti) contains two types of hair cells (inner and outer hair cells) and at least six different types of supporting cells, which are arranged in precise rows to surround the adjacent hair cells. During development, these cellular components become organized to form a highly ordered cellular mosaic. As a result of the development of this pattern, the cochlea is an excellent model to study morphogenesis, cell diversity, and cellular patterning. The cochlear duct develops from an out-pocketing of the ventral otocyst beginning around embryonic day (E) 10.5 in mice. However, the epithelial cells that comprise the duct are relatively uniform at this stage. Between E13 and E19, individual sensory progenitors (prosensory cells) within the cochlear duct become committed to specific cell fates and then organize into the characteristic cellular mosaic of the organ of Corti. The processes required for the formation of this sensory epithelium are presumed to be regulated by various molecular and cellular mechanisms; however, these factors remain largely unknown.

The importance of the IGF signaling pathway in development of the inner ear has been highlighted by several recent publications examining different aspects in the auditory system. In humans, Woods et al. reported a 15-year-old boy with a homozygous IGF1 mutation associated with severe bilateral hearing loss as well as short stature and mental retardation (Woods et al., 1996). In mice, deletion of Igf1 leads to a decrease in the initial size of the spiral ganglion as well as subsequent progressive loss of neurons postnatally (Camarero et al., 2001). The initial decrease in size is apparently a result of the loss of dividing neuroblasts within the developing statoacoustic ganglion (Camarero et al., 2003). Deletion of the Igf1 receptor (Igf1r) causes shortening of the cochlear duct and a delay in sensory cell development (Okano et al., 2011). These results demonstrate crucial roles for IGF signaling in development of both the cochlea and spiral ganglion.

The IGF signaling family is evolutionarily conserved (McMurtry et al., 1997; Duan, 1998) and plays important roles in development of multiple organs, including the ear, the brain, heart, eyes, and olfactory epithelium (Shirke et al., 2001; Hsieh et al., 2004; Laurino et al., 2005; Laustsen et al., 2007; Scolnick et al., 2008). The family consists of two ligands (IGF1, IGF2), two receptors (IGF1R, IGF2R), and six binding proteins (IGFBP1–6) with high affinity for the IGF ligands. Although the IGFBPs are named for their high affinity to IGF ligands, they have been shown to have a wide variety of actions (Mohan and Baylink, 2002). These include multiple effects on IGF ligands, including sequestration and effective inhibition, prolonged circulating half-life leading to a circulating storage reservoir, and concentration in target tissues (Duan and Xu, 2005). The tissue availability of IGFs and their access to cell receptors are thought to be regulated by IGFBPs, with local synthesis of IGFBPs by tissues that require IGF function, such as the gestational endometrium and the vascular smooth muscle cells (Han et al., 1996; Duan, 2002). Moreover, at least two of the IGFBPs, (IGFBP3 and 5) have been suggested to function independently of IGF (Mohan and Baylink, 2002; Schneider et al., 2002; Yamada and Lee, 2009). Based on these data, IGFBPs appear to act as key modulators of the IGF signaling pathway.

Although previous studies demonstrated that Igf1, Igf2, and Igf1r are expressed in the embryonic cochlea and play key roles in cochlear development (Sanchez-Calderon et al., 2010; Okano et al., 2011), roles for IGFBPs have not been determined. In zebrafish, Li et al. (2005) reported that knockdown of Igfbp3 disrupted hair cell differentiation and semicircular canal formation, which suggests that IGFBPs might play an essential part in the control of IGF signaling activity within the inner ear. To understand the specific aspects of cochlear development that may be controlled or modified by IGFBPs, it will be important to understand the functional properties of these factors as well as the mechanism through which they regulate or modulate the activity of IGF signaling.

To begin an analysis of IGFBPs in the developing cochlea, we studied the mRNA expression patterns for all six by reverse transcription polymerase chain reaction (RT-PCR) and in situ hybridization. We report a complete overview of the spatial and temporal pattern of expression of Igfbps in the cochlea between mid-gestation and birth. Results indicate distinct patterns of expression for each Igfbp, suggesting roles in the regulation of IGF signaling in a cell type-specific manner.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. ACKNOWLEDGMENTS
  8. REFERENCES

Igfbp2–5 Are Expressed in the Developing Cochlea

To determine which Igfbps are expressed in the developing mouse cochlea, we conducted RT-PCR using total RNA extracted from embryonic cochleae of wild type mice at E13.5, 16.5, and postnatal day (P) 0. RT-PCR analysis indicates that Igfbp2, 3, 4, and 5 are all expressed in the cochlea between E13.5 and P0. By contrast, Igfbp1 and -6 were not detected in the cochlea at any of the tested developmental time points (Fig. 1A). Arbp was used as an internal control (Yabe et al., 2003). The results for Igfbp1 and 6 are in concordance with existing data for these mRNAs during mouse embryonic development (Schuller et al., 1993). Next we determined whether there are any changes in the level of mRNA expression for any of the Igfbps over the course of cochlear development. Quantitative PCR (qPCR) data indicate that the expression of Igfbp3 and 4 dramatically and significantly increases between E13.5 and P0 (Fig. 1B). Significant increases in the levels of expression for Igfbp2 and 5 were also observed, although the overall increases for both of these factors were more modest by comparison with changes in Igfbp3 and 4. Based on the results described above, we undertook a comprehensive analysis of the temporal and spatial expression patterns of Igfbp2–5 in the developing cochlea by in situ hybridization.

