Abstract
Purpose
Low numbers of primary auditory neurons (ANs) may compromise the clinical performance of a cochlear implant. The focus of this research is to determine whether stem cells can be used to replace the ANs lost following deafness. To successfully replace these neurons, stem cells must be capable of directed differentiation into a sensory neural lineage in vitro and, subsequently, of survival and integration into the deafened cochlea.
Materials and Methods
In this study, we compared three in vitro treatments for directing the differentiation of mouse embryonic stem cells toward a sensory neural fate using neurotrophins, conditioned media from early post-natal cochlear epithelium, or media containing BMP4.
Results
In all treatments, stem cells were first exposed to retinoic acid, which was sufficient to induce Brn3a-positive patterning in 8-day differentiated embryoid bodies. After a further 8 days of differentiation in adherent culture conditions, BMP4 media-treated cultures produced higher proportions of cells expressing sensory neural markers in comparison to both the conditioned media and neurotrophin treatments, including significantly greater numbers of cells expressing peripherin ( P ≤ .001), tyrosine receptor kinase B ( P ≤ .001), and β -III tubulin ( P ≤ .001).
Conclusions
This study illustrated that combined treatment with retinoic acid and BMP4 was most effective at directing differentiation of mouse stem cells into sensory-like neurons in vitro . This finding further supports the role of bone morphogenetic proteins in the differentiation of sensory neurons from neural progenitors, and provides a basis for allotransplantation studies for auditory neuron replacement in the deaf mouse cochlea.
1
Introduction
In humans and other mammals, sensorineural hearing loss results in the loss of both the sensory hair cells and auditory neurons (ANs) of the cochlea, and neither cell type is capable of regeneration. Although a cochlear implant can be used to electrically stimulate residual ANs following hearing loss, these neurons progressively degenerate and low numbers may compromise the efficacy of this device . The purpose of this research is to determine whether stem cells can be used to provide replacement neurons to the deaf mammalian cochlea for combined therapy with a cochlear implant . In developing such a therapy, stem cells must be capable of differentiation into the sensory neural subpopulation required. Although neural differentiation of embryonic stem cells (ESCs) has been widely described , differentiation into more defined lineages is still a major challenge. This is certainly the case for differentiation of primary ANs, and few reports to date have described the production of auditory-like sensory neurons from ESCs in vitro ; one using human ESCs treated with BMP4 and another using genetic modification to produce glutamatergic neurons . However, the derivation of a defined population of ANs in vitro is complicated by the fact that there is no single specific marker that characterizes this cell type.
A logical approach to directing the differentiation of ESCs toward an auditory neural fate is to attempt to recapitulate the developmental stages of inner ear formation in the embryo. The primary ANs of the mammalian cochlea arise from the cranial placodes and, more specifically, from the medial ectoderm of the otic placode . The otic placode is the earliest morphologically visible event in the development of the inner ear , and this ectodermal thickening develops in close proximity to the hindbrain. The development of neurosensory progenitors from the otic placode is thought to follow the same molecular principles of brain development, including the combined interaction of diffusible signals including fibroblast growth factors (bFGF; 3, 8, and 10), sonic hedgehog, wingless proteins and bone morphogenetic proteins (BMPs ). The interaction between these 4 factors produces a patterning gradient which induces target cells into otic placode derivatives, and neurosensory identity is facilitated when Neurogenin 1 is up-regulated . Neurosensory progenitors in the inner ear have been shown to sequentially express Neurogenin 1, neurogenic differentiation factor 1 ( NeuroD ), brain-specific homeobox/POU domain protein 3a (Brn3a), and transcription factor GATA3 (GATA3) before delaminating from the otocyst and maturing into the sensory neurons of the inner ear . The importance of each of these factors is substantiated by knockout studies, in which mutant animals show a substantial or complete loss of ANs when they lack Neurogenin 1 and 2 , NeuroD , Brn3a , and GATA3 . In the mature system, ANs express many of the pan neuronal cytoskeletal markers including the neurofilament triplet proteins and peripherin, and more specifically, the tyrosine receptor kinase for brain-derived neurotrophic factor (BDNF) and neurotrophin 3 (NT3; tyrosine receptor kinase B [TrkB] and C [TrkC], respectively). Notably, the TrkB and TrkC receptors are critical for normal development, as evidenced by knockout studies in which deletion of one or both of these receptors results in a complete loss of ANs at birth .
