Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2019 Sep 1.
Published in final edited form as: Dev Growth Differ. 2018 Jun 19;60(7):431–441. doi: 10.1111/dgd.12543

Dorsovental Differences in cAMP Levels and Correlated Changes in the Subcellular Distribution of the PKA Catalytic Domain, Provide Further Evidence that PKA Signaling Coordinates Dorsoventral Patterning of the Otocyst

Sho Ohta 1,2, Gary C Schoenwolf 1,3
PMCID: PMC6133747  NIHMSID: NIHMS968630  PMID: 29920660

Abstract

Dorsoventral (DV) patterning of the otocyst gives rise to formation of the morphological and functionally complex membranous labyrinth composed of unique dorsal and ventral sensory organs. DV patterning results from extracellular signaling by secreted growth factors, which presumably form reciprocal concentration gradients across the DV axis of the otocyst. Previous work suggested a model in which two important growth factors, BMP and SHH, undergo crosstalk through an intersecting pathway to coordinate DV patterning. cAMP-dependent protein kinase A (PKA) lies at the heart of this pathway. Here, we provide further evidence that PKA signaling coordinates DV patterning, showing that both BMPs and SHH regulate cAMP levels, with BMPs increasing levels in the dorsal otocyst and SHH decreasing levels in the ventral otocyst. This, in turn, results in regional changes in the subcellular distribution of the catalytic domain of PKA, as well as DV regulation of PKA activity, increasing it dorsally and decreasing it ventrally. These new results fill an important gap in our previous understanding of how ligand signaling acts intracellularly during otocyst DV patterning and early morphogenesis, thereby initiating the series of events leading to formation of the inner ear sensory organs that function in balance and hearing.

Keywords: BMPs, Growth Factors, Inner Ear, Membranous Labyrinth, Patterning, SHH, Signaling, Transcription Factors

Graphical abstract

graphic file with name nihms968630u1.jpg

Introduction

The inner ear or membranous labyrinth is an exquisitely patterned organ that functions in both balance and hearing. These functions are partitioned dorsoventrally within the inner ear, with the dorsomost semicircular canals acting in balance by responding to rotational movements (angular acceleration) and the ventromost cochlea acting in hearing by detecting vibrations. At the junction between these two major sensory domains lies the utricle and saccule, two discrete organs that act in balance by responding to changes in head position vis-à-vis gravity (linear acceleration).

A major question driving the research in our laboratory over the last several years is, how is the initially simple, apparently uniformly-structured, otocyst transformed into the complex membranous labyrinth with regionally specific sensory functions? Past work has revealed that growth factors, such as members of the WNT, BMP, and Hedgehog families, secreted from embryonic tissues surrounding the otocyst, act on the otocyst to induce localized patterns of expression of transcription factors that regionally specify distinct areas of the otocyst (reviewed by Groves & Fekete 2012, Wu & Kelley 2012, Whitfield 2015, Basch et al. 2016). Because these growth factors are secreted into the extracellular space, either dorsal to or ventral to the otocyst, they are presumed to form concentration gradients that act differentially across the dorsoventral (DV) axis of the otocyst. That multiple factors impact the otocyst at different concentrations begs the question of how these factors are coordinated in space and time to induce a properly patterned inner ear.

Recent studies have led to a model of how DV patterning of the otocyst is at least partially coordinated. At the heart of this model is the enzyme cAMP-dependent protein kinase A (PKA) (Ohta et al. 2016a, b; reviewed by Ohta & Schoenwolf 2018). According to this model, BMPs secreted from the dorsal neural tube (and perhaps the cristae of the early otocyst) provide a dorsal signal that increases PKA activity in the dorsal otocyst, which in turn increases the post-translational proteolytic processing of full-length GLI3, increasing the ratio of GLI3R/GLI3A in the dorsal otocyst and inducing dorsal otocyst transcription factors such as DLX5 and NKX5.1/HMX3. Ventrally, SHH secreted from the notochord and overlying floor plate of the neural tube decreases PKA activity in the ventral otocyst, which in turn decreases the proteolytic processing of full-length GLI3, decreasing the ratio of GLI3R/GLI3A in the ventral otocyst and inducing ventral otocyst transcription factors such as PAX2 and OTX2.

Here, we provide further support for this model by showing that both BMPs and SHH regulate cAMP activity in the dorsal and ventral otocyst, respectively, with BMPs increasing activity and SHH decreasing it. This, in turn, results in regional changes in the subcellular distribution of the catalytic domain of PKA as well as DV regulation of PKA activity, increasing it dorsally and decreasing it ventrally.

Materials and Methods

Embryos

Fertilized White Leghorn chicken eggs were purchased from Merrill's Poultry Farm (Paul, Idaho, USA). Eggs were incubated at 38.5°C and staged (HH stages) according to the criteria of Hamburger and Hamilton (1951).

In ovo gene transduction

Chick eggs were windowed using standard techniques, and embryos were either electroporated (Ohta et al. 2010) or sonorporated (Ohta et al. 2003) in ovo to transfect expression vectors. For electroporation, the plasmid solution, consisting of either 2.0 μg/μl of pCAGGS-Gfp or 2.0 μg/μl of pCAGGS-caGli3A (Gli3AHIGH), pCAGGS-Gli3R, pCS2-Smad6, or pT2KCAGGS-caSmoA1-Gfp, containing 0.1% fast green to enable visualization of the injection site, was injected into the lumen of the developing otocyst (or cavity of the otic cup) using fine glass micropipettes. Embryonic stages were selected to target electroporation to the ventromedial (pCAGGS-Gfp and/or pCAGGS-Gli3R; HH stages 9-11) or the dorsolateral (pCAGGS-Gfp and/or pCS2-Smad6, pT2KCAGGS-caSmoA1-Gfp, or pCAGGS-caGli3A; HH stages 15-16) quadrant of the otocyst wall, and positive and negative electrodes were positioned accordingly. Two 50-msec pulses at 10 volts were then applied using a CUY21 electroporator. Embryos were collected after an additional 12 or 24 hours of incubation.