image

Figure 1. Expression of Igfbps in the developing cochlea. A: Reverse transcription polymerase chain reaction (RT-PCR) analysis of expression of Igfbp1–6 in the cochlea at embryonic day (E) 13.5, E16.5, and postnatal day (P) 0. Igfbp2–5 are expressed in the cochlea at all the time points examined, whereas Igfbp1 and 6 are not expressed at any of these developmental stages. Arbp is a ubiquitously expressed mRNA that is used as an internal control. B: Quantitative PCR study of expression of Igfbp2–5 at E13.5, E16.5, and P0. The expression level for Igfbp3 increases by 2.72-fold between E13.5 and E16.5, and by P0 reaches a 4.68-fold increase relative to E13.5. In addition, expression levels of Igfbp2, 4, and 5 increase significantly between E13.5 and P0. Error bars represent standard error of the mean. *P < 0.05, analysis of variance followed by Dunnett's Test.

Download figure to PowerPoint

Igfbp2 Is Expressed in Reissner's Membrane and Periotic Mesenchymal Cells

In situ hybridization for Igfbp2 at E12.5 and E13.5 indicated broad expression in periotic mesenchymal cells surrounding the cochlear duct, including regions that will develop as fibrocytes in the spiral ligament and the spiral limbus (asterisks, Fig. 2A,B). Mesenchymal expression persists through P0 (Fig. 2B–E). In addition, beginning at E13.5, Igfbp2 mRNA was also observed in the roof of the cochlear duct in a region that will develop as Reissner's membrane (Fig. 2B,C). In contrast with mesenchymal expression, Igfbp2 was absent from the cochlear duct at E12.5 (Fig. 2A), indicating an onset of expression around E13.5. Expression of Igfpb2 in developing Reissner's membrane was also present in sections of E16 and P0 cochleae (arrows in Fig. 2C,E). In addition, robust expression was observed in the tympanic border cells, the spiral limbus, and the lateral cochlear wall (Fig. 2E, arrow heads and asterisks). At P0, Igfbp2 expression was also observed in the organ of Corti (Fig. 2E, arrow) in cells located directly above the spiral vessel, a position that correlates with the region of the developing pillar cells. To determine whether Igfbp2 correlates with the pillar cell region, we conducted immunohistochemistry for S100 and Prox1 at P0. S100 is expressed in the cell bodies of pillar cells and Deiters' cells at P0 (Buckiova and Syka, 2009) (Fig. 2G), while Prox1 is expressed in the nuclei of pillar cells and Deiters' cells (Bermingham-McDonogh et al., 2006) (Fig. 2H). Comparison of the expression pattern for Igfbp2 (Fig. 2F) with S100 and Prox1 indicates that Igfpb2 is expressed in the pillar cell region at P0.

image

Figure 2. Expression pattern of Igfbp2 in the developing cochlea. A: Expression of Igfbp2 in a section through the cochlear duct at embryonic day (E) 12.5. Medial–lateral orientation is indicated and is the same for panels B, C, E–H. B: Expression is present in the area that will develop as Reissner's membrane (blank arrow) and in mesenchymal tissue surrounding the cochlear duct (asterisks) at E13.5. C: At E16.5, expression persists in mesenchymal regions that will give rise to the spiral limbus and spiral ligament (asterisks). D: At postnatal day (P) 0, Igfbp2 is strongly expressed throughout the cochlea, including in Reissner's membrane (blank arrows). Base and apex are indicated. E: At P0, strong expression is present in Reissner's membrane (blank arrows), the fibrocytes of the spiral ligament and the spiral limbus (asterisks), tympanic border cells (arrowheads) and in pillar cell region (arrow). F: High magnification view of the organ of Corti from the middle turn at P0. The pillar cells (arrow) and the spiral vessel (arrowhead) are indicated. G: A comparable section to that in F, but with pillar cells and Deiters' cells labeled with anti-S100 and actin. For reference, the spiral vessel is indicated (arrowhead). H: Comparable section to that in E and F labeled with anti-Prox1. The spiral vessel is indicated (arrowhead). Scale bars = 50 μm in A (applies to B,C,E), 100 μm in D, 20 μm in F (applies to G,H).