To recapitulate the differentiation of neurosensory progenitors, we must first induce the formation of ectoderm. Ectoderm can be induced in mouse ESCs following floating sphere culture in the presence of retinoic acid (vitamin A; ). This has been shown to produce ectoderm at the expense of mesoderm and, importantly, induce the formation of hindbrain neural subtypes when used at high concentrations . Furthermore, retinoic acid is required for normal auditory development, and disrupting production of retinoic acid in the mammalian embryo causes defects in the size and number of otic vesicles produced and alters the growth and regional patterning of the otocyst . Previous experimentation in our laboratory used a combination of retinoic acid treatment and co-culture with early post-natal sensory epithelium in an attempt to produce ANs from ESCs . The described study was promising in that the combined treatment produced significantly greater numbers of bipolar auditory-like neurons . However, the subtype of neuron produced from this treatment needed further characterization and the soluble factors responsible for the differentiation needed to be identified.
The BMPs are emerging as important diffusible factors in inner ear development and cell specification . BMP4 mRNA is expressed in the developing mammalian otic placode, and expression persists in the sensory epithelium and mesenchyme surrounding the cochleae until post-natal day 1 . In addition, there are both functional and anatomical defects in BMP4 heterozygous null mice , further supporting their important role in inner ear development. Importantly, BMP4 can also be used to induce differentiation of human stem cell-derived neural progenitors into sensory-like neurons . Although the BMPs are generally known to antagonize neural differentiation , BMPs are capable of promoting sensory neural differentiation if applied to specific progenitors that express Neurogenin 1 .
The purpose of the present study was to examine whether retinoic acid-treated embryoid bodies could be preferentially directed toward a sensory neural phenotype, by comparing three established in vitro differentiation models. The models tested the effect of the human recombinant neurotrophins BDNF and NT3, the effect of conditioned media from early post-natal cochlear epithelium (containing biological levels of BMP4 and BMP7 ), or the effect of treatment with commercially available BMP4 to produce greater numbers of sensory-like neurons in vitro . In situ immunocytochemical analyses were used in order to investigate changes in protein expression at the translational level.
2
Materials and methods
2.1
Maintenance and passaging of mouse ESCs
The mouse ESC line R1 (from 129X1/SvJ × 129S1, F1 3.5-day blastocyst) ( http://www.mshri.on.ca/nagy/r1.htm ) was used for this study. Undifferentiated mouse ESCs were grown on mitomycin-treated mouse embryonic fibroblasts in standard ESC media comprising Dulbecco’s modified Eagle’s medium (DMEM; Invitrogen, Melbourne, Australia), supplemented with 10% ESC-qualified fetal bovine serum (Invitrogen), 1% penicillin/streptomycin (Invitrogen), 1% nucleosides (Invitrogen), 1% nonessential amino acids (Invitrogen), 1% l -glutamine (Invitrogen), 1 mL/L 1000× β -mercaptoethanol (Invitrogen), and 1 mL/L (1000 U) leukemia inhibitory factor (LIF; Chemicon, Melbourne, Australia). Stem cells were grown at 37°C, 5% CO 2 and passaged every 2–3 days using 2 mL TrypLE (Invitrogen). Once cultures were established on feeders (approximately 3 weeks), mouse ESCs were grown for a further 2 weeks without mouse embryonic fibroblasts, in flasks coated with 0.1% gelatin (Millipore), before being induced to form embryoid bodies.
2.2
Embryoid body sphere formation using retinoic acid
Flasks containing rounded, undifferentiated colonies were selected for further experimentation. Briefly, undifferentiated mouse ESCs were passaged as described above and then resuspended in standard ESC media without LIF. The resulting cell suspension was then collected and centrifuged at 1000× g for 5 min, the supernatant discarded, and the pellet resuspended in 1 mL fresh ESC media (no LIF). A viable cell count was performed to accurately deliver approximately 1 × 10 4 cells to 4 petri dishes containing a total of 10 mL ESC media (no LIF). This process was repeated using flasks from 3 consecutive passages. Media was changed on embryoid bodies after 2, 4, and 6 days in vitro (DIV). At the 4 and 6 DIV time points, 0.05 μ M all-trans retinoic acid (Sigma) was added to the culture media to induce neural differentiation . After 8 DIV, the embryoid bodies were collected in a 15 mL tube and rinsed 3 × 10 min in ESC media (no LIF). After the final rinse, embryoid bodies were plated (one per well) into 60-chamber slide wells (Nunc), precoated with 500 μ g/mL poly- d -ornithine (Sigma) and 0.01 mg/mL laminin (Invitrogen). Each well contained 200 μ L freshly prepared ESC media (no LIF) containing 10 ng/mL bFGF (Invitrogen). Embryoid bodies were then incubated at 37°C, 10% CO 2 for 1 day, before receiving a further 7 days of specified media treatments (detailed below). The plating of embryoid bodies (1 per well into 60 individual chamber slide wells) was repeated twice more for cells derived from the remaining 2 consecutive passages (above). Surplus embryoid bodies were collected at 8 DIV and were fixed and processed as described below.
2.2.1
Preparation of media for experimental conditions
Three media treatments were prepared for directing embryoid body differentiation toward a sensory neural lineage; neurotrophin media (control), conditioned media, and BMP4 media. All media was made from DMEM/F12 media (Invitrogen) and contained N2 and B27 supplements (Invitrogen), 6 g/L d -glucose (Invitrogen) and 50 ng/mL of each of the human recombinant neurotrophins BDNF (Millipore) and NT3 (Millipore). The neurotrophin media (control media) contained only these constituents. The conditioned media was collected from early post-natal cochlear tissue as described below. The BMP4 containing media included 10 ng/mL human BMP4 (Peprotech). A total of 12 mL of each media type was prepared to add to the plated embryoid bodies.
2.3
Collection of conditioned media from cochlear sensory explants
Sensory explants were dissected and cultured from post-natal rat pup cochleae (P3–P9), as previously described . Each day between P3 and P9, sensory explants were isolated and plated onto 0.4 μ m organotypic membranes (Millipore; 2–3 explants per membrane) and grown in DMEM/F12 with supplements, as above. Explants were maintained at 37°C, 10% CO 2 for 1 day before conditioned media was collected and stored at 4°C for up to 2 weeks. A total of 6 mL of conditioned media was collected each day between P3 and P9 and this was added to 6 mL of fresh DMEM/F12 base media to make 12 mL of media required in total for each time point. In addition, ANs isolated and cultured from P3-6 rat pups were used to provide positive controls for immunocytochemistry experiments. All experimentation was conducted in accordance with the Royal Victorian Eye and Ear Hospital Animal Research and Ethics Committee Guidelines (project approval number 07/139A).
2.3.1
Addition of media to experimental treatments
Eight-day differentiated embryoid bodies were plated into individual wells (10 per slide) and maintained for 1 day at 37°C, 10% CO 2 . The slides were then allocated randomly to one of the three treatments groups; neurotrophin-treated (group 1), conditioned media-treated (Group 2), or BMP4-treated (Group 3). Experimental timelines for all treatments are illustrated schematically in Fig. 1 . For each group, the appropriate media was warmed to 37°C and then 200 μ L per well was added to the plated cells. All wells were then re-incubated at 37°C, 10% CO 2 . Every 2 days, 100 μ L of media was removed from each well and replaced with 100 μ L fresh media corresponding to the treatment group. Cells were maintained in adherent culture for 8 days using this regime.
2.4
Fixation and immunocytochemistry
All cells were fixed in ice-cold 4% paraformaldehyde (BDH Laboratories) for 10 minutes at 0, 8, and 16 DIV to investigate the extent of sequential differentiation toward a sensory neural lineage in each treatment group. Early, middle, and late time points were chosen for fixation as follows: undifferentiated mouse ESCs (0 DIV), partially differentiated mouse ESCs following treatment with retinoic acid (embryoid bodies, 8 DIV), and mouse ESCs treated with retinoic acid and grown for a further 8 days in one of three experimental culture conditions described (16 DIV). Following fixation, cells and embryoid bodies were rinsed three times for 10 min in phosphate-buffered saline (PBS; Invitrogen). The embryoid bodies were then embedded in Tissue-Tek O.C.T. compound (ProSciTech), frozen at −20°C, and the spheres cryosectioned at 10 μ m until sections from the centre of the sphere were obtained. These sections were collected onto Fischerbrand SuperFrost Plus slides (ProSciTech) and processed for immunocytochemistry in conjunction with cells grown on tissue culture glass as described below.
A cohort of known sensory neural markers was employed to characterize and quantify the numbers of sensory-like neurons produced from each treatment. The primary antibodies used in this study including the optimal concentration and manufacturer’s details are described in Table 1 . All antibody concentrations were determined individually and negative controls (without primary antibody) eliminated all immunoreactivity. Standard immunocytochemical procedures were used to fluorescently label fixed cells. Briefly, cells were blocked in 0.1% triton-X in PBS containing 2% goat serum for 1 hour at room temperature; before primary antibodies were added at the desired concentration, 200 μ L per well for 2 hours at room temperature and then overnight at 4°C. The following day, cells were rinsed in blocking solution and the relevant secondary antibodies added at the concentrations described in Table 1 , for 2 hours at room temperature with gentle rotation. All cells were then rinsed 3 times for 10 min in PBS and mounted with ProLong Gold antifade reagent containing the nuclear stain 4′,6-diamidino-2-phenylindole (DAPI) (Invitrogen). Slides were dried at room temperature overnight and the cover glass edges sealed with nail varnish the following day.
| Antibody | Type (clone) | Dilution | Source | Secondary used | Positive controls |
|---|---|---|---|---|---|
| SSEA1 | Mouse monoclonal IgM | 1:200 | AbCam ab16285 | Goat anti-mouse AlexaFluor IgG (488) 1:400 | Undifferentiated mouse ESC (cell surface) |
| Brn3a | Mouse monoclonal IgGR 1 | 1:1000 | Millipore MAB1585 | Goat anti-mouse AlexaFluor IgGR 1 (R488) 1:400 | Early post-natal ANs (nuclear) |
| Peripherin | Mouse monoclonal IgG | 1:1000 | Millipore MAB1527 | Goat anti-mouse AlexaFluor IgG (488) 1:400 | Early post-natal rat ANs, adult mouse and guinea pig ANs (soma) |
| Trk B | Rabbit polyclonal IgG | 1:500 | Millipore AB5372 | Goat anti-rabbit AlexaFluor IgG (594) 1:400 | Adult guinea pig ANs and brain (cytoskeleton) |
| β -III tubulin | Chicken polyclonal IgY | 1:1000 | Millipore AB9354 | Goat anti-chicken AlexaFluor IgG (647) 1:400 | Early post-natal rat ANs (cytoskeleton) |
| Neurofilament 200 kDa (NF200) | Chicken polyclonal IgY | 1:800 | Millipore AB5539 | Goat anti-chicken AlexaFluor IgG (647) 1:400 | Early post-natal rat ANs, adult guinea pig ANs, adult mouse brain (cytoskeleton) |
2.5
Image acquisition and quantification
Fluorescence photomicrographs were taken using an LSM 510 META confocal scanning laser system attached to a Zeiss AxioImagerZ1 microscope and image analysis was performed using LSM Image Browser software. Immunostaining was examined in cryosectioned embryoid bodies (n = 512) and spheres containing positively labeled cells were recorded and expressed as a percentage of the total counted (identified by DAPI nuclear staining). In adherent cultures, the number of positively labeled cells was counted in 10 randomly selected fields of view around the peripheral edge of the plated embryoid bodies (n = 120 per treatment group). The total numbers of positively labeled cells in each treatment group were added and expressed as a percentage of the total number of cells counted (quantified by DAPI nuclear staining), plus or minus the SEM. All data were statistically analyzed using SigmaPlot 11 software to identify significant differences in the numbers of positively labeled cells between treatment groups. Statistically significant effects were detected using a Kruskal-Wallis 1-way analysis of variance (ANOVA) on Ranks, with Dunn’s All Pairwise Multiple Comparison Procedure to isolate the groups that differed.
2
Materials and methods
2.1
Maintenance and passaging of mouse ESCs
The mouse ESC line R1 (from 129X1/SvJ × 129S1, F1 3.5-day blastocyst) ( http://www.mshri.on.ca/nagy/r1.htm ) was used for this study. Undifferentiated mouse ESCs were grown on mitomycin-treated mouse embryonic fibroblasts in standard ESC media comprising Dulbecco’s modified Eagle’s medium (DMEM; Invitrogen, Melbourne, Australia), supplemented with 10% ESC-qualified fetal bovine serum (Invitrogen), 1% penicillin/streptomycin (Invitrogen), 1% nucleosides (Invitrogen), 1% nonessential amino acids (Invitrogen), 1% l -glutamine (Invitrogen), 1 mL/L 1000× β -mercaptoethanol (Invitrogen), and 1 mL/L (1000 U) leukemia inhibitory factor (LIF; Chemicon, Melbourne, Australia). Stem cells were grown at 37°C, 5% CO 2 and passaged every 2–3 days using 2 mL TrypLE (Invitrogen). Once cultures were established on feeders (approximately 3 weeks), mouse ESCs were grown for a further 2 weeks without mouse embryonic fibroblasts, in flasks coated with 0.1% gelatin (Millipore), before being induced to form embryoid bodies.
2.2
Embryoid body sphere formation using retinoic acid
Flasks containing rounded, undifferentiated colonies were selected for further experimentation. Briefly, undifferentiated mouse ESCs were passaged as described above and then resuspended in standard ESC media without LIF. The resulting cell suspension was then collected and centrifuged at 1000× g for 5 min, the supernatant discarded, and the pellet resuspended in 1 mL fresh ESC media (no LIF). A viable cell count was performed to accurately deliver approximately 1 × 10 4 cells to 4 petri dishes containing a total of 10 mL ESC media (no LIF). This process was repeated using flasks from 3 consecutive passages. Media was changed on embryoid bodies after 2, 4, and 6 days in vitro (DIV). At the 4 and 6 DIV time points, 0.05 μ M all-trans retinoic acid (Sigma) was added to the culture media to induce neural differentiation . After 8 DIV, the embryoid bodies were collected in a 15 mL tube and rinsed 3 × 10 min in ESC media (no LIF). After the final rinse, embryoid bodies were plated (one per well) into 60-chamber slide wells (Nunc), precoated with 500 μ g/mL poly- d -ornithine (Sigma) and 0.01 mg/mL laminin (Invitrogen). Each well contained 200 μ L freshly prepared ESC media (no LIF) containing 10 ng/mL bFGF (Invitrogen). Embryoid bodies were then incubated at 37°C, 10% CO 2 for 1 day, before receiving a further 7 days of specified media treatments (detailed below). The plating of embryoid bodies (1 per well into 60 individual chamber slide wells) was repeated twice more for cells derived from the remaining 2 consecutive passages (above). Surplus embryoid bodies were collected at 8 DIV and were fixed and processed as described below.
2.2.1
Preparation of media for experimental conditions
Three media treatments were prepared for directing embryoid body differentiation toward a sensory neural lineage; neurotrophin media (control), conditioned media, and BMP4 media. All media was made from DMEM/F12 media (Invitrogen) and contained N2 and B27 supplements (Invitrogen), 6 g/L d -glucose (Invitrogen) and 50 ng/mL of each of the human recombinant neurotrophins BDNF (Millipore) and NT3 (Millipore). The neurotrophin media (control media) contained only these constituents. The conditioned media was collected from early post-natal cochlear tissue as described below. The BMP4 containing media included 10 ng/mL human BMP4 (Peprotech). A total of 12 mL of each media type was prepared to add to the plated embryoid bodies.
2.3
Collection of conditioned media from cochlear sensory explants
Sensory explants were dissected and cultured from post-natal rat pup cochleae (P3–P9), as previously described . Each day between P3 and P9, sensory explants were isolated and plated onto 0.4 μ m organotypic membranes (Millipore; 2–3 explants per membrane) and grown in DMEM/F12 with supplements, as above. Explants were maintained at 37°C, 10% CO 2 for 1 day before conditioned media was collected and stored at 4°C for up to 2 weeks. A total of 6 mL of conditioned media was collected each day between P3 and P9 and this was added to 6 mL of fresh DMEM/F12 base media to make 12 mL of media required in total for each time point. In addition, ANs isolated and cultured from P3-6 rat pups were used to provide positive controls for immunocytochemistry experiments. All experimentation was conducted in accordance with the Royal Victorian Eye and Ear Hospital Animal Research and Ethics Committee Guidelines (project approval number 07/139A).
2.3.1
Addition of media to experimental treatments
Eight-day differentiated embryoid bodies were plated into individual wells (10 per slide) and maintained for 1 day at 37°C, 10% CO 2 . The slides were then allocated randomly to one of the three treatments groups; neurotrophin-treated (group 1), conditioned media-treated (Group 2), or BMP4-treated (Group 3). Experimental timelines for all treatments are illustrated schematically in Fig. 1 . For each group, the appropriate media was warmed to 37°C and then 200 μ L per well was added to the plated cells. All wells were then re-incubated at 37°C, 10% CO 2 . Every 2 days, 100 μ L of media was removed from each well and replaced with 100 μ L fresh media corresponding to the treatment group. Cells were maintained in adherent culture for 8 days using this regime.
