The goal of this protocol is to demonstrate an effective method to decellularize and decalcify mouse cochleae for utilization as scaffolds for tissue engineering applications.
In mammals, mechanosensory hair cells that facilitate hearing lack the ability to regenerate, which has limited treatments for hearing loss. Current regenerative medicine strategies have focused on transplanting stem cells or genetic manipulation of surrounding support cells in the inner ear to encourage replacement of damaged stem cells to correct hearing loss. Yet, the extracellular matrix (ECM) may play a vital role in inducing and maintaining function of hair cells, and has not been well investigated. Using the cochlear ECM as a scaffold to grow adult stem cells may provide unique insights into how the composition and architecture of the extracellular environment aids cells in sustaining hearing function. Here we present a method for isolating and decellularizing cochleae from mice to use as scaffolds accepting perfused adult stem cells. In the current protocol, cochleae are isolated from euthanized mice, decellularized, and decalcified. Afterward, human Wharton's jelly cells (hWJCs) that were isolated from the umbilical cord were carefully perfused into each cochlea. The cochleae were used as bioreactors, and cells were cultured for 30 days before undergoing processing for analysis. Decellularized cochleae retained identifiable extracellular structures, but did not reveal the presence of cells or noticeable fragments of DNA. Cells perfused into the cochlea invaded most of the interior and exterior of the cochlea and grew without incident over a duration of 30 days. Thus, the current method can be used to study how cochlear ECM affects cell development and behavior.
The cochlea is an intricate spiral structure found in the temporal bone. It is composed of an outer bony labyrinth and a concentric, inner membranous labyrinth1. The membranous labyrinth consists of three fluid spaces: Scala vestibuli, Scala media, and Scala tympani1. The scala media houses the sensory epithelium, which is composed of a multitude of cell types, but the sensory hair cells (HC), which transduce mechanical energy in sound waves to nerve impulses2, are of particular interest. Exposure to acoustic trauma3,4,5, medication6, disease7,8, and aging9 can all result in impaired auditory function via HC death. Hair cell loss in mammals is permanent, unlike avian HCs, which can regenerate after injury10.
A variety of contemporary research efforts have sought to restore lost HCs, although the specific experimental approaches vary. Manipulation of gene expression in the sensory epithelium and implantation of stem cells differentiated outside the body are dominant approaches in the field, although induction methods that seek to differentiate stem cells into cochlear organoids have been attempted11,12,13. Each of these approaches is either reliant directly on stem cells, or the developmental cues used by stem cells; however, a second shared, and potentially critical, element is the ECM of the cochlea itself14,15.
The ECM not only provides physical support for cells and tissue, which includes a surface for cell adhesion, proliferation, survival, and migration, but also plays critical roles in the development of HCs and the spiral ganglion15,16,17. Naturally occurring ECM provides inductive signals that may guide cell phenotype determination and/or cell adhesion, proliferation, and survival18. Consequently, the use of decellularized cochlea in combination with cultured hWJCs offer a unique opportunity to explore the role of the ECM and HC regeneration. HWJCs are a readily available, non-controversial cell type isolated from human umbilical cords that behave like mesenchymal stem cells19. HWJCs have shown the ability to differentiate down neurosensory cell lineages20,21. Thus, the current protocol details the isolation, decellularization, and perfusion of cochleae from C57BL mouse carcasses with hWJCs for inner ear tissue engineering.
All procedures, including animal euthanasia, were conducted according to the approved Institutional Animal Care and Use Committee (IACUC) protocol (ACUP #2014-2234) at the University of Kansas Medical Center (KUMC).
NOTE: HWJCs were isolated from human umbilical cords that were donated by patients that provided informed consent and specimens were used in accordance with the protocols approved by the University of Kansas Human Subjects Committee (KU-IRB #15402).
1. Temporal Bone Harvest and Cochlea Isolation
2. Cochlear Processing
3. Procurement and Expansion of hWJCs
4. Infusion of hWJCs into Decellularized Cochleae
5. Cochlea Harvest and Preservation
NOTE: Cochleae may be cultured and harvested at any time point up to 30 days post-perfusion.