Download figure to PowerPoint

Igfbp3 Expression Correlates With Expression of Prosensory Domain Markers

Initial expression of Igfbp3 within the cochlea was observed at E12.5 in a group of cells located in the medial edge of the cochlear duct in a region that correlates with the presumptive prosensory domain (Fig. 3A). Cross-sections at E13.5 indicate that Igfbp3 is expressed in a basal-to-apical gradient (Fig. 3B) that corresponds with the known gradient of cellular differentiation (reviewed in Kelley, 2007). At E15.5, expression of Igfbp3 resolves into two distinct bands in the basal region of the cochlea (Fig. 3C), while expression in the apex remains as a single medial band (Fig. 3E, data not shown). The two bands in the base appear to be a medial band centered around the developing inner hair cell, and a more lateral band located around the developing outer hair cells. This pattern becomes more distinct at E16.5 (Fig. 3D), however, expression in specific cell types cannot be determined. By P0, expression of Igfbp3 in the basal part of the cochlea is restricted to border cells, inner phalangeal cells, and Deiters' cells, but expression in the apical turn remains broad and localized to the medial part of the cochlear sensory epithelium (Fig. 3E,F,F′). Consistent with a basal-to-apical gradient of development, expression of Igfbp3 in the middle turns of the cochlea at P0 has resolved into two bands while at the very apex a single broad band is still present (Fig. 3E).

image

Figure 3. Expression of Igfbp3 corresponds with the prosensory domain. A: Cross-section through the basal turn of the cochlea at embryonic day (E) 12.5, Igfbp3 is expressed in the cochlear duct at the medial edge. Medial–lateral orientation is indicated and is the same for panels B, C, D, F–N. B: Cross-section illustrating the apical and basal turns of the cochlear duct at E13.5. Igfbp3 is expressed in the base (arrow), but is absent in the apex. C: At E15.5, expression is restricted to two bands (arrows). D: Expression of Igfbp3 at E16.5 (arrows) is similar to the pattern at E15.5. E: Low magnification cross-section at P0 illustrates the apical-to-basal gradient of Igfbp3 expression (arrowheads). The pattern in the mid-apical region at postnatal day (P) 0 resembles that in the base at E16.5. F: High magnification view of the basal cochlear duct at P0. Igfbp3 is expressed in the sensory domain in two bands of cells, a medial band that extends from the basement membrane to the lumenal surface and a lateral band that only extends part way from the basement membrane. F′: Zoomed in view of the organ of Corti illustrates expression of Igfbp3 in cells adjacent to hair cells. G–J: Expression of Igfbp3 (G) overlaps with expression of Jagged1 (Jag1; magenta in H and I) at E13.5. The expression domain for p27Kip1 (green in H) is located lateral to the Igfbp3 domain. Sox2 expression (green in I) partially overlaps with both the Igfbp3/Jag1 and p27Kip1 domains. Panel J shows a pseudo-colored overlaid image of G and I to illustrate the close correlation between Igfbp3 and Jag1. K–N: At P0, Igfbp3 is expressed in Deiters' cells (arrows), inner phalangeal cells (IPhC), border cells (BC), and parts of Kolliker's organ (brackets) (K–M) at P0. Hair cells (K) are negative for Igfbp3. As was the case at E13.5, expression of Igfbp3 still overlaps with Jag1 (L). Pseudo-colored overlay of Fig 3K and M is shown in N. Scale bars = 100 μm in B,E, 50 μm in A (applies to C,D,F), 20 μm in F′, G (applies to H–J), and K (applies to L–N).

Download figure to PowerPoint

Next we examined the relationship between Igfbp3 expression and expression of prosensory markers by immunohistochemistry for Jagged1 (Jag1), Cdkn1b (p27Kip1), and Sox2. At E13.5, the expression of Igfbp3 appears to strongly correlate with the domain of Jag1 expression (Fig. 3G–J). Overlap with Sox2 and p27Kip1 was less well correlated. In fact, the band of p27Kip1 expression appears to be located just adjacent and lateral to the domain of Igfbp3 expression (Fig. 3H). To confirm that Igfbp3 becomes restricted to supporting cells at P0, we compared Igfbp3 expression with a hair cell marker, Myosin VI (Myo6), and a supporting cell marker (Jag1; Fig. 3K–N). While Myo6-positive hair cells are located near the lumenal surface of the organ of Corti, Igfbp3 is expressed in the area adjacent to the basement membrane in a complementary manner to Myo6 (Fig. 3K,L). In contrast, Igfbp3 expression overlaps strongly with Jag1 (Fig. 3M,N). These results suggest that the expression pattern for Igfbp3 strongly correlates with Jag1 at several time points between E13 and P0. At early time points, both Igfpb3 and Jag1 are expressed broadly within the prosensory domain, but as development proceeds, both become restricted to supporting cells.

