Netrin-1 and deleted in colorectal carcinoma (DCC) messenger RNA are identified by reverse transcription (RT)-polymerase chain reaction amplification from total RNA isolated from the organ of Corti and spiral ganglia of postnatal C57 mice, respectively. A total of 10 μL of each reaction product was analyzed on agarose gel, 2%, and stained with Cybergreen (Molecular Probes, Eugene, Oregon). Reaction with netrin-1 primers on the organ of Corti RNA target shows the expected 464–base pair (bp) product and with DCC primers on spiral ganglion RNA target shows expected 300-bp product (+RT). The 100-bp ladder size marker bands (M) seen in the gel represent 300, 400, and 500 bp. Each reaction set contained a no-RT negative control (−RT) toward the source of netrin-1 (P < .05).
The DCC protein was identified in mouse spiral ganglion cells. Immunohistochemical staining of DCC in cultured postnatal mouse spiral ganglion cells demonstrated localization in both the cell bodies and axons of these cells. The bar indicates 50 μm.
Netrin-1 promotes neurite extension in cultured spiral ganglion cells (CSGCs). Mouse spiral ganglion cell axons were cultured with varying concentrations of recombinant netrin-1, and neurite lengths were measured. The bars show the mean length of neurites plotted as a function of the percentage of control neurons cultured without added netrin-1. At least 50 neurons were measured per concentration of netrin-1 added. Axon lengths were significantly larger for ganglion cells cultured with netrin-1 at 100 (P = .01), 200 (P = .01), and 500 ng/mL (P = .001)
(115%, 130%, and 146%, respectively). The error bars represent the SEM.
Human embryonic kidney cells (HEK293)–netrin-1–secreting cells. A, Schematic drawing of HEK293–netrin-1–secreting cells embedded in growth factor–reduced Matrigel (Becton-Dickenson, Bedford, Massachusetts)
(arrow) cocultured with chick acoustic ganglion cells. B, Photomicrograph of HEK293–netrin-1–secreting cells embedded in growth factor–reduced Matrigel (black arrow) and cocultured with chick acoustic ganglion cells. The white arrows point to cell bodies of 2 acoustic ganglion cells with axons (arrowheads) extending toward netrin-1–secreting cells (black arrow). The bar indicates 100 μm.
Cultured spiral ganglion cells extend axons toward netrin-1. Quantitative analysis of direction of neurite extension in chick acoustic ganglion cells cocultured with human embryonic kidney cells (HEK293)–netrin-1–secreting cells. Neurites were classified as extending toward (+), away from (−), or neutral (0) relative to the source of netrin-1, based on their overall direction of growth. The means of these classifications from 4 independent neuronal preparations and coculture experiments were calculated and plotted in the graph with bars representing the mean number of neurites in each category and error bars representing the SEM. Ganglion cells in the cocultures with HEK293–netrin-1
cells showed a significant preference for extending their axons toward the source of netrin-1 (P = .04).
Customize your JAMA Network experience by selecting one or more topics from the list below.
Lee KH, Warchol ME. Promotion of Neurite Outgrowth and Axon Guidance in Spiral Ganglion Cells by Netrin-1. Arch Otolaryngol Head Neck Surg. 2008;134(2):146–151. doi:10.1001/archoto.2007.6
To identify the expression of netrin-1, a diffusible chemoattractive molecule, and its receptor, deleted in colorectal carcinoma (DCC), in the developmentally mature inner ear, and to determine its effects on axon length and guidance in cultured auditory neurons.
Messenger RNA (mRNA) and protein expression of netrin-1 and DCC were identified in the organ of Corti and spiral ganglion cells using reverse transcription–polymerase chain reaction (RT-PCR)
and fluorescence immunohistochemical analysis. In vitro experiments examined the effects of exogenous netrin-1 on spiral ganglion cell axon length. Auditory neurons were cocultured with a cell line secreting netrin-1 to determine the effect on direction of axon extension.
Young C57 mice and posthatched white leghorn chicks.
Netrin-1 and DCC mRNA expression were found in the mouse organ of Corti and spiral ganglion cells by RT-PCR. Application of exogenous netrin-1 led to a dosage-dependent increase in neurite length in cultured spiral ganglion cells. Chick acoustic ganglion cells cocultured with netrin-1–secreting cells demonstrated statistically significant preferential extension toward the source of netrin-1 (P = .04).