Using the methods presented here, successful decellularization of cochleae was assessed by examining the presence or absence of DNA through 4',6-diamidino-2-phenylindole (DAPI) staining. Cochleae were considered fully decellularized if DNA was not identified within the decellularized cochlea. A native cochlea from a previous experiment that did not undergo decellularization or decalcification was used as a positive control to illustrate the structures and cells traditionally found in a C57BL mouse cochlea. The decellularized-decalcified cochlea, which did not have any hWJCs infused, was used as a negative control. No quantifiable DAPI stained cell nuclei were observed in any region of the tissue section (Figure 3). Two cochleae were decellularized and decalcified, and then infused with hWJCs for the current project. The decellularized-decalcified cochleae that were perfused with hWJCs were observed to have numerous DAPI stained nuclei within the cochlea and growing on the bony shell outside of the cochlea (Table 1). Total estimated cell densities for the cell infused first cochlea were 113,759 cells/cochlea after 30 days of culture (Figure 4 and Figure 5) and for the second cochlea were 118,732 cells/cochlea after 30 days of culture (Figure 6 and Figure 7). These cell density estimates include the Scala tympani, Scala media, and Scala vestibule, and volumes of the structures found within the three fluid spaces, but these estimates excluded any of the remaining bony labyrinth. In addition, samples from each cochlea were stained with Hematoxylin and Eosin (H&E) to show the presence of cells and anatomical features within each cochlea (Figure 8). The gross anatomy of the cochlear microstructures remained predominantly unchanged throughout all stained sections; however, no cells were identifiable in the decellularized-decalcified section (Figure 8B). The identification of cell nuclei was dramatically different between the native cochlea (Figure 8A) and the decellularized-decalcified cochleae infused with cells (Figure 8C–D).
Figure 1: Mouse temporal bone isolation. Once the mouse has been appropriately euthanized, decapitate the animal by cutting between the base of the skull and the first cervical vertebra, and spray the head with ethanol to disinfect (A). Bisect the skull in a sagittal plane of section (B–C) using a sharp pair of surgical scissors, cutting through both the brain tissue and bone. The bisected mouse head (D, lower mandible removed for unrelated experimental work) must then have the brain tissue removed to reveal the temporal bone. The two foramina through which the auditory and vestibular nerves pass (E, red arrowheads) are useful cues for successfully identifying the temporal bone. Remove the flesh from the skull (F). The external ear canal (F, arrow, top panel) provides a useful external cue as well. Carefully trim away the excess bone, leaving just the isolated temporal bone (G). The bulla (H, top panel, red highlighted area) must then be delicately removed using fine forceps (H, lower panel). The bulla is thin bone and can be readily chipped away (I) until the bony labyrinth of the cochlea is revealed (J, red brackets). Scale bars on all images are 1 mm. Please click here to view a larger version of this figure.
Figure 2: Cochlear Perfusion. Once the temporal bone has been isolated and the bulla has been removed, the cochlea can then be set up for perfusion at a microscope. When perfusing using a manually driven syringe with tubing attached (A, dotted line) it may be helpful to stabilize the cochlea using a pair of self-closing forceps. Gently attach the forceps to the vestibular part of the temporal bone (B) and rest the forceps against the lip of the Petri dish (A). This leaves both hands free to manipulate instruments during perfusion. One hand can then manipulate the syringe, controlling the flow rate of the fluid, while the other hand can position the tubing over the oval window using a second pair of forceps (B). Cutting the free end of the tubing at an angle allows for easier positioning, allowing the tubing to be placed correctly without blocking the field of view. Scale bar is 1 mm. Please click here to view a larger version of this figure.