Igfbp4 Is Expressed in Periotic Mesenchyme and Claudius' Cells

At E13.5, Igfbp4 is broadly expressed in periotic mesenchymal cells located in the future scala tympani and scala vestibuli, some of which will develop as fibrocytes in the spiral limbus and the spiral ligament (Fig. 4A,F, asterisks). Expression in these regions persists from E13.5 to P0. In addition, beginning at E14.5, Igfbp4 begins to be expressed at the lateral side of the cochlear duct in a relatively limited group of cells (Fig. 4B, arrow) This pattern of expression is maintained at E16.5 (Fig. 4C) and appears to correlate with cells that will develop as Claudius' cells. To confirm this, expression of Igfbp4 was compared with Bmp4 (Fig. 4D), which has been shown to be restricted to developing Claudius' cells (Takemura et al., 1995). At P0, expression of Igfbp4 in the cochlear sensory epithelium persists in Claudius' cells along the length of the cochlea and is also present in the modiolus, the spiral limbus, and the lateral cochlear wall (Fig. 4E,F).

image

Figure 4. Igfbp4 is expressed in periotic mesenchyme and Claudius' cells. A: Cross-section through the base of the cochlea at embryonic day (E) 13.5. Igfbp4 is broadly expressed in the periotic messenchyme (asterisks) surrounding the cochlear duct; however, no expression of Igfpb4 is seen in the cochlear epithelium. Medial–lateral orientation is indicated and the same for panels B–D, and F. B: At E14.5, expression of Igfbp4 persists in the periotic mesenchyme and limited expression is present in the lateral part of floor of the cochlear duct (arrow). C: At E16.5, strong expression of Igfbp4 is still present in the periotic mesenchyme and expression in the lateral region of the cochlear duct is more pronounced (arrow). D: Expression of Bmp4 (arrow), which is restricted to Claudius cells, appears to overlap with Igfbp4. E: At postnatal day (P) 0, Igfbp4 is expressed in the cochlear modiolus, the spiral ligament (asterisks), and the spiral limbus (white arrowheads), but is not expressed in the spiral ganglion (arrows). F: Cross-section through the basal turn of the cochlea at P0. Expression of Igfbp4 within the duct is restricted to the Claudius cells (arrow). Expression is also observed in fibrocytes in the spiral ligament and spiral limbus (asterisks). Scale bars = 100 μm in D, 50 μm in A (applies to B,C,E,F).

Download figure to PowerPoint

Igfbp5 Is Expressed in Outer Sulcus Cells and Kolliker's Organ

At E10.5, Igfbp5 is expressed in the medio-ventral portion of the otocyst, the region that corresponds with the tip of the elongating cochlear duct (Fig. 5A). Igfbp5 is the only member of IGFBP family that is expressed at the beginning of cochlear development. At this stage, Igf1r is expressed in the ventral portion of the otocyst in a broad and diffuse pattern (Fig. 5B, arrows), while Sox2 is broadly expressed in the medio-ventral quadrant of the otocyst (Fig. 5C). A comparison of Figs. 5A with 5C suggests that Igfbp5 is expressed in a subdomain of the larger Sox2 expression region. At E13.5, Igfbp5 is broadly and robustly expressed in the floor (the dorsal side of the duct that will give rise to the inner sulcus, organ of Corti and outer sulcus) of the apical turn of the cochlear duct, while expression in the basal region is largely restricted to the lateral edge of the duct with some weak expression in the prosensory domain (Fig. 5D, bracket). This result indicates a graded pattern of expression along the apical-to-basal axis of the cochlea. At E15.5, Igfbp5 expression persists in the lateral edge of the cochlear duct, but expression is also present in the developing sensory domain in a pattern that, to some extent, offsets the pattern observed for Igfbp3 expression (Fig. 5E, arrowheads, compare with Fig. 3C). By P0, the expression pattern for Igfbp5 has become even more varied (Fig. 5F). In the apex, Igfbp5 is still expressed broadly in the lateral half of the floor of the duct, while, in more basal regions, expression becomes restricted to distinct subsets of cells within the floor of the duct. In particular, Igfbp5 is strongly expressed in the lateral half of Kolliker's organ (KO), supporting cells within the organ of Corti (OC), and in the outer sulcus cells (OS; Fig. 5G). Finally, it should be noted that while at E15.5 Igfbp5 is expressed in the mesenchyme that will form the scala tympani and scala vestibuli, no expression of Igfbp5 is observed in the fibrocytes of the spiral ligament or spiral limbus at P0. This pattern clearly contrasts with the patterns of expression for Igfbp2 and 4 which persist in the spiral ligament and spiral limbus.

image

Figure 5. Igfbp5 is expressed broadly throughout the cochlear duct. A: Cross-section through the otocyst at embryonic day (E) 10.5. The medioventral extension of the cochlear duct toward the bottom of the panel is positive for Igfbp5. D, dorsal; L, lateral. B: At E10.5, Igf1r is diffusely expressed throughout the ventral region of the otocyct (arrows). C: Sox2 is also expressed in the medioventral portion of otocyst, but in a broader pattern by comparison with Igfbp5. The apex and base of the elongating cochlear duct are indicated. D: At E13.5, Igfbp5 is expressed broadly in the floor of cochlear duct in the apex, but expression in the basal turn is restricted to the lateral edge. Prosensory domain; bracket. E: Expression of Igfbp5 resolves into two bands within the developing organ of Corti (arrowheads) and a third domain of expression in the extreme lateral region of the cochlear duct at E15.5. The medial–lateral orientation is indicated and is the same in G. F: Low magnification view at P0. Igfbp5 is expressed in the modiolus and some residual mesenchymal cells (arrows). G: High magnification view of the basal cochlear duct at P0. Igfbp5 is expressed in Kolliker's organ (KO), some supporting cells in the organ of Corti (OC), and the outer sulcus cells (OS). Scale bars = 50 μm in C (applies to A,B) and E (applies to G), 100 μm in D,F.