Netrin-1 and DCC are expressed in the organ of Corti and spiral ganglion cells of developmentally mature mice. Exogenous netrin-1 promotes dosage-dependent neurite growth in vitro. Mature auditory neurons preferentially direct neurite extension toward netrin-1 released in culture. These findings may lead to the development of strategies to optimize the interface between electrode arrays and spiral ganglion cells, resulting in improved cochlear implant performance.
Recent advances in cochlear implant hardware design have included strategies to position the electrode array more medially to hug the modiolus and bring the stimulating electrodes closer to the spiral ganglion cells. In theory, this would lower stimulation thresholds, improve battery life by decreasing power consumption, and expand the dynamic range of the stimulus.1-3 Implementing such strategies has led to improved cochlear implant function.4,5
Stimulating the outgrowth of spiral ganglion cell axons would be another means to achieve this objective. Decreasing the distance between the electrodes and their stimulus targets may provide a mechanism that would enhance frequency selectivity and dynamic range. Such benefits would not only augment speech recognition but may also improve the tonal quality of sound and potentially allow implant users to better appreciate music. However, simple stimulation of spiral ganglion neurite outgrowth may lead to random patterns of axon extension, resulting in decreased cochlear implant function. An understanding of the innate molecular mechanisms that establish and maintain the precise tonotopic organization of the inner ear may be essential to develop strategies for highly controlled neurite outgrowth necessary to achieve improved cochlear implant function.
Netrin-1 is a member of a family of secreted laminin-related proteins and has been shown to attract commissural axons to the midline of the spinal cord during development.6,7 The family of peptides known as netrins are highly conserved across several species and can act both at short range as well as long range to guide growing axons to appropriate innervation targets.8,9 Netrins are bifunctional molecules in that they can mediate either attraction or repulsion, based on their interactions with different receptor families10,11 or based on levels of second messengers such as cyclic adenosine monophosphate.12
Deleted in Colorectal Carcinoma (DCC) is a gene that has shown to be mutated in metastatic human colorectal cancer.13 This gene encodes a receptor that binds netrin-1 and confers an axon-attracting response.10 In addition to the attractive response of this interaction, netrin-1 and the DCC gene product (hereinafter, DCC) also lead to axon outgrowth mediated by second messengers.14-16
Although most of the studies of netrin-1 have focused on the central nervous system, some investigations17,18 have also demonstrated its activity in the periphery. Netrin-1 has been implicated in the guidance of retinal ganglion cells to appropriate central targets in the visual system. Additional data suggest that netrin-1 is involved in the embryogenesis of the otic epithelium8 as well as in the development of the semicircular canals.19 Recently, netrin-1 protein has been shown to be expressed in early postnatal rat cochlea.20
In the present study, we investigated the expression of DCC in the inner ear of developmentally mature mice at both the messenger RNA (mRNA) and protein levels and netrin-1 expression at the mRNA level. We also show functional data indicating that the auditory neurons of young developmentally mature animals maintain their ability to elongate and change the direction of axon extension in response to netrin-1 in vitro. Our results may be the basis for further investigation leading to new strategies to advance cochlear implant function.
All experiments were performed using tissue from C57 mice, postnatal day 28 to 35 (P28-P35) obtained from Charles River Laboratories (Wilmington, Massachusetts) and acoustic ganglia from white leghorn chickens, posthatch day 10 to 14 from Spafas (Preston, Connecticut).
Total RNA was isolated from pooled samples of spiral ganglia and organs of Corti microdissected from 5 to 6 C57 mice using Ultraspec RNA (Biotecx Laboratories, Houston, Texas) and contaminating genomic DNA was removed with RNase-free DNase I. Published sequences for netrin-1
(Serafini et al7) and DCC (Fearon et al13) in mice were used to design the following primers: 5′AGTSTGTCTCAACTGCCGCC3′ (netrin-1
sense), 5′TACACGGAGATGATGTTCACGG3′ (netrin-1 antisense), 5′AGTGCCTCTCATTCAGGTCAGG3′ (DCC sense), and 5′TCACAGACTGAGTTCTTCCTGC3′
(DCC antisense). Reverse transcription–polymerase chain reaction (RT-PCR) was performed using the RNA Gene Amp kit (Applied Biosystems, Foster City, California). Thirty-five cycles of amplification were performed with 95°C denaturing, 55°C annealing, both for 30
seconds, and 72°C extension for 60 seconds. Each experiment included a no-RT negative control. Reaction products (10 μL) were then analyzed on an agarose gel, 2%, with 100–base pair DNA ladder size standard and stained with Cybergreen (Molecular Probes, Eugene, Oregon).
