2Ear Science Institute Australia, Perth, Australia
3Division of Medicine, University College London, UK
4Centre for Cell Therapy and Regenerative Medicine, University of Western Australia, Perth, Australia
Tissue engineering approaches using growth factors and various materials for repairing chronic perforations of the tympanic membrane are being developed, but there are surprisingly few relevant tissue culture models available to test new treatments. Here, we present a simple three-dimensional model system based on micro-dissecting the rat tympanic membrane umbo and grafting it into the membrane of a cell culture well insert. Cell outgrowth from the graft produced sufficient cells to populate a membrane of similar surface area to the human tympanic membrane within 2 weeks. Tissue grafts from the annulus region also showed cell outgrowth but were not as productive. The umbo organoid supported substantial cell proliferation and migration under the influence of keratinocyte growth medium. Cells from umbo grafts were enzymatically harvested from the polyethylene terephthalate (PET) membrane for expansion in routine culture and cells could be harvested consecutively from the same graft over multiple cycles. We used harvested cells to test cell migration properties and to engraft a porous silk scaffold material as proof-of-principle for tissue engineering applications. This model is simple enough to be widely adopted for tympanic membrane regeneration studies and has promise as a tissue-equivalent model alternative to animal testing.
The rapid spontaneous healing of injuries to the tympanic membrane (TM) gives rise to the concept of a robust regenerative mechanism. Studies have localized areas of high cell turnover in the TM to the center of the membrane (the umbo region) and the outer rim (annulus region), which have been suggested to contain stem cells (1,2) or regenerative centers (3,4). However, there is limited understanding of the biology of the cells responsible for regeneration or their niche within the TM, and in vitro models of the TM are currently unavailable.
Tympanic membrane cell cultures have been generated (2,5,6) using trypsin digestion or TM tissue explants in conjunction with selective collagen IV adhesion in vitro. Previously, we showed that TM cells can be isolated via tissue explant culture from humans (5); however, these methods can yield inconsistent results. An opportunity to generate consistent cell cultures arises from a unique aspect of TM biology: large numbers of cells are produced at the umbo, which then migrate en masse across the TM surface and along the ear canal to a point where they are ultimately shed externally. An ink dot placed on the keratinized surface at the TM umbo will migrate to the peripheral margin of the membrane and then along the ear canal; the pattern and rate of migration in vivo has been described for various species (7-10). This property of the TM helps maintain a clear ear canal and a minimal TM keratin burden for sensitive transmission of sound through the middle ear. This also has a profound effect on repair mechanisms, which in contrast to other skin-like tissues, is led by keratinocyte layer responses rather than granulation tissue formation (11-13).
Here, we present a tympanic membrane organoid explant culture using a tissue culture insert membrane that allows rapid isolation of tympanic membrane primary cells for biocompatibility testing and tissue engineering.
We have used this property of the TM epidermis to generate cell cultures from umbo organoids placed in a collecting membrane. Here, we describe our method and its use to generate a TM tissue–equivalent model as well as a tissue-engineered TM construct in vitro using state-of-the-art scaffolds. Materials and methods Animals
Male Sprague-Dawley rats, weighing 250–300 g, were obtained from the Animal Resources Centre (Murdoch, Western Australia, Australia). This study was approved by the University of Western Australia Animal Ethics Committee (No. 100/1249). The experiments were performed in accordance with the Australian Code of Practice for the Care and Use of Animals for Scientific Purposes. Extraction of tympanic membrane
The animals (n = 4) were euthanized by intraperitoneal administration of pentobarbitone (160 mg/kg). Using en bloc excision, the left and right external ears were removed at the osteocartilaginous junctions, and the tympanic bulla was isolated from each. The TM was then dissected free. The harvested TMs were rinsed in phosphate buffered saline (PBS) pH 7.2 containing 2% antibiotic-antimycotic solution (Gibco, Grand Island, NY) for 2 min prior to processing. Enzymatic digestion
For tissue culture, freshly extracted temporal bones were rinsed thoroughly in 2 mL PBS and transferred to a 35-mm glass-bottom culture dish (Greiner Bio-One) containing 2 mL 0.25% Dispase II solution (Gibco). Following overnight incubation at 4°C, the TMs were peeled from the bony bulla and separated from the ear canal skin using forceps. Rat TM explant culture on culture well insert membrane
A culture well insert with a polyethylene terephthalate (PET) membrane (0.4 μm pore size, 12-well Thincert; Greiner Bio-One, Frickenhausen Germany) was rinsed in PBS, and a 1 mm scalpel incision was made in the center. The malleus bone attached to the TM umbo was then partially inserted into the incision with the outer (lateral) surface of the umbo facing vertically and close to the level of the upper surface of the PET membrane. A length of annulus tissue was dissected from the pars tensa and similarly placed through a PET membrane incision. The inserts were then mounted into 12-well plates and filled with 1 mL keratinocyte growth media: keratinocyte serum-free medium (KSFM) supplemented media with human recombinant epidermal growth factor (EGF) a.a. 1–53, bovine pituitary extract, 10% fetal bovine serum (FBS) (Bovogen, East Keilor, Vic, Australia) and 1% penicillin-streptomycin (KSFM and the additives other than FBS were purchased from Gibco). Explant cultures were incubated at 37°C in 5% CO2 with twice-weekly medium changes for up to 2 weeks. Culture expansion