Download figure to PowerPoint

Igfbp3 and 5 Are Expressed in a Complementary Pattern in the Cochlea at E14.5

Of the four Igfbps expressed in the developing cochlea, the patterns of expression for Igfbp3 and 5 were most intriguing in terms of possible roles in prosensory formation. To examine this possibility further, we compared the expression patterns of Igfbp3 and 5 with that of p27Kip1 at E14.5, At this stage, p27Kip1 is probably the most definitive marker of the prosensory domain and expression of p27Kip1 in the cochlea at this time is thought to mark the prosensory cells that have exited the cell cycle and are poised to become committed to a specific cell fate (Ruben, 1967; Chen and Segil, 1999; Lowenheim et al., 1999). At the extreme apex, Igfbp3 is expressed in the medial corner of the cochlear duct (Fig. 6A), but at progressively more basal positions, Igfbp3 expression moves toward the middle of the floor of the duct (Fig. 6B,C). By contrast, at the same stage Igfbp5 is expressed broadly in the lateral two-thirds of the floor of the duct in the apex (Fig. 6D). At more basal positions, expression becomes restricted to the lateral edge (Fig. 6E,F). These results are consistent with complementary domains of expression for Igfbp3 and 5. A comparison with the domain of p27Kip1 expression indicates that p27Kip1 expression seems to overlap partially with the domain of Igfbp5 expression in the apical and middle part of the cochlear duct at E14.5 (Fig. 6D,G,J,E,H,K), but correlates more precisely with expression of Igfbp3 in the most basal turn (Fig. 6C,I). In the base, p27Kip1 expression almost completely overlaps with the Sox2 domain characterizing the prosensory cell population at E14.5, and onward (Fig. 6L). These data suggest that Igfbp3 and 5 may have unique roles in cochlear prosensory formation and differentiation of cochlear sensory cells.

image

Figure 6. Expression of Igfbp3 and 5 partially overlaps with expression of p27Kip1 at E14.5. Cross-sections through the indicated turns of the cochlear duct at E14.5. The medial–lateral orientation indicated in A is applied to all panels. The developing spiral vessel located underneath the presumptive organ of Corti is indicated in each panel (asterisks) for orientation. A–C: Expression of Igfbp3 is restricted to the medial edge of cochlear duct in the apex (A), but moves more laterally in the middle turn (B), and is localized to the middle of the floor of the duct in the base (C). D: In contrast, Igfbp5 is expressed broadly in the apex in a pattern that complements that of Igfbp3. E: In the middle turn, Igfbp5 expression extends laterally, but is down regulated in the middle region of its expression domain. F: This down-regulation continues in the base such that expression is restricted to the extreme lateral edge. G–I: Comparable views illustrating expression of p27Kip1 in the apex, middle, and base are shown in G, H, and I, respectively. G,I: p27Kip1 is expressed in a gradient from apex to base with broader expression in the apex (G) and progressive restriction in more basal regions (I). J–L: Overlap between Sox2 and p27Kip1 defines the prosensory domain. C,L: In the apex, there is no overlap between Igfbp3 and the prosensory domain, but in the middle and basal turns, there is partial to extensive overlap between the two domains. D–F,J–L: Expression of Igfbp5 overlaps with the prosensory domain in the apex (D,J) and the middle turn (E, K) but not in the base (F,L). Scale bar = 50 μm in A (applies to B–L).

Download figure to PowerPoint

Igfbp2–5 Are Expressed in Specific Cell Types Within the Cochlear Duct at P0

Although none of the Igfbps were present in differentiated hair cells or spiral ganglion neurons, their relative patterns of expression within the organ of Corti and its surrounding tissues at P0 are intriguing. Igfbp2 is expressed exclusively in the pillar cell region (arrow, Fig. 7A), while Igfbp3 appears to be expressed in the border cells, inner phalangeal cells, and Deiters' cells (Fig. 7B, arrowheads). Igfbp4 appears to be expressed in expressed in the Claudius' cells (Fig. 7C, bracket), and Igfbp5 appears to be expressed in Kolliker's organ (KO), Deiters' cells in the organ of Corti (OC), and the outer sulcus cells (OS) (Fig. 7D). As is summarized in Figure 8, these expression patterns are largely complementary with limited overlap, and suggest possible complex actions for IGF signaling in different supporting cell populations.