Spiral ganglia were microdissected and placed in M199 medium (Sigma, St Louis, Missouri) with Hank salts and 25 mM Hepes. Each ganglion was further dissected into smaller fragments, pooled, and digested in trypsin, 25%, (Sigma) for 25 minutes at 37°C. Digested tissue isolates were resuspended in M199 with Earle salts, supplemented with 10% fetal bovine serum (Invitrogen, Carlsbad, California). Samples were then triturated in this solution to generate suspensions of individual neuronal and glial cells and plated on culture dishes precoated with laminin (Sigma). Subsequently, the cells were cultured for 48 to 120
hours with 50% fresh medium changes every 2 days for cells cultured longer than 48 hours.
Following 48 hours in culture, the neurons were fixed with 4%
paraformaldehyde for 15 minutes, incubated in 90% methanol and 0.3%
hydrogen peroxide, and blocked with 2% normal horse serum, 1% bovine serum albumin, and 0.2% Triton X-100 detergent (Sigma). The cultured cells were then incubated with mouse monoclonal anti-DCC antibody (5 μg/mL) (Oncogene Research Products, Boston, Massachusetts) diluted 1:500 in phosphate-buffered saline with 2% normal horse serum and 0.2% Triton X-100 detergent overnight at 4°C, then in biotinylated antimouse IgG (Vector Laboratories, Burlingame, California) for 2
hours at room temperature, followed by streptavidin Alexa Fluor 488
(Molecular Probes) for 2 hours at room temperature. A no–primary antibody negative control was performed with each experiment. The labeled neurons were viewed with an epifluorescence microscope (Nikon Instruments, Melville, New York).
Dissociated spiral ganglion cells were prepared as described in the “Ganglion Cell Cultures” subsection and cultured for 120 hours with 1-, 10-, 100-, 200-, or 500-ng/mL recombinant chick netrin-1 (R&D Systems, Minneapolis, Minnesota), or with no exogenous netrin-1 (negative control). The neurons were fixed and immunolabeled with the neuron-specific ßIII tubulin antibody marker TuJ-1 (2 μg/mL diluted 1:500, mouse monoclonal anti-ßIII tubulin antibody; Covance, Berkeley, California), followed by biotinylated secondary antibody and the ABC Elite Kit with diaminobenzidine detection (Vector Laboratories). These specimens were viewed with an Axiovert 135 inverted microscope (Zeiss MicroImaging, Thornwood, New York). All images were digitized with a Nikon Coolpix 990 digital camera, and neurite lengths were quantified using National Institutes of Health Image J free downloadable software.
Human embryonic kidney (HEK293) cells stably transfected with human netrin-1 were obtained as a gift from Yi Rao, PhD. The HEK293–netrin-1–secreting cells were suspended in 10 μL growth factor–reduced matrix gel (Matrigel; Becton-Dickenson, Bedford, Massachusetts). The cell suspensions were placed along the edges of 35-mm culture dishes (MatTek Corp, Ashland, Massachusetts) and warmed to 37°C to allow Matrigel polymerization. Subsequently, dissociated chick acoustic ganglion cells were prepared as described in the previous subsection and plated on laminin-coated culture dishes. The ganglion cells were cocultured with the HEK293-netrin-1 cells for 48 hours in M199 with Earl salts and 10% fetal bovine serum, and the neurons were then processed for labeling with TuJ-1, as described in the previous subsection, and analyzed for overall direction of neurite extension. Cocultures with nontransfected HEK293 were performed in parallel as controls.
Total RNA was extracted from mouse spiral ganglia and from mouse organ of Corti. Next, RT-PCR was used to amplify specific regions of netrin-1 and DCC mRNA. Analysis by agarose gel electrophoresis demonstrated the expected 464-bp netrin-1 product from organ of Corti RNA and the expected 300-bp DCC product from the spiral ganglion RNA. In addition, netrin-1 mRNA was detected in spiral ganglion cells and DCC mRNA in the organ of Corti (Figure 1). Negative-control PCR without RT showed no products, demonstrating that the bands seen reflect amplification of mRNA and not contaminating genomic DNA. These results indicate that the genes for netrin-1 and DCC are actively transcribed in postnatal mouse inner ear.