For sonoporation, a DNA-microbubble mixture was prepared by adding 10 μl of a plasmid DNA solution (concentration 2.0-4.0 μg/μl), consisting of pCAGGS-Gfp, pCAGGS-Bmp4, or pCAGGS-Nog (Noggin), to 10 μl of SonoVue (BRACCO, Protech)—a microbubble solution used for gene transfection. The DNA-microbubble mixture was then injected into the head mesenchyme adjacent to the developing otocyst using a glass micropipette (GD-1.2: Narishige). Embryonic stages were selected to target sonoporation ventrally (pCAGGS-Gfp and/or pCAGGS-Bmp4; HH stages 9-11) or dorsally (pCAGGS-Gfp and/or pCAGGS-Nog; HH stages 15-16). Injected embryos were immediately exposed to ultrasound using a 3-mm diameter ultrasound probe (Sonitron 2000N, Nepagene) with an input frequency of 1 MHz, an output intensity of 2.0 W/cm2, and a pulse duty ratio of 20% for a duration of 60 seconds. Embryos were collected after an additional 12 or 24 hours of incubation.

Expression vectors

Expression vectors were obtained from several sources. pCAGGS-Gfp was kindly provided by Dr. H. Ogawa (Ogawa et al. 1995). pCAGGS-caGli3A and pCAGGS-Gli3R were kindly provided by Dr. J. Briscoe (Persson et al. 2002, Stamataki et al. 2005). pCAGGS-Bmp4 was described by Ohta and coworkers (Ohta et al. 2010). Full-length mouse Nog cDNA was inserted into a pCAGGS vector using an EcoRI site. pCS2-Smad6 was purchased from Addgene (Addgene plasmid no. 14960). pT2KCAGGS-caSmoA1-Gfp was kindly provided by Dr. T. Suzuki.

Immunolabeling and Quantification

To detect GFP expression in transfected cells and pCREB activity in otocysts, embryos were fixed with 4% paraformaldehyde and embedded in paraffin. Serial transverse sections were cut using standard procedures. To identify transfected cells in tissue sections, we used a primary antibody that detects GFP (mouse monoclonal, Roche, #11814460001). As a proxy for PKA activity, we assessed pCREB labeling using a primary antibody that detects phospho-CREB (rabbit polyclonal, MILLIPORE, cat. # 06-519) diluted 1:1000 in blocking solution (0.1% FBS).

To quantify pCREB signal intensity in tissue sections, we used an ImageJ macro kindly developed by Dr. H. Otsuna, VoxelPress, LLC. The signal to noise (background) labeling of pCREB was variable from embryo to embryo, even in groups of the same type of embryo at the same stage and processed identically in the same batch. Hence, to ensure that accurate measurements of pCREB labeling could be made independently of observer bias, we used this macro to automatically normalize the signal intensity of pCREB labeling among sections of embryos based on the levels of background labeling.

To identify the subcellular distribution of the PKA catalytic subunit (PKAC) and the density of apical primary cilia in otocysts, embryos were fixed as described above and transversely cryosectioned. To detect the catalytic subunit, we used a primary antibody that recognizes PKAC (mouse monoclonal, BD transduction Laboratories, cat. # 610980), diluted 1:500 in blocking solution, followed by labeling of the primary antibody with the secondary antibody Alexa Fluor 488 (Thermo Scientific, cat. #A21202), diluted 1:1000. The primary antibody was raised against the human PKA catalytic subunit, which is 95% conserved between human, mouse, and chick in the portion of the protein (amino acids 18 to 347 used as the immunogen to generate the antibody (according to the TDA and a BLAST search). It had been shown previously by others that this antibody labels PKAC in the apical epithelium of the mouse neural tube, just beneath (basal to) the centrosome (Tucon et al. 2011). To test whether the antibody also detected chick protein, we did a pilot study on cryosections of the chick neural tube (Fig. 1). As expected, the antibody labeled the chick neural tube in a highly restricted subapical pattern, reminiscent of that in the mouse, strongly suggesting its specificity for both mouse and chick PKAC.

Figure 1.

Figure 1

Open in a new tab

Cross section through the neural tube of a stage 24-25 chick embryo shows labeling of PKAC along the dorsolateral apex of the neuroepithelium (white line), adjacent to the neural tube lumen. The section was labeled with α-PKAC (green) and counterstained with the nuclear marker TOPRO3 (blue). D: dorsal, L: lateral.

To map the distribution of otocyst apical primary cilia, we used a primary antibody that detects an ADP-ribosylation factor-like 13B protein found in primary cilia, Arl13b (rabbit polyclonal, Proteintech, cat. # 17711-1-AP), diluted 1:500 in blocking solution, followed by labeling of the primary antibody with the secondary antibody Alexa Fluor 594 (Thermo Scientific, cat. # A11012), diluted 1:1000.

After co-labeling with both sets of antibodies, we photographed sections using brightfield optics, which resulted in a pseudo-phase contrast effect that partially defined cell boundaries. We then removed the cover slips from the slides, labeled sections with α-E-cadherin to further enhance cell boundaries (mouse monoclonal mouse IgG2a, BD Bioscience, cat. #610181; Ohta et al. 2010), and superimposed images. The outlines of individual cells were then traced for quantifying the subcellular distribution of antibody labeling using a second macro to do this, also developed by Dr. H. Otsuna, VoxelPress, LLC, quantifying the signal intensity of PKAC labeling along the apicobasal extent of selected otocyst cells. At any one apicobasal level, the macro averaged the labeling intensity across the full width of the cell at that level; the mean intensities were then plotted.