image

Figure 7. Cell type-specific expression of Igfbps in the organ of Corti at postnatal day (P) 0. A–D: High magnification images of the basal turn of the cochlear duct at P0. The medial–lateral orientation indicated in A is the same for all panels. A,B: Igfbp2 is expressed in a region that correlates with the pillar cells (A, arrow), while Igfbp3 is expressed in the border cells, inner phalangeal cells, and Deiters' cells (B, arrowheads). C,D: Igfbp4 is expressed in the Claudius' cell (C, bracket) and Igfbp5 is expressed in Kolliker's organ (D, KO), border cells, inner phalangeal cells, Deiters' cells, and the outer sulcus cells (OS). Igfbp3 and Igfbp5 are co-expressed in supporting cells within the organ of Corti, although Igfbp5 is also expressed in additional cells in KO and the OS. Scale bar = 100 μm in A (applies to B–D).

Download figure to PowerPoint

image

Figure 8. Schematic depiction of expression of members of the insulin-like growth factor (IGF) signaling family in the basal part of the cochlea and the spiral ganglion at different developmental time points. The expression of IGF ligands, receptor, and binding proteins is shown at four different stages: embryonic day (E) 10.5, E13.5, E16.5, and postnatal day (P) 0. Only Igf1r and Igfbp5 are expressed in the cochlear domain of the otocyst at E10.5. For E10.5, orientations for dorsal (D) and lateral (L) are indicated. For all other diagrams, medial is to the left and lateral is to the right. Igf2 is the major IGF ligand expressed in the cochlea at E13.5, but is gradually replaced by Igf1 by P0. Igf1r is expressed broadly in the sensory epithelium, including developing hair cells, throughout embryonic development. By contrast, Igfbps are expressed in cell type-specific patterns between E13.5 and P0. Igfbp3 and 5 are expressed in nonoverlapping domains at E13.5, but both are expressed in supporting cells by P0. While Igf1r is expressed in the spiral ganglion during embryonic development, none of the IGF ligands or binding proteins are present. Expression of Igf1, Igf2, and Igf1r is cited from two previous studies by Sanchez-Calderon, et al. (2010) and Okano, et al. (2011). GER, greater epithelial ridge; BC, border cells; IHC, inner hair cells; IPC, inner phalangeal cells; PC, pillar cells; OHC, outer hair cells; DC, Deiters' cells; HC, Hensen's cells; CC, Claudius' cells; OS, outer sulcus cells.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. ACKNOWLEDGMENTS
  8. REFERENCES

In this study, we analyzed the spatial and temporal patterns of expression for Igfbps in the developing mouse cochlear duct between E10.5 and P0 using RT-PCR and in situ hybridization. Unique and dynamic patterns of expression were observed for Igfbp2, 3, 4, and 5 throughout cochlear development (Fig. 8). At early time points, before the initiation of hair cell differentiation, all four Igfbps are expressed in unique regions of the cochlear duct, with Igfbp3 and 5 expressed in regions that apparently overlap, at least transiently, with the prosensory domain. In contrast, Igfbp2 and 4 are expressed in restricted subsets of cochlear duct cells, as well as in surrounding periotic mesenschyme. At later developmental time points, all four Igfbps are expressed in specific subsets of cells within the organ of Corti and are retained in various nonsensory tissues in the cochlea. As will be discussed below, considering the known modulatory abilities of IGFBPs on the IGF signaling pathway, these data suggest that IGFBP expression may act to create unique regional and cell type-specific IGF signaling within the developing cochlea and organ of Corti.

Igfbps and Prosensory Cell Formation

At early stages of sensory differentiation, Igfbp3 and 5 are largely expressed in reciprocal gradients that occupy most of the floor of the duct. At E13.5 and E14.5, Igfbp5 is expressed in the apex in a broad band that includes at least the lateral 3/4s of the floor of the duct. At the same time points, apical expression of Igfbp3 is restricted to the medial 1/4 of the duct. At more basal, and therefore more mature, positions, Igfbp5 expression in the floor narrows as a result of decreased expression on its medial side and appears to shift laterally. At the same time, Igfbp3 expression also shifts laterally toward the prosensory domain. At later developmental time points, Igfbp5 expression expands medially to overlap with Igfbp3 within the prosensory domain, and specifically in the Deiters' cells and inner phalangeal cells.

The specific roles of either Igfbp3 or 5 within the cochlea are unclear. Many IGFBPs, including 3 and 5, have been reported to act as either inhibitors or enhancers of IGF signaling depending on concentration and context. The coincident expression of Igfbp3 and 5 in developing supporting cells beginning around E15 could reflect a change in the responsiveness of these cells to IGF signaling. The timing coincides with the first obvious cochlear defects in Igf1r mutants (Okano et al., 2011), suggesting an increased importance of IGF signaling. Moreover, inhibition of IGF signaling at different developmental time points in vitro suggests specific roles for IGF signaling in both cellular differentiation and patterning. Whether co-expression of both Igfbp3 and 5 in supporting cells increases, decreases, or modulates the response of cells to IGF signaling remains to be determined.