Spiral ganglion cell suspensions were prepared from postnatal mice and cultured for 48 hours. These cultures were then processed for immunohistochemical localization of DCC, a receptor that interacts with netrin-1 leading to its axon-attracting activity. Positive immunostaining of cultured spiral ganglion cells with the anti-DCC antibody was observed in both the cell bodies and axons (Figure 2). Negative controls with no primary antibody showed no DCC immunolabeling in the neurons (data not shown). This finding demonstrated that in addition to gene transcription, DCC protein is expressed and further suggests that netrin-1 may have an active role in postnatal mouse spiral ganglion cell function.
To determine the effect of netrin-1 on the neurite length of mouse spiral ganglion cells, dissociated neurons were cultured for 120 hours with recombinant chick netrin-1 at concentrations of 1, 10, 100, 200, or 500 ng/mL. The cells were then immunolabeled with TuJ-1, a neuron-specific marker. Neurite lengths were determined by tracing digitized images of the axons. These lengths were compared with those of control neurons cultured similarly, but without any addition of netrin-1. Exogenous netrin-1 led to a dosage-dependent increase (SEM) in neurite length in cultured spiral ganglion cells with lengths at 115% (4.4%), 130% (6.3%), and 146% (7.0%) of controls at 100 ng/mL (P = .01), 200 ng/mL (P = .01), and 500 ng/mL (P = .001), respectively (Figure 3). This demonstrates that netrin-1
promotes axon outgrowth in cultured postnatal mouse spiral ganglion cells.
The HEK293 cells were transfected with human netrin-1 to stably secrete this protein in culture. These HEK293–netrin-1 cells were cultured in a collagen matrix to produce a localized continuous source of netrin-1. These localized cell suspensions were cocultured with dissociated chick acoustic ganglion cells (avian analogue of mammalian spiral ganglion cells) to assess the influence of localized netrin-1 on the direction of outgrowth (Figure 4). The ganglion cells in these cocultures were analyzed for the overall direction of neurite extension after immunolabeling the cells with TuJ-1. This analysis showed a significant preference of acoustic ganglion cells for neurite extension toward a defined source of netrin-1 (P = .04; Figure 5). In contrast, control cocultures containing nontransfected HEK293 cells (negative controls) did not show this preferential direction of neurite outgrowth. This indicates that netrin-1 may play a role in axon guidance of developmentally mature auditory neurons.
Netrin-1 is a secreted protein that is well characterized for its activity in neurite lengthening and axon guidance by attraction when interacting with the receptor DCC. In our study, we have shown that netrin-1 mRNA is expressed in the organ of Corti of developmentally mature mice and that DCC is expressed both at the mRNA and protein levels in the spiral ganglion cells of these animals. In addition, we have demonstrated that exogenous netrin-1 promotes spiral ganglion cell axon extension in a dosage-dependent manner in vitro. Lastly, with coculture experiments, we have shown that acoustic ganglion cells of young chickens preferentially extend their axons toward a localized source of netrin-1.
We found that both netrin-1 and DCC are expressed in both the organ of Corti and spiral ganglion cells at the mRNA level. This indicates that, in addition to the expected findings that netrin-1 is expressed in the organ of Corti and DCC is expressed in spiral ganglion cells, the spiral ganglion neurons also express the ligand and the organ of Corti expresses the receptor. This is consistent with previous findings that netrin-1 is expressed in adult mammalian spinal cord neurons.20 Expression of DCC in organ of Corti cells suggests the possibility of an autocrine feedback mechanism in netrin-1–mediated guidance of cochlear innervation.
A previous study21 reported expression of netrin-1 in the early postnatal Wistar rat cochlea. Semiquantitative Western immunoblotting was used to show stable levels of netrin-1 peptide expression at P1 to P7, with decreasing levels at P10 and P15. At P22, neither netrin-1 nor DCC protein was detected in rat cochlear tissue. Our studies were performed in early postnatal C57 mice (P28-P35); netrin-1 mRNA was detected from total RNA isolated from organ of Corti tissue, and expression of DCC mRNA was identified in preparations isolated from spiral ganglia. We also found positive immunoreactivity for DCC localized in both the cell bodies as well as axons of cultured spiral ganglion harvested from animals of the same age.