Inhibition of SHH signaling

Hybridoma cells secreting either a SHH-blocking antibody (5E1, provided by Developmental Studies Hybridoma Bank [DSHB], Iowa City, Iowa) or a control antibody (QCPN, an irrelevant, α-quail nucleus antibody, provided by DSHB, Iowa), were implanted in ovo beneath the otocyst in chick embryos at HH stages 9-11. Twelve hours later, embryos were collected and processed for cryosectioning. In some experiments, sections were also labeled for PKAC.

cAMP assay

Intact dissected otocysts (from chick embryos at HH stages 15-16, collected 12 hours after either implantation of control or α-SHH hybridoma cells or sonoporation with Bmp4) were pooled (10 for each sample) and placed on ice until homogenized. 0.1M HCl (5× the volume of each sample) was added and tissues were homogenized with a polytron-type homogenizer, followed by centrifugation at top speed for 20 min at 4°C; supernatants were then collected. cAMP was assayed using a cAMP Direct Immunoassay kit (Abcam # ab65355), and the concentration of cAMP was measured using a Bio-Rad Benchmark Plus microplate reader.

Statistical analysis

Quantitative data was obtained from experiments repeated multiple times (n values are given in figure legends). Means and standard derivations were calculated and plotted, with P-values obtained using the Student's t-test.

Results

The intracellular labeling pattern of PKAC differs in the dorsal and ventral otocyst, consistent with differences in the levels of BMP and SHH signaling expected to impinge on these regions

A recent study using cerebellar granule neuron precursors (CGNPs) reported that the subcellular distribution of the PKA catalytic subunit (PKAC) is regulated by SHH signaling (Barzi et al. 2010). When CGNPs were cultured in the presence of SHH, PKAC was localized at the base of the primary cilium; upon withdrawal of SHH, this localization was lost. Our previous results demonstrated that a dorsal-high to ventral-low gradient of PKA activity is present in the otocyst during its dorsoventral patterning, and that BMP signaling increased PKA activity and SHH decreased it (Ohta et al. 2016a, b). Because of the result with CGNPs in the presence and absence of SHH, we hypothesized that the subcellular distribution of the PKA catalytic subunit (PKAC) would differ across the apicobasal extent of wall of the otocyst in dorsal versus ventral regions, which receive low and high levels, respectively, of SHH signaling, and reciprocal levels of BMP signaling.

To test this hypothesis, we labeled otocyst sections with an antibody that detects the catalytic subunit of PKA. Our pilot study on chick neural tube sections, showed that this antibody readily detected chick protein in the expected location (Fig. 1). In the ventromedial otocyst, PKAC labeling was tightly restricted to the apical side of the otic epithelium (Fig. 2). Co-labeling sections with α-Arl13b, an antibody that detects primary cilia (and, which we found also artifactually labels cell nuclei in chick sections), showed that the apical surface of the ventromedial otocyst was rich in primary cilia. In the dorsolateral otocyst, PKAC labeling was more dispersed apicobasally, extending throughout the apicobasal extent of the epithelium to the basal side of the otic epithelium, and primary cilia were only sparsely represented on the apical surface of the dorsolateral otocyst. Moreover, the overall amount of PKAC labeling seemed to be substantially greater in the dorsolateral otocyst than in the ventromedial otocyst (Fig. 2).

Figure 2.

Figure 2

Open in a new tab

In the ventromedial otocyst (the region most strongly impacted by SHH signaling, and least impacted by BMP signaling), PKAC labeling is restricted to the apical side of the epithelium, but in the dorsolateral otocyst (the region least impacted by SHH signaling, and most impacted by BMP signaling), PKAC labeling is distributed more basally. PKAC and Arl13b immunolabeling (green, blue, respectively) of a transverse section of a chick HH stage 16 otocyst comparing the subcellular distribution of PKAC and the apical localization of primary cilia (pc) in the ventromedial (vm) and dorsolateral (dl) otocyst. n=4. Other abbreviations, BF: bright field illumination, D: dorsal, L: lateral. The graphs show the gray scale values of the mean PKAC signal intensity (scale range = 0-255) within the entire area of each cell outlined in white (minus the region occupied by the nucleus) measured along their apicobasal axis (colored numbers correspond to same colors plotted in the graph). GSV, gray-scale values.

Thus, a dorsal apicobasally-dispersed to ventral apically-restricted gradient of PKAC labeling was present in the otocyst, along with a sparse-dorsal to dense-ventral gradient of apical primary cilia. These results suggest that SHH and/or BMP signaling might regulate the subcellular distribution of PKAC in the otocyst, and that such regulation might be controlled in part by SHH signaling, which is believed to be mediated perhaps exclusively through primary cilia (reviewed by Goetz & Anderson 2010).

BMP signaling increases PKA activity, inhibiting ventral development of the otocyst

We showed previously (Ohta et al. 2016b) that overexpression of BMP in the head mesenchyme lateral to the otocyst inhibited expression of some ventral otocyst transcription factors, induced expression of some dorsal transcription factors in the ventral otocyst, and inhibited development of the cochlear duct. Moreover, as assessed with an enzyme activity assay in whole otocysts, PKA activity was increased.

If BMP signaling plays an essential role in regulating PKA activity, and if the level of PKA activity is important for otocyst patterning and development, we would expect that overexpression of BMP would increase PKA activity specifically in the ventral otocyst where BMP signaling is normally weak. To test whether this occurred, we overexpressed BMP by sonoporating Bmp4 into the head mesenchyme lateral to the otocyst. Embryos were collected 12 and 24 hours after Bmp was overexpressed, otocysts were sectioned, an antibody was used to label pCREB expression (a proxy for detecting PKA activity in situ), and labeling was quantified using a macro in ImageJ (Fig. 3A). In sections of otocysts collected both at 12 and 24 hours after Bmp was overexpressed, pCREB labeling was significantly increased in the ventral otocyst, supporting our expectation. As a control, we transfected Gfp alone; doing so had no effect on the level of pCREB labeling.

Figure 3.

Figure 3

Open in a new tab

BMP signaling inhibits ventral development of the otocyst by increasing PKA activity. (A) Graph showing the mean signal intensities of pCREB labeling in chick otocysts collected 12 and 24 hours after overexpression (by head mesenchyme sonoporation) of Bmp4, compared with controls (Con) (n=18, 12 h; n=13, 24 h); or overexpression (by ventral otocyst electroporation) of Gli3R and Gfp, compared with controls (Con; Gfp-electroporated only) (n=5, 12 h; n=4, 24 h). pCREB labeling was assessed in the entire ventral otocyst of sonoporated embryos, and only in transfected cells (expressing GFP) in electroporatedotocysts (B) Graph showing the mean signal intensities of pCREB labeling in chick otocysts collected 12 and 24 hours after overexpression (by head mesenchyme sonoporation) of Noggin (Nog), compared with controls (Con) (n=4, 12 h; n=6, 24 h); or overexpression (by dorsal otocyst electroporation) of Smad6 and Gfp, compared with controls (Con; Gfp-electroporated only) (n=5, 12 h; n=7, 24 h). pCREB labeling was assessed in the entire dorsal otocyst of sonoporated embryos, and only in transfected cells (expressing GFP) in electroporated otocysts. Single asterisk: P <0.05, double asterisk: P <0.01, no asterisk: not statistically significant.

We showed previously that BMP regulates the expression of regional transcription factors in the dorsal otocyst by signaling through two intracellular pathways: the canonical pathway (pSMAD), and a non-canonical pathway (PKA-high GLI3R/GLI3A ratio) (Ohta et al. 2016b). In addition, we showed previously that overexpression of either Noggin or Smad6 inhibited gene expression and dorsal development of the otocyst (Ohta et al. 2010; Ohta et al. 2016b). Thus, we next conducted a series of experiments to ask whether inhibition of BMP signaling reduced pCREB expression, and if so, which BMP pathway might be involved.

First, we decreased BMP signaling by sonoporating Noggin, an extracellularly secreted, natural inhibitor of both canonical and non-canonical BMP signaling, into the head mesenchyme lateral to the otocyst. Consistent with the result obtained by overexpression of Bmp4, pCREB labeling was significantly reduced in the dorsal otocyst at both 12 and 24 h after overexpression (Fig. 3B), demonstrating that BMP signaling regulates PKA activity.

Next, we decreased BMP signaling specifically through the canonical pathway by electroporating Smad6, an endogenous inhibitor of this pathway. Based on our previous result (Ohta et al. 2016b), we expected that inhibition of the canonical BMP pathway—the pSMAD—pathway would have no effect on pCREB labeling. To test this, we overexpressed Smad6, an intracellular inhibitory SMAD, in the dorsal otocyst by electroporation. pCREB labeling was unchanged at both 12 and 24 hours after electroporation (Fig. 3B), demonstrating that PKA activity is regulated independently of pSMAD signaling, as expected.

Finally, we asked whether overexpression of Gli3R regulated PKA activity. Previously, we showed roles for Gli3R and Gli3A in regulating otocyst patterning (Ohta et al. 2016a, b), which led to our model (Ohta & Schoenwolf 2018). In this model, Gli3R is downstream of PKA activity/pCREB labeling, and, therefore, would not be expected to alter pCREB labeling. To test this expectation, we electroporated Gli3R into the ventral otocyst and examined pCREB labeling in sections 12 and 24 hours later. Consistent with this expectation, pCREB labeling was unchanged in transfected cells at both time points (Fig. 3A), strongly suggesting that PKA activity occurs upstream of changes in the GLI3R/GLI3A ratio.

SHH signaling decreases PKA activity, inhibiting dorsal development of the otocyst

We showed previously that overexpression of SHH in the head mesenchyme lateral to the otocyst inhibited expression of some dorsal otocyst transcription factors, induced expression of some ventral transcription factors, and inhibited development of the semicircular canals (Ohta et al. 2016a). During SHH signaling in Drosophila development, PKA activity is downregulated in the presence of SHH ligand (Chen et al. 2017). Similarly, in vertebrate development, it is generally believed that the presence of SHH ligand also downregulates PKA activity (Barzi et al. 2001, Trousse et al. 2001, Tuson et al. 2010). Consistent with this we showed recently that the ventral otocyst, the region likely to be exposed to the highest dose of SHH due to its proximity to the notochord and floor plate of the neural tube, has low PKA activity (Ohta et al. 2016a). Moreover, overexpression of Shh by sonoporating the head mesenchyme lateral to the otocyst, reduced PKA activity in the dorsal otocyst where PKA activity is normally high, as assessed both by using an enzyme activity assay in whole otocysts 24 hours after sonoporation, and by labeling otocyst sections for pCREB expression, which were then quantified (Ohta et al. 2016a). pCREB labeling was significantly decreased in the dorsal otocyst, as compared to control embryos transfected with Gfp alone, at both 12 and 24 hours after Shh overexpression (Fig. 2c' in Ohta et al., 2016a).

Here, we further examined the role of SHH signaling in otocyst DV patterning. First, we overexpressed Smoothened, a seven transmembrane G protein coupled receptor that is believed to be an obligatory member of the hedgehog signaling pathway (Corbit et al. 2005), in the dorsal otocyst by electroporating its constitutively active form, caSmo. pCREB labeling was significantly decreased at both 12 and 24 hours, demonstrating that activated SMO acts like SHH ligand by inhibiting PKA activity (Fig. 4). Next, we overexpressed Gli3A in the dorsal otocyst. Like Gli3R, Gli3A is downstream of PKA activity/pCREB labeling in our model (Ohta & Schoenwolf 2018). As expected, overexpression of Gli3A did not significantly affect pCREB labeling 12 hours later. Thus, this result, like that of overexpression of Gli3R described above, also strongly suggests that PKA activity occurs upstream of changes in the GLI3R/GLI3A ratio. Nevertheless, at 24 hours later, pCREB labeling was significantly reduced (Fig. 4). This result raises the unexpected possibility that over the longer term, GLI3A may be capable of regulating PKA activity in some indirect manner, such as through a feedback loop.