These results suggest that the dynamic expression of IGFBPs could play a crucial role in modulating differential activation of IGF signaling in specific regions of the duct at specific developmental time points.

Cell Type-Specific Expression of Igfbps and Regulation of IGF Signaling

Despite differences in their initial patterns of expression, both Igfbp3 and 5 are expressed in Deiters' cells, inner phalangeal cells, and border cells at P0, with Igfbp5 expression extending more medially than that of Igfbp3 within Kolliker's organ (Fig. 8). At the same time point and region, Igfbp2 expression is restricted to the pillar cell region while expression of Igfbp4 is limited to the outer sulcus. By comparison, expression of Igf1r at P0 is broad and includes all of the cell types that are positive for one or more of the Igfbps. Of interest, hair cells, which are significantly affected in Igf1r mutants, express none of the Igfbps once they progress beyond the progenitor stage. These observations would be consistent with a straight-forward hypothesis in which expression of Igfbps acts to inhibit IGF signaling in progenitor cells and supporting cells. However, the unique pattern of cell-type specific expression of different Igfbps in different supporting cell populations suggests more complex signaling interactions.

Similar cell or tissue specific expression patterns for Igfbps have been reported previously in other tissues. Igfbp3 and 5 are expressed in the terminal end buds and ducts of mammary glands, while Igfbp2 and 4 are predominantly expressed in surrounding stromal cells (Allar and Wood, 2004). In the mature kidney, predominant expression of Igfbp4 is confined to proximal tubules, whereas Igfbp2 is expressed in thin limbs of Henle's Loops. By contrast, Igfbp5 expression is restricted to glomerular mesangial cells and peritubular capillaries of the medulla, while Igfbp3 is expressed in the endothelial cells of the renal capillary system (Lindenbergh-Kortleve et al., 1997). In skeletal development, Igfbp5 and 6 are expressed in chondrocyte precursors, while Igfbp2, 4, 5, and 6 are co-expressed in osteoblasts (Wang et al., 1995). The observation of multiple systems in which different Igfbps are expressed in closely associated cell types suggest that different IGFBPs may play distinct roles in modulating IGF signaling during embryonic development. However, dissecting these roles has proven to be difficult because Igfbp mutant mice often show no obvious phenotypes, suggesting redundancy or functional compensation among Igfbps (Ning et al., 2006).

In addition to expression in the epithelium of the cochlear duct, Igfbp2, 4, and 5 are widely expressed in the periotic mesenchyme surrounding the cochlear duct. These results, along with the demonstration of a shortened cochlear duct in Igf1r mutants, suggests a possible role for IGF signaling in epithelial–mesenchymal interactions that mediate outgrowth of the duct and formation of the bony labyrinth. Similar roles for IGF signaling have been suggested based on studies of mammary gland development (Heckman et al., 2007). However, in the cochlea, it is not clear whether these effects are mediated directly through activation of IGF1R in periotic mesenchyme or indirectly as a result of changes in development of the cochlear epithelium which might then influence mesenchymal development.

Interactions Between IGF and Other Signaling Pathways

As discussed, the expression of Igfbp3 appears to largely overlap with that of Jag1, suggesting a potential interaction between IGF and Notch signaling pathways. In skin development, deletion of Notch1 alters the expression of the components of the IGF signaling pathway, including Igfbp3, and leads to an apparent inhibition of IGF signaling (Lee et al., 2007). In the developing inner ear, deletion of Jag1 or the Notch-effector Rbpj, leads to a significant decrease in the size of the prosensory domain and defects in hair cell differentiation (Brooker et al., 2006; Kiernan et al., 2006; Basch et al., 2011; Yamamoto et al., 2011). Some of these phenotypes, in particular those related to hair cell differentiation, are similar to those observed in the Igf1r mutants, suggesting that Notch-dependent expression of Igfbp3 could play a key role in regulating IGF signaling within the developing cochlea.

Similarly, expression of Igfbp4 in the cochlear epithelium appears to overlap both temporally and spatially with the Bmp4 expression domain. Previous work has demonstrated that IGF signaling acts to inhibit BMP by means of MAPK-mediated phosphorylation and subsequent degradation of Smad1 (Pera et al., 2003; Eivers et al., 2008). In the cochlea, Ohyama et al. (2010) reported that BMP signaling is required for specification of the outer sulcus and prosensory domains as well as for patterning sensory and nonsensory regions of the cochlear duct. The specific effects of IGFBP4 are unclear, but because Igf1r is expressed throughout the epithelium, Igfbp4 expression in the Bmp4 expression domain could act to prevent IGF-mediated inhibition of BMP signaling in this region.

In conclusion, four of the six Igfbps are expressed within the developing cochlear duct in unique and fairly specific patterns. Recent studies have demonstrated the importance of IGF signaling in the auditory system. Given the demonstrated roles of IGFBPs in modulating IGF signaling, we believe a better understanding of the roles of these molecules will provide valuable insights regarding IGF signaling in auditory development and homeostasis.

EXPERIMENTAL PROCEDURES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. ACKNOWLEDGMENTS
  8. REFERENCES

Animals

Timed pregnant ICR mice were purchased from Charles River Laboratories. All animals were maintained based on the standards outlined in the National Institutes of Health Guide for the Care and Use of Laboratory Animals. Noon on the day when the vaginal plug was identified was designated as E0.5, and P0 was defined as the day of birth. All embryos were staged according to The Atlas of Mouse Development (Kaufman, 1992).

Quantitative RT-PCR

For the analysis of expression levels of IGF signaling components, cochlear ducts were collected from four pups per time points at E13.5, E16.5, and P0, and total RNA was extracted using the RNAquous Kit (Ambion, Austin, TX) according to the manufacturer's protocol. cDNA was synthesized using the Superscript III cDNA Synthesis Kit (Invitrogen, Carlsbad, CA). cDNA and primer sets were mixed with Power SYBR Green PCR Master Mix (Applied Biosystems Foster City, CA) and subjected to real-time PCR quantification using an ABI PRISM 7000 Sequence Detection System (Applied Biosystems). The relative amounts of mRNAs were calculated using the standard curve method. Mouse Arbp mRNA was used as an invariant control (Yabe et al., 2003). Sequences for primer sets for Arbp and Igfbps 1–6 are as follows; Arbp: 5′-atcaccacgaaaatctccag-3′ and 5′-ttcagcatgttcagcagtgt-3′, Igfbp1: 5′-ccgacctcaagaaatggaa-3′ and 5′-catctcctgctttctgttgg-3′, Igfbp2: 5′-atctctactccctgcacatcc-3′ and 5′-tccgttcagagacatcttgc-3′, Igfbp3: 5′-cacatcccaaactgtgacaa-3′ and 5′-ccatacttgtccacacacca-3′, Igfbp4: 5′-atccccattccaaactgtga-3′ and 5′-gatccacacaccagcacttg-3′, Igfbp5: 5′-actgtgaccgcaaaggattc-3′ and 5′-ttgtccacacaccagcagat-3′, Igfbp6: 5′-agaggcttctaccgaaagca-3′ and 5′-tccttgaccatctggagaca-3′.

In Situ Hybridization

In situ hybridization was performed as described previously (Jacques et al., 2007; Okano et al., 2011) for Igfpb2, Igfbp3, Igfbp4, Igfpb5, Igf1r, Sox2, and Bmp4 on 12 μm frozen sections from cochleae isolated at specific time points between E10.5 and P0. The sequences for the full length-mouse cDNAs were obtained from Open Biosytems (Thermo Scientific, Kalamazoo, MI), and used to generate digoxigenin-labeled sense and antisense RNA probes. At least two independent samples were analyzed for each time point to confirm expression of genes of interest.

Immunohistochemistry

For immunohistochemistry on frozen sections, cochleae were dissected out in ice-cold phosphate buffered saline and fixed in 4% paraformaldehyde overnight. Samples were cryoprotected through a sucrose gradient, embedded in Tissue-Tek O.C.T. Compound (Sakura, Torrance, CA) and sectioned on a cryostat at a thickness of 12 μm. The primary antibodies used in this study were as follows: rabbit anti-Myo6 (1:1,000; Proteus BioSciences, Ramona, CA), rabbit anti-Prox1 (1:2,000; Covance, Princeton, NJ), goat anti-Sox2 (1:250; Santa Cruz Biotechnology, Santa Cruz, CA), rabbit anti-Sox2 (Millipore, Billerica, MA), goat anti-Jagged1 (Jag1; 1:300; Santa Cruz Biotechnology), rabbit anti-p27Kip1 (1:300) (Thermo Scientific), rabbit anti-S100 (Thermo Scientific). Primary antibodies were visualized with Alexa Fluor (AF) 488- or AF555-conjugated secondary antibodies (1:500; Invitrogen). AF488-conjugated phalloidin (Invitrogen) was used to visualize F-actin. All images were obtained on a Nikon E800 (Nikon, Tokyo, Japan).

Statistics

One-way analysis of variance followed by Dunnett's test was used to determine significant differences for qPCR.

ACKNOWLEDGMENTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. ACKNOWLEDGMENTS
  8. REFERENCES

We thank Dr. Weise Chang for his excellent technical assistance and Dr. Junko Okano for critical discussion. This study was supported by funds from the Intramural Program at the National Institute on Deafness and Other Communication Disorders (M.W.K.), and by a grant from the Japan Society for the Promotion of Science Research Fellowship for Japanese Biomedical and Behavioral Researchers at NIH (T.O.).

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. ACKNOWLEDGMENTS
  8. REFERENCES