Multiple factors could account for these differences. Species differences (rat vs mouse) could be a simple explanation for the discrepancies in results. Also, in our studies, netrin-1 expression was detected only at the mRNA level by RT-PCR, so it is possible that at later stages in the early postnatal period, netrin-1 is transcribed but not translated to mature protein. The use of different techniques and preparations of the target tissue, as well as antibodies from different sources (Oncogene, as used by Gillespie et al,21 and R&D Systems in our studies), can account for our detection of DCC expression in postnatal rodents. Finally, our functional studies demonstrating that the spiral ganglion cells respond to netrin-1 by both axon extension and directional guidance support our finding of expression of DCC in these cells.
One caveat of the immunolocalization of DCC in both cell bodies and axons of spiral ganglion cells is that these studies were performed on neurons that had been cultured for 48 hours. Thus, we cannot determine with certainty if this expression pattern occurs in vivo or if it results from culturing the cells. The RT-PCR experiments amplifying DCC mRNA from isolated spiral ganglion cells, harvested from mice immediately following sacrifice, did indicate active transcription of DCC. Therefore, assuming that DCC is not translationally regulated, it can be inferred that the mature protein is expressed in vivo. Additional immunohistochemical experiments to localize DCC in spiral ganglia immediately after the animals were killed would elucidate the expression patterns of DCC peptide in vivo.
The addition of exogenous netrin-1 led to an increase in neurite length of cultured mouse spiral ganglion cells. Specifically, recombinant chick netrin-1 (R&D Systems) was used for these experiments. Chick and mouse netrin-1 have 86% amino acid homology, and thus the recombinant chick peptide has bioactivity in mouse tissue. Recombinant mouse netrin-1
is now available, and similar experiments using this protein with mouse spiral ganglion cells may demonstrate a more pronounced effect. The effect of promoting neurite length with netrin-1 in our experiments demonstrated a dosage-dependent response at concentrations of 100, 200, and 500 ng/mL. These concentrations coincide with the range at which the peptide demonstrates a linear bioactivity curve based on functional enzyme-linked immunosorbent assay measuring of binding to DCC (R&D Research specification sheet).
All data reported in this study were from spiral ganglion cells of young postnatal C57 mice with the exception of the coculture experiments with the HEK293–netrin-1–expressing cells. The neurons used in these experiments were from acoustic ganglia of chicks. Initial coculture experiments were also performed with C57 mice spiral ganglion cells, but the neuronal yields from these preparations were extremely low and not sufficient for quantitative analysis. In contrast, cell preparations from the chick acoustic ganglia were much higher. Owing to the high amino acid sequence homology of netrin-1 across species, the human peptide secreted by the HEK293–netrin-1 cells is active in cultures of chick neurons.22
In summary, we have shown that netrin-1 can promote spiral ganglion cell neurite outgrowth and guide their axons by attraction in vitro in postnatal mice. Further investigations of the axon-guidance activity of netrin-1 and other peptides with similar function may lead to an understanding of the molecular mechanisms that guide spiral ganglion cells to appropriate innervation targets during embryonic development. Studies performed in developing mice prior to cochlear hair cell innervation may provide additional insight to this understanding. Application of these mechanisms would allow for the development of strategies to promote controlled outgrowth of spiral ganglion cell axons toward specific electrodes on the array of a cochlear implant. The result of this would potentially improve speech understanding and the tonal quality of sound perceived by cochlear implant users and take us to the next level in treating patients with considerable hearing losses.
Correspondence: Kenneth H. Lee, MD, PhD, Department of Otolaryngology–Head and Neck Surgery, University of Texas Southwestern Medical Center at Dallas, Children's Medical Center Dallas, 5323 Harry Hines Blvd, Dallas, TX 75390-9035
Submitted for Publication: June 7, 2006; final revision received May 24, 2007; accepted July 3, 2007.
Author Contributions: Drs Lee and Warchol had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Lee and Warchol. Acquisition of data: Lee. Analysis and interpretation of data: Lee and Warchol. Drafting of the manuscript: Lee. Critical revision of the manuscript for important intellectual content: Lee and Warchol. Obtained funding: Lee and Warchol. Administrative, technical, and material support: Warchol. Study supervision: Warchol.
Financial Disclosure: None reported.
Funding/Support: This study was supported by National Institutes of Health grants NIH-NIDCD No. 1F32DC0043501
to Dr Lee, NIH-NIDCD grant No. DC03576 to Dr Warchol, and an Otologic Research Grant from the Deafness Research Foundation to Drs Lee and Warchol.
Create a personal account or sign in to: