Three-dimensional organotypic cultures of the murine utricle and cochlea in optically clear collagen I gels preserve innate tissue morphology, allow for mechanical stimulation through adjustment of matrix stiffness, and permit virus-mediated gene delivery.
The sensory organs of the inner ear are challenging to study in mammals due to their inaccessibility to experimental manipulation and optical observation. Moreover, although existing culture techniques allow biochemical perturbations, these methods do not provide a means to study the effects of mechanical force and tissue stiffness during development of the inner ear sensory organs. Here we describe a method for three-dimensional organotypic culture of the intact murine utricle and cochlea that overcomes these limitations. The technique for adjustment of a three-dimensional matrix stiffness described here permits manipulation of the elastic force opposing tissue growth. This method can therefore be used to study the role of mechanical forces during inner ear development. Additionally, the cultures permit virus-mediated gene delivery, which can be used for gain- and loss-of-function experiments. This culture method preserves innate hair cells and supporting cells and serves as a potentially superior alternative to the traditional two-dimensional culture of vestibular and auditory sensory organs.
The study of most aspects of mammalian organ development has been facilitated by in vitro systems. Two principal methods are now used for the culture of vestibular sensory organs: free-floating1 and adherent2 preparations. Both methods permit the investigation of hair cell vulnerabilities3 and regeneration1,4 in vitro. In addition, the developmental roles of the Notch5,6, Wnt7,8, and epidermal growth factor receptor (EGFR)9,10 signaling cascades in the inner ear have been established, in part, through the use of in vitro cultures of sensory epithelia. However, cell growth and differentiation are controlled, not only through signaling by morphogens, but also through physical and mechanical cues such as intercellular contacts, the stiffness of extracellular matrix, and mechanical stretching or constriction. The role of such mechanical stimuli is challenging to investigate in the developing inner ear in vivo. Moreover, existing free-floating and adherent culture methods are not suitable for such studies in vitro. Here we describe a method for three-dimensional organotypic culture in collagen I gels of varying stiffness. This method largely preserves the in vivo architecture of the vestibular and cochlear sensory organs and allows investigation of the effects of mechanical force on growth and differentiation11.
Because mechanical stimuli are known to activate downstream molecular events, such as the Hippo signaling pathway12,13,14,15, it is important to be able to combine mechanical stimulation with biochemical and genetic manipulations. The culture method described here permits virus-mediated gene delivery and can therefore be used to study both mechanical and molecular signaling during inner ear development11.
All methods described here have been approved by the Animal Care and Use Committees of Rockefeller University and of the University of Southern California.
1. (Optional) Preparation of Collagen I Solution from Mouse-tail Tendons
Note: Collagen I solutions are available commercially. Follow the manufacturer's instructions for gel preparation.
2. Dissection of Vestibular and Auditory Organs
3. (Optional) Adjust Collagen I Gel Stiffness by Adding Varying Concentrations of Chondrocytes
Note: The method for chondrocyte isolation was modified from Gosset et al.20
4. Place the Vestibular or Auditory Sensory Organ in a Collagen I Gel
5. Viral Injections in Three-dimensional Cultures of Vestibular and Auditory Sensory Organs
Vestibular and auditory sensory organs from embryonic ears, cultured in 40-Pa collagen I gels mimicking low stiffness embryonic conditions11, retain relatively normal three-dimensional structures (Figure 1) and maintain hair cells and supporting cells (Figure 2 and Figure 3). Although supporting cell density decreases by over 30% (Student's t-test: n = 4, p <0.004) and hair cell density decreases by 60% (Student's t-test: n = 5 , p <0.0001) after 3 days in utricular cultures (Figure 2), the area of the macula more than doubles over the same period of time11 (n = 3, p = 0.0002). This demonstrates that the method for three-dimensional culture described here allows for a significant increase in the number of supporting cells11, while maintaining 80% of existing hair cells in the utricle over a 3 day period. In the three-dimensional cultures established from E14.5 cochlea, Sox2-positive progenitor cells differentiate as morphologically distinguishable rows of inner and outer hair cells after 3 days in culture (Figure 3).