Cell outgrowth from the explant onto the upper surface of the PET membrane was allowed to continue until a monolayer of cells approached the periphery of the culture well insert membrane, after which the cells were passaged by enzymatic digestion into fresh culture dishes. Briefly, the cell monolayer was rinsed thoroughly with PBS and digested with 250 μL pre-warmed Tryple Express (Gibco) for 15 min at 37°C. Digestion was stopped by adding 2 mL KSFM containing 10% FBS. The supernatant containing the detached cells was collected and centrifuged at 600 × g for 5 min and cells were resuspended in 5 mL of medium before replating in a 25 cm2 cell culture flask (Greiner Bio-One). The cells were cultured in KSFM supplemented with 10% FBS and fed with fresh medium twice weekly. Upon reaching 60%–70% confluence, cells were then passaged into fresh dishes at a 1:3 ratio. The explant and membrane construct was kept intact, and fresh culture medium was added to investigate continued outgrowth. Immunofluorescent staining for characterization of the primary cells
Cells were cultured in 35-mm glass-bottom culture dishes (Greiner Bio-One) for 2–3 days at 37°C before staining. The cells were rinsed with PBS and fixed with 4% paraformaldehyde in PBS for 10 min at RT. They were then incubated with ice-cold 100% methanol for 10 min at -20°C before blocking with 5% BSA in PBS for 30 min at RT. The primary antibodies used for staining were mouse anti-pan cytokeratin (Biocare, Concord, CA) and anti-vimentin (Abcam, Cambridge, UK). Following an overnight incubation at RT, the cells were rinsed 3 times with PBS containing 0.01% Tween 20 before the addition of secondary antibodies: anti-rabbit IgG Alexafluor 555 (Abcam) or anti-mouse IgG Alexafluor 488 (Molecular Probes, Eugene, OR. The cells were incubated for 1 h at RT then rinsed in PBS/0.01% Tween 20 and counterstained with 4’,6-diamidino-2-phenylindole (DAPI) (0.05 mg/mL, Molecular Probes) for 20 min. The cells were mounted in PBS:glycerol (1:1) before viewing with with a fluorescence microscope (Olympus, Hachioji-shi, Tokyo, Japan) with appropriate fluorescence filters. Epidermal cell enrichment using timed trypsinization
Homogeneous epidermal cell cultures were obtained by carefully timed trypsinization. Confluent cells grown in a 25 cm2 cell culture flask were rinsed once in PBS and incubated with 2 mL pre-warmed Tryple Express. Cells were incubated at 37°C for 5–10 min, and the flask was gently tapped to detach the fibroblasts. Enzyme solution and suspended fibroblasts were discarded or cryopreserved for later use. The remaining epidermal colonies were fed with fresh KSFM containing 10% FBS and passaged as previously described upon reaching 70%–80% confluence. Wound-healing migration assay
Epidermal cells [passage 4–5 (p4–5)] were seeded (5 × 104 cells) into the reservoirs of a 2-well culture insert (Ibidi, Martinsried, Germany), adhered to a 35-mm glass-bottom dish, and grown overnight at 37°C. On the day of assay, the insert was removed using a pair of forceps, and the dish was briefly rinsed with PBS to remove cell debris. The dish was then filled with 1 mL KSFM containing 10% FBS and placed on a stage top incubator (Tokai Hit, Fujinomiya, Japan). Phase-contrast images were viewed and captured at 10× objective magnification using an Olympus IX-81 microscope system at 15 min intervals over 18 h. Tissue engineering using cells isolated from membrane cultures
TM cells at p2 were resuspended in KSFM (5 × 105 cells/mL) and seeded onto silk fibroin–coated coverslips in a 12-well culture plate. The scaffold was produced by immersing 13 mm borosilicate coverslips in 50 mg/mL silk fibroin solution (Advanced Biomatrix, Carlsbad, CA). The coated coverslips were immediately immersed into liquid nitrogen to achieve thermally induced phase separation. The frozen coverslips were placed into an Edwards Modulyo freeze drier (Edwards Ltd, Crawley, UK) and lyophilized for 24 h. The porous silk films were sterilized in 70% ethanol overnight before conditioning for 2 h in growth medium containing 10% FBS. Once seeded with cells, scaffolds were incubated at 37°C with 5% CO2 for 24 h, then rinsed twice in PBS, fixed in 4% paraformaldehyde for 10 min, and stained with DAPI for 20 min before visualizing with an epifluorescence microscope as described above. Results and discussion
TM regeneration is of considerable general interest due to its robust and relatively scarless healing compared with skin; however, it is most specifically relevant in otology due to chronic failure of wound healing after TM perforation and to the serious consequence of invasive growth of TM cells into the middle ear and mastoid cholesteatoma (14,15).
There are few models for exploring the biology of TM cells (2,6), and those currently available are limited by the very small amount of tissue and cells available. One approach we explore here is to establish a cell culture model that provides access to TM cells and allows exploration of their properties in relation to growth and use for tissue engineering. A flow diagram for this method of culturing rat TM umbo explants on culture well insert membranes is shown in Supplementary Figure S1. Rat TM was dissected as a thin sheet suspended in a shell-like bony ring structure, the tympanic ring. Figure 1A shows a TM peeled whole off the bone and still attached to the malleus after overnight digestion in 0.25% Dispase II. The annulus of the membrane was retained as a prominent fibrous ring, and the thin and transparent pars tensa was intact within it. As shown in Figure 1B, the umbo could be dissected free from the whole TM and the handle of the malleus removed, leaving only a remnant at the umbo. This method of rat TM extraction is superior to in situ harvest or other previously described methods (2,6) as it preserves the normal anatomical relationships of TM and, more importantly, enables cell isolation from specific areas of the TM (umbo and annulus).