Figure 4.

Figure 4

Open in a new tab

SHH signaling inhibits dorsal development of the otocyst by decreasing PKA activity, as assessed by pCREB signal intensity. Graph showing mean signal intensities of pCREB labeling in chick otocysts collected 12 and 24 hours after overexpression (by dorsal otocyst electroporation) of constitutively active Smoothened (caSmo) (n=5, 12 h; n=5, 24 h) or Gli3A (both with Gfp) (n=4, 12 h; n=10, 24 h), compared with controls (Con; Gfp-electroporated only). pCREB labeling was assessed only in transfected cells (expressing GFP) in control (Gfp electroporated) and in caSmo electroporatedand Gli3A-electroporated otocysts. Single asterisk: P <0.05, double asterisk: P <0.01, no asterisk: not statistically significant.

BMP signaling increases cAMP levels and broadens the apicobasal distribution of PKAC in the wall of the otocyst

It is well established that activation or inactivation of PKA depends on the intracellular level of cAMP (Taylor et al. 1990). Thus, we next asked whether overexpression of BMP increases cAMP levels in the otocyst. Bmp4 was overexpressed by sonoporating the head mesenchyme lateral to the otocyst, and cAMP levels were assayed in whole chick otocysts 12 hours later. cAMP levels increased about 1.7-fold in treated otocysts, as compared with control otocysts (Fig. 5A). Thus, BMP signaling increases the intracellular level of cAMP.

Figure 5.

Figure 5

Open in a new tab

BMP signaling increases cAMP levels and results in a broad apicobasal distribution of PKAC in the wall of the otocyst. (A) Mean relative levels of cAMP in whole chick otocysts (HH stages 14-16) collected 12 hours after overexpression of Bmp4 (by head mesenchyme sonoporation), as compared with controls (Con) (n=2). Single asterisk: P <0.05. (B) PKAC immunolabeling (green) of transverse sections of chick ventromedial otocysts after overexpression of Bmp4 (by head mesenchymne sonoporation), as compared with controls. The graphs show the gray scale values of the mean PKAC signal intensity (scale range = 0- 255) within the entire area of each cell outlined in white (minus the region occupied by the nucleus) measured along their apicobasal axis (colored numbers correspond to same colors plotted in the graph).GSV, gray-scale values.

As described above, the subcellular distribution of the PKA catalytic subunit (PKAC) is regulated in CGNPs by SHH signaling (Barzi et al. 2010), and we showed also above that the subcellular distribution of PKAC differs in dorsal and ventral regions of the otocyst, consistent with areas receiving high (ventral) and low (dorsal) levels of SHH signaling, as well as reciprocal levels of BMP signaling. First, we asked whether BMP can regulate PKAC distribution in the otocyst. To do so, we overexpressed Bmp4 by sonoporating the head mesenchyme lateral to the otocyst and assessing PKAC labeling in the ventromedial otocyst 12 hours later. After overexpression, PKAC was more broadly distributed from the apical side of the epithelium extending more basally, spanning on average of about 50% of the apicobasal extent of the epithelium, as compared with controls in which it spanned less than 25% of this extent (Fig. 5B). Thus, BMP signaling can regulate the subcellular distribution of PKAC in the early otocyst.

SHH signaling decreases cAMP levels and restricts the apical distribution of PKAC in the wall of the otocyst

We next asked whether cAMP levels in the developing otocyst are also regulated by SHH signaling. To do this, we blocked SHH signaling in the ventral otocyst by implanting hybridoma cells secreting a SHH-neutralizing antibody (α-SHH) beneath the otic placode in chick embryos at HH stages 9-11. Twelve hours later, cAMP levels were assayed in whole otocysts and compared to control otocysts (containing implanted hybridoma cells secreting QCPN, an α-quail antibody serving as a control antibody). cAMP levels increased about 1.8-fold in α-SHH-treated otocysts, as compared with control otocysts (Fig. 6A), suggesting that SHH signaling normally decreases the production of cAMP in the ventral otocyst.

Figure 6.

Figure 6

Open in a new tab

SHH signaling decreases cAMP levels and results in a restricted apical distribution of PKAC in the wall of the otocyst, as shown by the inhibition of SHH protein. (A) Mean relative levels of cAMP in whole chick otocysts (HH stages 14-16) collected 12 hours after transplantation of α-SHH-secreting hybridoma cells beneath the otic placode, as compared with controls (Con; from otocysts in which hybridoma cells secreting QCPN—an α-quail nucleus antibody—were transplanted instead) (n=3). Double asterisk: P <0.01. (B) PKAC immunolabeling (green) of transverse sections of chick ventromedial otocysts collected 12 hours after transplantation of α-SHH-secreting hybridoma cells beneath the otic placode, as compared with controls. The graphs show the gray scale values of the mean PKAC signal intensity (scale range = 0-255) within the entire area of each cell outlined in white (minus the region occupied by the nucleus) measured along their apicobasal axis (colored numbers correspond to same colors plotted in the graph). h: hybridoma cells (fluorescent because the α-mouse IgG secondary antibody used in this experiment binds to the SHHblocking antibody). GSV, gray-scale values.

To ask whether SHH is required to maintain the restricted apical distribution of PKAC in the ventromedial otocyst, we next examined PKAC labeling 12 hours after hybridoma cells secreting α-SHH or a control (QCPN) antibody were implanted beneath the otic placode of chick embryos at stages 9-11. After blocking SHH signaling, PKAC was distributed more basally (i.e., away from the location of the primary cilia), as compared with that of control ventromedial otocysts in which the distribution of PKAC was restricted to the apical region (Fig. 6B). Thus, SHH signaling is required for maintaining the restricted apical distribution of PKAC labeling in the otocyst.