Figure 3: Tissue section of a decellularized mouse cochlea. An upright epi-fluorescent microscope was used to image samples at a magnification of 200X (10X eyepiece and 20X objective). Images were stitched together to create approximately a 4 x 5 montage. DAPI nuclear staining showing absence of cells is shown in panel A using a DAPI common filter set (350 nm excitation, 470 nm emission), and an overexposed, grayscale image of the autofluorescence using a fluorescein isothiocyanate (FITC) common filter set (490 nm excitation, 525 nm emission), which provides anatomical landmarks, is shown in panel B. Automated cell counts were performed in three regions of interest, which are delineated on both panels A and B. Total cell counts were calculated for the entire tissue section (1, solid white line) and the cochlea (2, dashed line). These two numbers were used to calculate the number of cells in the bony structures outside of the cochlea but within the exterior, bony edges of the tissue section (3, space between the dotted line and dashed line). Scale bars on all images are 500 µm. Please click here to view a larger version of this figure.
Figure 4: A near-modiolar section of decellularized mouse cochlea 1 infused with hWJCs. DAPI nuclear staining is shown in panel A overlaid onto an overexposed, grayscale autofluorescence from the green channel, which provides anatomical landmarks. Automated cell counts were performed in three regions of interest, which are delineated in panel A. Total cell counts were calculated for the entire tissue section (1, solid white line) and the cochlea (2, dashed line). These two numbers were used to calculate the number of cells in the bony structures outside of the cochlea but within the exterior, bony edges of the tissue section (3, space between the dotted line and dashed line). Isolated DAPI staining is shown in panel B, and the threshold of the DAPI staining used during automated quantification is shown in panel C. Scale bars on all images are 500 µm. Please click here to view a larger version of this figure.
Figure 5: High magnification view of a near-modiolar section of decellularized mouse cochlea 1 infused with hWJCs. DAPI nuclear staining is shown in panel A overlaid onto an overexposed, grayscale image of the autofluorescence from the green channel, which provides anatomical landmarks. Isolated DAPI staining is shown in panel B, and the threshold image used during automated cell counting is shown in panel C. The fluid spaces and finer microstructures housed within the cochlea are nicely preserved during the decellularization and cell culture phases of the experiment. Scala Vestibuli (SV), Scala Tympani (SM), Basilar Membrane (BM), Tectorial Membrane (TM), Spiral Limbus (L), Rosenthal's Canal (RC), Modiolus (M), Spiral Ligament (SL), and Stria Vascularis (*). Reissner's membrane was not clearly defined in tissue sections, and as a result the Scala Media was not clearly defined. Scale bars on all images are 250 µm. Please click here to view a larger version of this figure.
Figure 6: A near-modiolar section of decellularized mouse cochlea 2 that was infused with hWJCs. DAPI nuclear staining is shown in panel A, overlaid onto grayscale autofluorescence from the green channel, which provides anatomical landmarks. Automated cell counts were performed in three regions of interest, which are delineated in panel A. Total cell counts were calculated for the entire tissue section (1, solid white line) and the cochlea (2, dashed line). These two numbers were used to calculate the number of cells in the bony structures outside of the cochlea but within the exterior, bony edges of the tissue section (3, space between the dotted line and dashed lines). Isolated DAPI staining is shown in panel B, and the threshold of the DAPI staining used during automated quantification is shown in panel C. Scale bars on all images are 500 µm. Please click here to view a larger version of this figure.
Figure 7: High magnification view of a near-modiolar section of decellularized mouse cochlea 2 that was infused with hWJCs. DAPI nuclear staining is shown in panel A, overlaid onto grayscale autofluorescence from the green channel, which provides anatomical landmarks. Isolated DAPI staining is shown in panel B, and the threshold image used during automated cell counting is shown in panel C. The fluid spaces and finer microstructures housed within the cochlear are preserved during the decellularization and cell culture phases of the experiment. Scala Vestibuli (SV), Scala Tympani (SM), Basilar Membrane (BM), Tectorial Membrane (TM), Spiral Limbus (L), Rosenthal's Canal (RC), Modiolus (M), Spiral Ligament (SL), and Stria Vascularis (*). Reissner's Membrane was not clearly defined in tissue sections, and as a result the Scala Media was not clearly defined. Scale bars on all images are 250 µm. Please click here to view a larger version of this figure.
Figure 8: Hematoxylin and Eosin staining of control and treated cochleae. Hematoxylin and Eosin stained sections displaying a typical cochlea with native cells present is shown in panel A. Panel B contains the same structures as panel A, but is from a fully decellularized cochlea, which leaves behind the extracellular matrix. The two previously shown cochleae infused with hWJCs are seen in panels C and D. The gross cochlear anatomy remains predominantly unchanged through the decellularization and cell culture processes, although the specific cell populations and locations are dramatically altered. Scala Vestibuli (SV), Scala Tympani (SM), Basilar Membrane (BM), Tectorial Membrane (TM), Spiral Limbus (L), Rosenthal's Canal (RC), Modiolus (M), Spiral Ligament (SL), and Stria Vascularis (*). Scale bars on all images are 250 µm. Please click here to view a larger version of this figure.
A1 Cochlea + Cells | A2 Cochlea + Cells | B2 Neg. Control Cochlea | |
Outside Bony Labyrinth | 3,758 | 549 | 0 |
Bone Outside Cochlea Spaces | 248 | 586 | 0 |
Within Cochlea | 518 | 701 | 0 |
Total Cells | 4,524 | 1,836 | 0 |
Section Volume (μl) | 0.0077 | 0.0100 | 0.0147 |
Percent of Cochlea Volume | 0.46% | 0.59% | 0.87% |
Estimated Cell Density within Cochlea | 113,759 | 118,732 | 0 |
Table 1: Cell Counts. DAPI stained cochlear sections were thresholded and quantified using automatic counting tools in ImageJ in 3 non-overlapping regions of interest (outside the bony labyrinth, temporal bone outside of the cochlear fluid spaces, and within the cochlea). Cross-sectional cochlear area was also measured in ImageJ, and this value was used to calculate section volume in combination with section thickness (10 µm). Using an estimate of total cochlear volume published by Santi et al.23, the relative cochlear volume of a given section was calculated. The cell density of the entire cochlea was estimated using the relative cochlear volume of a given section in combination with the paired cell count.
We have successfully demonstrated that native cochlear cells can be removed from the cochlea via a decellularization process, which allows for the use of the cochlea as an intricate, three-dimensional tissue scaffold. Santi et al.15 developed the initial method for decellularizing cochleae, and have accurately estimated the volumes of many cochlear structures through with the aid of light sheet microscopy23. Such early work served as a strong basis for the tissue engineering and cell culture techniques presented here. Decellularized cochlea can successfully be infused with cells, and then cultured for a prolonged period of time. Cochleae from a wide array of animals could theoretically be utilized in combination with a similarly wide array of cells types. These combinations allow the techniques presented here to be utilized in numerous possible experimental designs, such as tissue engineering applications that examine sensory epithelium development14, drug testing of new pharmaceutical agents24, and development of cochlear repair models25. However, despite the broad applicability of these techniques they are not without limitations.
One of the limitations of the current technique is that the perfusion is variable based on individual. Developing a holder, for the cochlea, and using a syringe pump to conduct perfusions may increase the accuracy and success of the process. Additionally, definitively proving successful decellularization could potentially be accomplished through several methods. We employed DAPI staining that was used to quantify remaining cell nuclei, and we observed no residual nuclei post-decellularization. Additionally, standard DNA quantification assays utilize the reagent, green dye detecting picogram quantities of DNA, to identify nucleic acid remnants, which may visually be more sensitive than DAPI for identifying the presence of small fragments of nucleic acids in decellularized tissues26. Regardless of which techniques are used to validate successful decellularization, it may not be possible to show complete decellularization of an individual cochlea prior to utilizing it as a scaffold. Nevertheless, we have found the decellularization process to be successful and robust.
The two most critical steps in the current protocol are the isolation and handling of the cochleae, and the perfusion of the cochleae. Great care must be taken when the cochlea is initially isolated, as the cochlea is delicate and brittle. If too much force is applied while handling the cochlea with forceps, the cochlea could fragment. Thus, it is imperative to handle each cochlea gently before decellularization and decalcification. The vestibular system, if left intact, serves as a great handle for grabbing the inner ear complex with forceps, and orienting the cochlea. Similarly, when perfusing the cochlea with cells, the pressure behind the perfusion must not be too great, otherwise cells may not survive the perfusion. As noted in the current protocol, attaching tubing to the end of a 1-mL insulin syringe allows for easier handling of the cochlea with one hand, and greater precision in perfusing cells through the cochlea with the other hand.
As a next step, it would be natural to assess developmental and functional markers to determine if the cochlear ECM is inducing differentiation of stem cells, and if so, which specific ECM components in the cochlea are inducing changes in cells. Perfusing different types of cells (e.g., stem cells, neuroprogenitors, fibroblasts, etc.) through the cochlea to monitor cell differentiation and behavior would be another fascinating follow-up investigation. In addition, cells that have been genetically modified or are exposed to a growth factor differentiation protocol may provide new insights into how the cochlear ECM supports neurosensory development, and possibly even function in hearing. Thus, the current protocol is a significant first step in using the cochlear ECM to study inner ear neurosensory development and maintenance, which may one day lead to the development of new therapies to restore hearing loss.
The authors have nothing to disclose.
The current project was funded by the University of Kansas Proof of Concept Fund. We would like to thank the nursing staff at KUMC (Kansas City, KS) for assisting us in obtaining human umbilical cords, and David Jorgensen for assisting with cochleae cultures.
Allegra X-14R Centrifuge | Beckman-Coulter | B08861 | |
Intramedic Semi-Rigid Tubing | Becton Dickinson | 427401 | |
New Brunswick Innova 2000 Orbital Shaler | Eppendorf | M1190-0002 | |
Surgical Scissors | Fine Science Tools | 14060-10 | |
Fine Forceps | Fine Science Tools | 11370-40 | |
Ultra-Fine Forceps | Fine Science Tools | 18155-13 | |
50-mL Conical Tubes | Fisher Scientific | 12565271 | |
Petri Dish | Fisher Scientific | FB087579B | |
U-100 Insulin Syringe | Fisher Scientific | 14-829-1B | |
Scintillation Vial | Fisher Scientific | 03-341-73 | |
Rotator | Fisher Scientific | 88-861-049 | |
Transfer Pipette | Fisher Scientific | 22-170-404 | |
Razor Blade | Fisher Scientific | 12-640 | |
Antibiotic-Antimycotic | Fisher Scientific | 15-240-062 | |
Penicillin-Streptomycin | Fisher Scientific | 15-140-122 | |
24-Well Plate | Fisher Scientific | 07-200-84 | |
SuperFrost PLUS Glass Microscope Slides | Fisher Scientific | 12-550-15 | |
Transfer Pipette | Fisher Scientific | 22-170-404 | |
ProLong Gold Antifade Mountant with DAPI | Fisher Scientific | P36935 | |
Clear-Rite 3 | Fisher Scientific | 22-046341 | |
Thermo Scientific Forma Series II 3110 Water-Jacekted CO2 Incubator | Fisher Scientific | 13-998-078 | |
Mesenchymal Stem Cell Growth Medium | Lonza | PT-3001 | |
Trypsin-EDTA | Lonza | CC-3232 | |
TPP T-75 Culture Flask | MidSci | TP90076 | |
TPP T-150 Culture Flask | MidSci | TP90151 | |
TPP T-300 Culture Flask | MidSci | TP90301 | |
Dissection Microscope | Nikon Instruments | SMZ800 | |
Nikon Eclipse Ts2R-FL Inverted Microscope | Nikon Instruments | MFA51010 | |
NuAire Class II, Type A2 Biosafety Cabinet | NuAire | NU-425-600 | |
1X PBS | Sigma-Aldrich | P5368-10PAK | |
1% SDS Solution | Sigma-Aldrich | 436143-100G | |
10% EDTA | Sigma-Aldrich | E9884-100G |