Gene expression can be manipulated in three-dimensional cultures of vestibular and auditory sensory organs by means of viral infection. 4hydroxytamoxifen is added to the culture medium to label supporting cells in the cochlear explants established from E15.5 embryos of Lfng-CreERT2/tdTomato mice17. Injection of adenovirus type 5 into the lumen of the culture results in infection of supporting cells at the organ's base (Figure 4A). Injection of the same virus into the lumen of utricular culture established from E17.5 embryo, results primarily in infection of supporting cells throughout the sensory epithelium (Figure 4B).
Figure 1. Schematic diagrams of dissections and light-microscopic images of representative cultures of utricle and cochlea in three-dimensional collagen I gels. (A) A schematic drawing portrays the sensory epithelia (green) of the six receptor organs of the murine inner ear. The red lines delineate the cuts introduced during the dissection of an utricle and cochlea. (B) Light microscopy images portray the E17.5 utricle (upper panel) and E14.5 cochlea (lower panel) embedded in collagen I gel and cultured for 48 h. The scale bars represent 100 µm. This figure has been modified from Gnedeva et al.11 Please click here to view a larger version of this figure.
Figure 2. Hair cells and supporting cells in three-dimensional utricular cultures. (A) Confocal-microscopic images portray an E18.5 utricle prior to explantation (top panels) and after 3 days in a three-dimensional culture in 40-Pa collagen I gel (bottom panels). Hair cells are labeled for Myo7A (green) and supporting cells for Sox2 (red). The scale bar represents 50 µm. (B) Quantifications of supporting cell densities (red bars), hair cell densities (green bars), and macular areas (grey bars) in E18.5 utricles prior to explantation and after 3 days in three-dimensional organ culture are represented as means ± SEMs (p <0.001 is represented as ** and p <0.0001 as ***). This figure has been modified from Gnedeva et al.11 Please click here to view a larger version of this figure.
Figure 3. Hair cells and supporting cells in three-dimensional cochlear cultures. Confocal-microscopic images portray an E14.5 cochlea prior to explantation (top panels) and after 3 days in a three-dimensional culture in 40-Pa collagen I gel (bottom panels). Myo7A- and Sox2-positive inner hair cells (IHC) and outer hair cells (OHC) appear in 4 – 5 rows after 3 days in culture. Supporting cells are also labeled for Sox2 (red). The scale bar represents 25 µm. Please click here to view a larger version of this figure.
Figure 4. Representative results of viral infections in three-dimensional organ cultures of the cochlea and utricle. (A) Injection of adenovirus serotype 5 into three-dimensional cochlear cultures established from Lfng-CreERT2/tdTomato26 E15.5 embryos results in infection (GFP, green) of supporting cells (Tomato, red) at the base of the organ. The sensory epithelium is delineated in gray. The scale bar represents 100 µm. (B) An identical injection into a three-dimensional utricular culture established from an E17.5 embryo results in infection (GFP, green) of supporting cells (Sox2, red) throughout the organ. The sensory epithelium is delineated in gray. The scale bar represents 100 µm. This figure has been modified from Gnedeva et al.11 Please click here to view a larger version of this figure.
The molecular signals that mediate growth and differentiation in the inner ear during development have been studied extensively5,6,7,8,9,10. However, evidence obtained from the utricular model system suggests that mechanical cues, sensed through cell junctions and the activation of Hippo signaling, also play an important role in these processes2,11,22. Moreover, both the extrusion of dying hair cells from the sensory epithelium and the subsequent formation of new sensory receptors through transdifferentiation can affect the mechanical force sensed by the residual supporting cells, causing them to re-enter the cell cycle during regeneration. The three-dimensional culture system described here provides an experimental means of investigating both the role of mechanical force in growth control during inner ear development11 and, potentially, the role of same force during hair cell regeneration. Moreover, the approach facilitates viral infection that provides a method of altering gene expression in the sensory epithelium, thus permitting a combination of mechanical and molecular manipulations for the investigation of growth and regeneration in the ear's sensory organs11.
The limitations of the method relate to the minimal information available on the endogenous forces and mechanical stimuli during inner ear embryogenesis. Measurements of tissue stiffness in the developing inner ear do not exist to our knowledge; hence it is hard to estimate what stiffness of collagen I gel corresponds to in vivo conditions. Our observations and the model suggest that the force opposing growth of the utricle is low initially and increases as the organ approaches its final size11. We therefore hypothesize that a collagen I gel without chondrocytes is a physiologically relevant substrate in which to culture embryonic vestibular and auditory sensory organs.
The three-dimensional culture method described here induces formation of new supporting cells in the utricular macula11, while also maintaining over 80% of hair cells after 3 days in culture (Figure 2). The method, therefore, represents a superior alternative to two-dimensional cultures of the utricle, in which only 40 – 50% of hair cells survive after the first 24 h in culture5,8, and can be used to study hair cell vulnerabilities and regeneration in vitro.
Although we demonstrate the formation of anatomically distinguishable organized rows of inner and outer hair cells in the cultured organ of Corti, more work is required to determine whether the three-dimensional culture method described here supports normal cochlear duct elongation during the process of convergent extension (CE)23,24,25. CE is a highly dynamic process involving cell migration, rearrangement, and cell-cell contact changes26 that is likely to be affected by the external force produced by the tissues surrounding the developing cochlear duct. This method does preserve relatively normal three-dimensional tissue architecture, and could potentially be beneficial for the study of CE in vitro.
The authors have nothing to disclose.
We thank Dr. A. Jacobo, Dr. J. Salvi, and A. Petelski for their contributions to the original research on which this protocol is based. We also thank J. Llamas and W. Makmura for technical assistance and animal husbandry. We acknowledge NIDCD Training grant T32 DC009975, NIDCD grant R01DC015530, Robertson Therapeutic Development Fund, and the Caruso Family Foundation for funding. Finally, we acknowledge support from Howard Hughes Medical Institute, of which Dr. Hudspeth is an Investigator.
#10 Surgical Blades | Miltex | 4-110 | |
#5 Forceps | Dumont | 11252-20 | |
100 mm Petri dish | Sigma | P5856-500EA | |
250 uL large orifice pipette tips | USA Scientific | 1011-8406 | |
30 mm glass-bottom Petri dish | Matsunami Glass USA Corporation | D35-14-1.5-U | |
4 well plate | Thermo Fisher Scientific | 176740 | |
4-Hydroxytamoxifen | Sigma | H7904 | |
60 mm Petri dish | Thermo Fisher Scientific | 123TS1 | |
Acetic acid | Sigma | 537020 | |
Ad-GFP | Vector Biolabs | 1060 | |
Anti-GFP, chicken IgY fraction | Invitrogen | A10262 | |
Anti-Myo7A | Proteus Biosciences | 25-6790 | |
Anti-Sox2 Antibody (Y-17) | Santa Cruz | sc-17320 | |
Bicinchoninic acid assay | Thermo Fisher Scientific | 23225 | |
Click-iT EdU Alexa Fluor 647 Imaging Kit | Thermo Fisher Scientific | C10340 | |
Collagenase I | Gibco | 17100017 | |
D-glucose | Sigma | G8270 | |
DMEM/F12 | Gibco | 11320033 | |
Epidermal growth factor | Sigma | E9644 | |
Fetal Bovine Serum (FBS) | Thermo Fisher Scientific | 16140063 | |
Fibroblast growth factor | Sigma | F5392 | |
Flaming/Brown Micropipette Puller | Sutter Instrument | P-97 | |
Glutamine | Sigma | G8540 | |
HBSS | Gibco | 14025092 | |
Hemocytometer | Daigger | EF16034F | |
HEPES | Sigma | H4034 | |
Insulin | Sigma | I3536 | |
Iridectomy scissors | Zepf Medical Instruments | 08-1201-10 | |
Microinjector | Narishige | IM-6 | |
Nicotinamide | Sigma | N0636 | |
PBS (10X), pH 7.4 | Gibco | 70011044 | |
PBS (1X), pH 7.4 | Gibco | 10010023 | |
Phenol Red pH indicator | Sigma | P4633 | |
Pure Ethanol, 200 Proof | Decon Labs | 2716 | |
RFP antibody | ChromoTek | 5F8 | |
Sodium bicarbonate | Sigma | S5761 | |
Sodium hydroxide | Sigma | S8045 | |
Sodium selenite | Sigma | S5261 | |
Tabletop vortex | VWR | 97043-562 | |
Transferrin | Sigma | T8158 | |
Trypan blue | Sigma | T6146 |