Collectively, the results of these experiments in which BMP signaling was increased and SHH signaling was decreased reveal that both factors can regulate cAMP levels in the otocyst (with BMP increasing levels and SHH decreasing levels). Moreover, both factors can regulate the subcellular distribution of PKAC in a manner sufficient to explain the DV gradient of PKAC labeling normally present in the early otocyst. Finally, after increasing BMP signaling or decreasing SHH signaling, the overall amount of PKAC labeling in the wall of the otocyst seemed to be substantially greater than that of non-treated embryos (Fig. 6B).

Discussion

Our results provide strong evidence that both BMP and SHH signaling regulate PKA activity in the otocyst during its DV patterning and do so by regulating cAMP levels, which, in turn, results in regional changes in the subcellular distribution of the catalytic domain of PKA and DV regulation of PKA activity, increasing it dorsally and decreasing it ventrally. Thus, our results suggest the following model (Fig. 7). BMP signaling, normally high in the dorsal otocyst and low in the ventral otocyst, increases cAMP levels in the dorsal otocyst, which in turn releases the catalytic domains of PKA from its regulatory domains, thereby increasing PKA activity in the dorsal otocyst. This increased PKA activity catalyzes the post-translational, partial proteolytic processing of GLI3FL, increasing the GLI3R/GLI3A ratio in the dorsal otocyst and resulting in the expression of dorsal otocyst transcription factors (for instance, Nkx5.1/Hmx3 and Dlx5) and development of dorsal otocyst structures such as the semicircular canals. Conversely, SHH signaling, normally high in the ventral otocyst and low in the dorsal otocyst, decreases cAMP levels in the ventral otocyst, which in turn prevents the release of the catalytic domains of PKA from it regulatory domains, maintaining low PKA activity in the ventral otocyst. With low PKA activity in the ventral otocyst, partial proteolytic processing of GLI3FL fails to occur, and the GLI3R/GLI3A ratio in the ventral otocyst remains low, resulting in the expression of ventral otocyst transcription factors (for instance, Pax2 and Otx2) and development of ventral otocyst structures such as the cochlea. Below, we discuss the new elements of this model added by our current study.

Figure 7.

Figure 7

Open in a new tab

Model integrating BMP and SHH signaling pathways to establish dorsoventral polarization of the otocyst. During formation of the inner ear from the otocyst, BMP acts on the dorsolateral otocyst to increase the GLI3R/GLI3A ratio, resulting in the expression of dorsal otocyst genes (e.g., Dlx5, Nkx5.1/Hmx3) and the formation of the vestibule, and SHH acts on the ventrolateral otocyst to decrease the GLI3R/GLI3A ratio, resulting in the expression of ventral otocyst genes (e.g., Pax2, Otx2) and the formation of the cochlear duct. Red and blue ovals on the otocyst indicate enlargements of dorsolateral and ventromedial otocyst cells, respectively, included to show the intracellular pathways underlying BMP and SHH signaling. c: PKA catalytic domain, r: PKA regulatory domain, AC, adenylate cyclase. See Discussion for further details.

BMP and SHH signaling differentially regulate cAMP levels in the dorsal and ventral otocyst

Our current results provide strong evidence that BMP and SHH signaling differentially and reciprocally regulate cAMP levels in the early otocyst, with BMP increasing cAMP levels and SHH decreasing levels. Overexpression of Bmp4 significantly increased cAMP levels in the otocyst (Fig. 5A). Consistent with this, blocking SHH signaling with hybridoma cells secreting α-SHH also significantly increased cAMP levels (Fig. 6A), suggesting that the normal role for SHH signaling in the otocyst is to decrease cAMP levels. Thus, both BMP and SHH signaling regulate cAMP levels in opposite ways and both are sufficient and necessary (along with other ligands such as WNTs and FGFs) for normal DV patterning of the otocyst (Riccomagno et al. 2002, 2005, Bok et al. 2005, 2007, Ohyama et al. 2006, Hatch et al. 2007, Brown & Epstein 2011, Ohta et al. 2010, 2016a, b).

Although the precise role of BMP signaling in regulating cAMP levels is unclear, SHH signaling has been shown to decrease cAMP levels in at least one previous study: when retinal ganglion axons were treated with SHH, cAMP levels were dramatically reduced (Trousse et al. 2001), a result consistent with our current results. Thus, our current study identifies another component of upstream intracellular events involved in the signaling cascades that lead to ligand-dependent DV patterning of the otocyst. This is important because activation or inactivation of PKA is known to depend on the intracellular level of cAMP (Taylor et al. 1990), and our previous work has shown an important role for PKA activity in otocyst DV patterning (Ohta et al. 2016a, b; reviewed by Ohta & Schoenwolf 2018).

BMP and SHH signaling differentially regulate PKA activity in the dorsal and ventral otocyst

Our current results also provide strong evidence that BMP and SHH signaling differentially and reciprocally regulate PKA activity in the early otocyst, with BMP increasing PKA activity and SHH decreasing it. First, manipulating BMP signaling altered PKA activity as assessed by using pCREB labeling, a direct readout of PKA activity. Overexpression of Bmp4 significantly increased pCREB labeling in the otocyst at both 12 and 24 hours after overexpression (Fig. 3A). in contrast, blocking BMP signaling, by overexpressing Noggin, significantly decreased pCREB labeling (Fig. 3B). Patterning of the dorsal otocyst through BMP signaling involves two intracellular pathways: a pSMAD-dependent pathway and a pSMAD-independent pathway (Ohta et al. 2016b). Because Noggin inhibits the binding of BMP ligands to BMP receptors at the cell surface, it blocks both pathways. In contrast, Smad6 overexpression blocks only the pSMAD (intracellular) pathway. Blocking the pSMAD pathway had no effect on pCREB levels (Fig. 3B), suggesting that only the pSMAD-independent pathway depends upon PKA activity.

Second, as shown previously, manipulating SHH signaling had opposite effects to those obtained by manipulating BMP signaling. Overexpression of Shh significantly decreased pCREB labeling in the otocyst at both 12 and 24 hours after overexpression (Ohta et al. 2016a). Moreover, increasing SHH signaling by overexpressing caSmo also significantly decreased pCREB labeling in the otocyst at both 12 and 24 hours after overexpression (Fig. 4).

Third, the intracellular distribution of PKAC differed dramatically in the dorsolateral otocyst (the portion of the otocyst expected to receive the highest level of BMP signaling and the lowest level of SHH signaling) and the ventromedial otocyst (the portion of the otocyst expected to receive the highest level of SHH signaling and the lowest level of BMP signaling). Specifically, dorsolaterally, PKAC is distributed throughout most of the apicobasal extent of the otocyst epithelium, whereas ventromedially PKAC is restricted to the apical side of the epithelium in close proximity to copious apical primary cilia (Fig. 2), essential organelles for SHH signaling (reviewed by Goetz & Anderson 2010). In cerebellar granule neuron precursors, the subcellular distribution of the PKA catalytic subunit (PKAC) is regulated by SHH signaling; namely, accumulating at the basal body of the primary cilium in the presence of SHH ligand and losing this localization in its absence (Barzi et al. 2010). Thus, our results on the localization of PKAC in the otocyst provide further evidence of increased PKA activity in the dorsal otocyst, which receives high levels of BMP signaling and low levels of SHH signaling, and decreased PKA activity in the ventral otocyst, which receives high levels of SHH signaling and low levels of BMP signaling.

Fourth, overexpressing Bmp4 caused a redistribution of PKAC in the otocyst, mimicking the pattern normally found in the dorsal otocyst (Fig. 5B). This redistribution indicates increased PKA activation as a result of increased BMP signaling.

Fifth, inhibiting SHH signaling by transplanting hybridoma cells secreting α-SHH also caused a redistribution of PKAC in the otocyst, mimicking the pattern normally found in the dorsal otocyst (Fig. 6B). This redistribution indicates increased PKA activation as a result of decreased SHH signaling. Thus, PKA activity is regulated regionally in the otocyst by both BMP and SHH signaling, a finding consistent with our earlier result based on whole otocyst PKA enzyme activity assays.

Finally, increasing BMP signaling (Fig. 5B) or decreasing SHH signaling (Fig. 6B) seems to increase the amount of PKAC labeling present in the otocyst wall, mimicking the normal difference between the dorsolateral and ventromedial walls of the untreated otocyst (Fig. 2). Why altering BMP or SHH signaling would have this effect is a puzzle. Perhaps, BMP and SHH signaling play important roles in PKA synthesis and/or degradation. Alternatively, perhaps when PKAC is apically sequestered and bound to its regulatory domain, it is under-labeled by the antibody because of spatial/antibody-access constraints, which are relieved upon unbinding and basal spreading away from the basal body. Further studies are necessary to solve this interesting puzzle.

PKA activity patterns the dorsal and ventral otocyst, and PKA activity is not merely the result of such patterning

Our current results, based on regional changes in pCREB labeling in the dorsal and ventral otocyst after manipulation of BMP and SHH signaling, as well as changes in the distribution of PKAC after these manipulations, strongly support the following model (Fig. 7). In the present of high levels of BMP signaling (and low levels of SHH signaling) impacting the dorsal otocyst, the intracellular level of cAMP is increased in the dorsal otocyst, resulting in increased PKA activity. In the presence of high levels of SHH signaling (and low levels of BMP signaling) impacting the ventral otocyst, the intracellular level of cAMP is decreased in the ventral otocyst, resulting in decreased PKA activity. As shown by our previous work (Ohta et al. 2016a, b; reviewed by Ohta & Schoenwolf 2018), this establishes a DV gradient in PKA activity across the otocyst that results in DV-specific patterning of the otocyst: namely, 1) enhanced post-translational proteolytic processing of the full-length form of GLI3 dorsally to establish high dorsal levels of GLI3R (repressor form); and 2) maintained high levels of GLI3A (activator form) ventrally. The resulting high-dorsal to low-ventral gradient of the GLI3A/GLI3R ratio induces regionally-characteristic patterns of transcription factor expression that establish otocyst DV patterning, regional morphogenesis, and formation of distinct dorsal and ventral sensory organs.

Although extensive epistasis-type experiments have not been conducted to definitely determine the sequence of steps in this cascade of signaling events, our experiments in which Gli3R and Gli3A were overexpressed in the otocyst (ventrally and dorsally, respectively) provide preliminary support for the proposed sequence of events. In both sets of experiments, no change in pCREB staining was detected 12 hours after overexpression (Figs. 3A, 4), providing compelling evidence that PKA activity is upstream of GLI3 activity and not downstream of it. Further experiments will be required to examine the exact positioning of cAMP activity in the sequence, but it acting upstream of PKA activity is most consistent with our understanding from previous work (e.g., Li et al. 2017).

Acknowledgments

The research reported herein was supported by the National Institute on Deafness and Other Communication Disorders of the National Institutes of Health (R01 DC011819). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

References

  1. Barzi M, Berenguer J, Menendez A, Alvarez-Rodriguez R, Pons S. Sonic-hedgehog-mediated proliferation requires the localization of PKA to the cilium base. J Cell Sci. 2010;123:62–69. doi: 10.1242/jcs.060020. [DOI] [PubMed] [Google Scholar]
  2. Basch ML, Brown RM, 2nd, Jen HI, Groves AK. Where hearing starts: the development of the mammalian cochlea. J Anat. 2016;228:233–254. doi: 10.1111/joa.12314. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bok J, Bronner-Fraser M, Wu DK. Role of the hindbrain in dorsoventral but not anteroposterior axial specification of the inner ear. Development. 2005;132:2115–2124. doi: 10.1242/dev.01796. [DOI] [PubMed] [Google Scholar]
  4. Bok J, Dolson DK, Hill P, Ruther U, Epstein DJ, Wu DK. Opposing gradients of Gli repressor and activators mediate Shh signaling along the dorsoventral axis of the inner ear. Development. 2007;134:1713–1722. doi: 10.1242/dev.000760. [DOI] [PubMed] [Google Scholar]
  5. Brown AS, Epstein DJ. Otic ablation of smoothened reveals direct and indirect requirements for Hedgehog signaling in inner ear development. Development. 2011;138:3967–3976. doi: 10.1242/dev.066126. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Chen P, Zhou Z, Yao X, et al. Capping enzyme mRNA-cap/RNGTT regulates hedgehog pathway activity by antagonizing Protein Kinase A. Sci Rep. 2017;7:2891. doi: 10.1038/s41598-017-03165-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Goetz SC, Anderson KV. The primary cilium: a signalling centre during vertebrate development. Nat Rev Genet. 2010;11:331–344. doi: 10.1038/nrg2774. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Groves AK, Fekete DM. Shaping sound in space: the regulation of inner ear patterning. Development. 2012;139:245–257. doi: 10.1242/dev.067074. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Hamburger V, Hamilton HL. A series of normal stages in the development of the chick embryo. J Morphol. 1951;88:49–92. [PubMed] [Google Scholar]
  10. Hatch EP, Noyes CA, Wang X, Wright TJ, Mansour SL. Fgf3 is required for dorsal patterning and morphogenesis of the inner ear epithelium. Development. 2007;134:3615–3625. doi: 10.1242/dev.006627. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Li J, Wang C, Wu C, et al. PKA-mediated Gli2 and Gli3 phosphorylation is inhibited by Hedgehog signaling in cilia and reduced in Talpid3 mutant. Dev Biol. 2017;429:147–157. doi: 10.1016/j.ydbio.2017.06.035. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Ogawa H, Inouye S, Tsuji FI, Yasuda K, Umesono K. Localization, trafficking, and temperature-dependence of the Aequorea green fluorescent protein in cultured vertebrate cells. Proc Natl Acad Sci U S A. 1995;92:11899–11903. doi: 10.1073/pnas.92.25.11899. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Ohta S, Mansour SL, Schoenwolf GC. BMP/SMAD signaling regulates the cell behaviors that drive the initial dorsal-specific regional morphogenesis of the otocyst. Dev Biol. 2010;347:369–381. doi: 10.1016/j.ydbio.2010.09.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Ohta S, Schoenwolf GC. Hearing crosstalk: the molecular conversation orchestrating inner ear dorsoventral patterning. Wiley Interdiscip Rev Dev Biol. 2018;7 doi: 10.1002/wdev.302. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Ohta S, Wang B, Mansour SL, Schoenwolf GC. SHH ventralizes the otocyst by maintaining basal PKA activity and regulating GLI3 signaling. Dev Biol. 2016a;420:100–109. doi: 10.1016/j.ydbio.2016.10.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Ohta S, Wang B, Mansour SL, Schoenwolf GC. BMP regulates regional gene expression in the dorsal otocyst through canonical and non-canonical intracellular pathways. Development. 2016b;143:2228–2237. doi: 10.1242/dev.137133. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Ohyama T, Mohamed OA, Taketo MM, Dufort D, Groves AK. Wnt signals mediate a fate decision between otic placode and epidermis. Development. 2006;133:865–875. doi: 10.1242/dev.02271. [DOI] [PubMed] [Google Scholar]
  18. Persson M, Stamataki D, te Welscher P, Andersson E, Bose J, Ruther U, Ericson J, Briscoe J. Dorsal-ventral patterning of the spinal cord requires Gli3 transcriptional repressor activity. Genes Dev. 2002;16:2865–2878. doi: 10.1101/gad.243402. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Riccomagno MM, Martinu L, Mulheisen M, Wu DK, Epstein DJ. Specification of the mammalian cochlea is dependent on Sonic hedgehog. Genes Dev. 2002;16:2365–2378. doi: 10.1101/gad.1013302. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Riccomagno MM, Takada S, Epstein DJ. Wnt-dependent regulation of inner ear morphogenesis is balanced by the opposing and supporting roles of Shh. Genes Dev. 2005;19:1612–1623. doi: 10.1101/gad.1303905. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Stamataki D, Ulloa F, Tsoni SV, Mynett A, Briscoe J. A gradient of Gli activity mediates graded Sonic Hedgehog signaling in the neural tube. Genes Dev. 2005;19:626–641. doi: 10.1101/gad.325905. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Taylor SS, Buechler JA, Yonemoto W. cAMP-dependent protein kinase: framework for a diverse family of regulatory enzymes. Annu Rev Biochem. 1990;59:971–1005. doi: 10.1146/annurev.bi.59.070190.004543. [DOI] [PubMed] [Google Scholar]
  23. Trousse F, Marti E, Gruss P, Torres M, Bovolenta P. Control of retinal ganglion cell axon growth: a new role for Sonic hedgehog. Development. 2001;128:3927–3936. doi: 10.1242/dev.128.20.3927. [DOI] [PubMed] [Google Scholar]
  24. Tuson M, He M, Anderson KV. Protein kinase A acts at the basal body of the primary cilium to prevent Gli2 activation and ventralization of the mouse neural tube. Development. 2011;138:4921–4930. doi: 10.1242/dev.070805. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Whitfield TT. Development of the inner ear. Curr Opin Genet Dev. 2015;32:112–118. doi: 10.1016/j.gde.2015.02.006. [DOI] [PubMed] [Google Scholar]
  26. Wu DK, Kelley MW. Molecular mechanisms of inner ear development. Cold Spring Harb Perspect Biol. 2012;4:a008409. doi: 10.1101/cshperspect.a008409. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES