This methodology aims to illustrate the mechanisms by which extracellular matrix cues such as substrate stiffness, protein composition and cell morphology regulate Schwann cell (SC) phenotype.
Traumatic peripheral nervous system (PNS) injuries currently lack suitable treatments to regain full functional recovery. Schwann cells (SCs), as the major glial cells of the PNS, play a vital role in promoting PNS regeneration by dedifferentiating into a regenerative cell phenotype following injury. However, the dedifferentiated state of SCs is challenging to maintain through the time-period needed for regeneration and is impacted by changes in the surrounding extracellular matrix (ECM). Therefore, determining the complex interplay between SCs and differing ECM to provide cues of regenerative potential of SCs is essential. To address this, a strategy was created where different ECM proteins were adsorbed onto a tunable polydimethylsiloxane (PDMS) substrate which provided a platform where stiffness and protein composition can be modulated. SCs were seeded onto the tunable substrates and critical cellular functions representing the dynamics of SC phenotype were measured. To illustrate the interplay between SC protein expression and cellular morphology, differing seeding densities of SCs in addition to individual microcontact printed cellular patterns were utilized and characterized by immunofluorescence staining and western blot. Results showed that cells with a smaller spreading area and higher extent of cellular elongation promoted higher levels of SC regenerative phenotypic markers. This methodology not only begins to unravel the significant relationship between the ECM and cellular function of SCs, but also provides guidelines for the future optimization of biomaterials in peripheral nerve repair.
Peripheral nervous system (PNS) injuries remain a major clinical challenge in healthcare by compromising the quality of life for patients and creating a significant impact through a multitude of socioeconomic factors1,2. Schwann cells (SC), as the major glial cells in the PNS, provide necessary molecular and physical cues to induce PNS regeneration and aid in functional recoveries in short gap injuries. This is due to the remarkable ability of SCs to dedifferentiate into a “repair” cell phenotype from a myelinating or Remak phenotype3. The repair SC is a distinctive cell phenotype in several ways. Following injury, SCs increase their proliferation rate by re-entering the cell cycle and begin expression of several transcriptional factors to facilitate reinnervation. These factors, such as c-Jun and p75 NTR, are upregulated while myelinating SC markers, such as myelin basic protein (MBP), are downregulated4,5. In addition, SCs change morphology to become elongated and aligned with each other to form Büngner bands across the injury site6. This provides a physical guidance mechanism for the axons to extend to the correct distal target7. However, despite the ability that SCs possess to promote nerve regeneration in short gap injuries, the outcome of functional recovery remains poor in severe injuries. This is due in part to a loss of extracellular matrix (ECM) guidance cues, as well as the inability of SCs to maintain the regenerative phenotype over long periods of time8.
The nerve regeneration and recovery process are intimately tied to the state of the basal lamina following injury. The basal lamina is a layer of ECM around the nerve that facilitates guidance and provides ECM-bound cues for axons and SCs in cases where it remains intact following injury9. The state of the ECM and its ability to deliver matrix bound cues to cells is vitally important and has been previously explored in a variety of different contexts10,11,12,13,14. For example, it has been shown that the stiffness of the ECM can guide cell functions such as proliferation and differentiation11,15,16. Composition of the ECM can also lead to a distinct cellular response and regulate cell behaviors such as migration and differentiation through intracellular signaling pathways17,18. Furthermore, cell morphology, including spreading area and cellular elongation, play a major role in regulating the function and can be governed by ECM-bound cues19,20. Many previous studies have focused on stem cells differentiating into defined lineages, yet SCs possess a similar ability to alter phenotype from a homeostatic, adult SC within a healthy nerve, to a repair SC capable of secreting proteins and growth factors while remodeling the ECM following nerve injury5,21. Therefore, it is especially crucial to identify mechanisms underlying the relationship between the innate SC regenerative capacity and ECM bound cues for the insight to ultimately harness this capacity for nerve regeneration.
To address this, we have developed a detailed methodology to produce a cell culture substrate where mechanical stiffness and ligand type can be easily tuned in physiologically relevant ranges. Polydimethyl siloxane (PDMS) was chosen as a substrate due to its highly tunable mechanics as compared to polyacrylamide gel, where the maximum Young’s modulus is around 12 kPa contrasted to PDMS at around 1000 kPa22,23,24. This is beneficial to the work at hand, as recent studies have shown the Young’s modulus of a rabbit sciatic nerve can exceed 50 kPa during development, thereby suggesting that the range of stiffness of nerves within the PNS is wider than previously examined. Different proteins are capable of adsorption onto PDMS substrates to analyze the combinatorial regulation of mechanics and ligands on SC behavior. This allows for the investigation of multiple microenvironmental cues present in the PNS regeneration process and comparison of a high degree of tunability to the work focusing solely on stiffness of the substrate25. Further, these engineered cell culture substrates are compatible with a multitude of quantitative analysis methods such as immunohistochemistry, western blot, and quantitative polymerase chain reaction (q-PCR).
This engineered cell culture platform is highly suitable for analyzing mechanistic pathways due to the high level of individual tunability of each ECM-bound signal. In addition, popular methods for cell micropatterning, including microcontact printing, can be achieved on the substrates to allow for controlled cellular adhesion to analyze cell shape in relation to other ECM bound cues24. This is critical because line patterned substrates, which promote elongation in cell populations, provide a tool to mimic and study elongated and regenerative SCs within Büngner bands during nerve regeneration. Further, cellular morphology is a potent regulator of multiple cell functions and can potentially introduce confounding experimental results if not controlled26,27. Significant attention is now being provided to the mechanisms governing the SC regenerative phenotype as regulated by ECM cues28,29,30. This is essential to provide insight into the design of biomaterials that can be applied as nerve guidance conduits for aid in PNS nerve regeneration. These detailed protocols can ultimately be applied as a potential tool to decipher the mechanisms of SC and other cell type function as regulated by ECM bound cues.
1. Tunable cell culture substrate preparation and characterization
2. Quantification of cellular properties on tunable substrates
To analyze and quantify the interplay between substrate stiffness and protein composition on SC phenotype, a tunable PDMS cell culture substrate was developed (Figure 1A). Compression testing of the polymer at differing base: curing agent ratios was utilized to quantify the Young’s modulus (E) of the substrate (Figure 1B). The resulting range of modulus values represents physiologically relevant substrate conditions. Following preparation of substrates, SCs were cultured and cellular properties analyzed on the tunable microenvironment. Proliferation rates of SCs on substrates of differing protein composition were first analyzed. Laminin coated substrates resulted in a higher proliferation rate when compared to collagen I and fibronectin adsorption all at 10 µg/mL (Figure 2A,B). SCs on substrates with laminin coating and differing moduli showed that relatively softer substrates (E=3.85kPa) decrease cell proliferation rates across all conditions (Figure 2C,D). However, differences between stiff substrates (E=1119kPa) and relatively soft substrates (E=8.67kPa) were insignificant (Figure 2D).
SCs were also analyzed for protein expression through immunohistochemistry and western blot. Levels of transcriptional factor c-Jun were analyzed by immunofluorescent microscopy (Figure 3A) and represented by the mean pixel fluorescent intensity (Figure 3B-D). c-Jun expression was shown to be upregulated as substrates became softer (E=1119 kPa to E=8.67 kPa), however, on the softest substrates (E=3.85 kPa), c-Jun expression was significantly downregulated. On stiff substrates (E=1119 kPa), collagen I coated substrates resulted in the highest c-Jun expression, yet as substrates became softer (E=8.67 kPa and 3.85 kPa), laminin showed the highest levels of c-Jun (Figure 3E). Western blot was also used to analyze both c-Jun and myelin basic protein (MBP) with c-Jun levels upregulated and MBP downregulated on softer substrates (Figure 3F). Further, SCs seeded on laminin coated substrates resulted in the highest c-Jun expression when compared to collagen I and fibronectin.
Cells of differing seeding densities were then cultured and stained with rhodamine-phalloidin to explore the role of cell spreading and area in c-Jun expression (Figure 4A,B). To control nuclear elongation of cells, a common micropatterning technique (microcontact printing36) was utilized to create cell adhesive lines on the cell culture substrates. The nuclear aspect ratio of cells seeded on a line patterned substrate was shown to be significantly higher than cells seeded on unpatterned substrates (Figure 4C,D). It was found that expression of both c-Jun and another marker significant in SC regenerative phenotypes, p75 neurotrophin receptor (p75 NTR), were upregulated in dense cells with a smaller spreading area (Figure 4E). Line-patterned cells also resulted in a higher expression of both c-Jun and p75 NTR when compared to nonpatterned cells (Figure 4F). Therefore, microcontact printed cell adhesive geometries were created to precisely control cell spreading area and elongation while eliminating cell-cell interactions (Figure 5A). Dimensions of total cell adhesive areas were 900 µm2, 1,600 µm2, and 2,500 µm2 with an aspect ratio of either 1 or 4 (cell length: cell width). Fluorescent bovine serum albumin (fBSA, Texas red) staining was used to reveal the status of micropatterns on cell culture substrate after microcontact printing (Figure 5B). The nuclear aspect ratio of SCs for each cell area was measured and it was shown that by increasing the nuclear aspect ratio, the cellular elongation increases (Figure 5C). In addition, as SC aspect ratio increased, c-Jun was upregulated (Figure 5D). Interestingly, however, it was found that as cell spreading area increases c-Jun expression is downregulated (Figure 5E). Immunofluorescence staining for both nuclei and actin confirmed cell spreading area and elongation were highly controlled through this micropatterning method (Figure 5F).
Figure 1: Cell culture substrates with tunable stiffness and protein composition. (A) Schematic showing the development of PDMS cell culture substrates. (B) The initial PDMS mixing ratio of base: curing agent determines Young’s modulus. Please click here to view a larger version of this figure.
Figure 2: SC proliferation rates regulated by substrate stiffness and protein composition. (A) Representative images showing BrdU staining when cultured on substrates of the same modulus. (B) Histogram showing the percentage of BrdU positive cells for each protein coating. (C) Representative images showing BrdU incorporation in SCs seeded on substrates of the same protein coating. (D) Histogram showing the percentage of BrdU positive cells for each Young’s modulus value. Scale bars = 50 μm. Data are presented as mean ± SEM. *p < .05, **p < .005, ***p < .0005. Portions of the figure have been modified from ref.24. Please click here to view a larger version of this figure.
Figure 3: Substrate stiffness and protein regulated SC protein expression. (A) Representative images show c-Jun immunofluorescence staining for SCs seeded on substrates of different stiffness and protein composition. Mean pixel fluorescent intensity of c-Jun was measured for SCs seeded on (B) collagen I (C) fibronectin and (D) laminin coated substrates of different stiffness. (E) c-Jun fluorescence level of SCs grouped by Young’s modulus of substrate. (F) Western blot showing c-Jun and myelin basic protein (MBP) of SCs seeded on substrates. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as loading control. Scale bar = 50 μm. Data are presented as mean ± SEM. *p < .05, **p < .005, ***p < .0005. Portions of the figure have been modified from ref.24. Please click here to view a larger version of this figure.
Figure 4: Cellular spreading area influences protein expression of SCs. (A) Cell spreading area of different seeding densities was visualized through rhodamine-phalloidin (red) and nuclei staining (blue). (B) Histogram showing the average spreading area of SCs in each condition. (C) The nuclei of SCs seeded on unpatterned or line-patterned substrates were stained with DAPI (blue) to show morphology. (D) Histogram showing quantification of nuclear aspect ratio on patterned and unpatterned substrates. (E) Western blot showing expression of c-Jun and p75NTR of cells with different spreading area. (F) Western blot showing the protein expression of SCs on unpatterned and line-patterned substrates. Scale bar = 50 μm. Data are presented as mean ± SEM. *p < .05, **p < .005, ***p < .0005. Please click here to view a larger version of this figure.
Figure 5: SC morphology and elongation impact c-Jun expression of SCs. (A) Schematic showing cell micropatterning for shapes of different aspect ratio. (B) fBSA staining (red) showing the shape of micropatterns following microcontact printing. Scale bar = 10 μm. (C) Histogram showing nuclear aspect ratio of micropatterned SCs. (D, E) Histogram showing the mean pixel fluorescent intensity of c-Jun for each of the geometrical conditions. (F) Rhodamine-phalloidin (red), nuclei (blue) and c-Jun (green) were stained on the differing micropatterns. Scale bar = 10 μm. Data are presented as mean ± SEM. *p < .05, **p < .005, ***p < .0005. Portions of the figure have been modified from ref.24. Please click here to view a larger version of this figure.
SCs can promote nerve regeneration due to their phenotypic transformation and regenerative potential following nerve injury. However, how ECM cues regulate this regenerative capacity remains mostly unclear, potentially hindering not only the development of biomaterials that aim to promote nerve regeneration but also the understanding of the mechanisms involved in nerve regeneration. To begin to examine this interplay, cell culture substrates were created where ECM cues such as stiffness, protein coating, and adhesive topography can be controlled. The ability to micropattern adhesive topography is a key feature within the protocol, utilizing the common method of microcontact printing. However, this substrate is different from the glass in that the compression force being applied to the PDMS stamps must be at an appropriate level to achieve the desired shape of cell adhesive areas on cell culture substrates. Utilizing fluorescent bovine serum albumin (fBSA) as a model protein to visualize cell adhesive areas and adjust the compression forces on the PDMS stamps can ultimately mitigate this issue. Another key difference from glass is the removal of excess Pluronic F-127 that is used to render the remainder of the substrate non-adhesive. After this treatment, cell culture substrates are highly hydrophobic, making it challenging to maintain wet cell culture substrates throughout washes, which is important for the structural integrity of micropatterned protein37. Therefore, it is recommended to use multiple pipettes to aspirate and inject solutions nearly simultaneously to prevent substrates from completely dehydrating.
Although micropatterning can precisely control cell shape and does not require complicated synthetic procedures using cell blocking polymers such as OEGMA, the methods used to quantify protein expression of micropatterned cells are sometimes limited38. For example, samples that are available from micropatterning are generally limited to a few hundred cells per coverslip, which is inadequate to prepare cell lysates to be used for western blots or qPCR. Considering this, immunofluorescence staining alone was used to quantify protein expression for micropatterned cells, limiting the variety of proteins that can be quantified. To address this, we utilized line patterns to promote cell elongation for larger cell populations and successfully prepared cell lysates that were analyzed by western blot. Other methods such as applied electric fields or aligned electrospun fibers can also be applied to control the elongation of cell populations39,40. However, electric fields may not fit into all applications for NGCs since the conductivity of commonly used biomaterials such as collagen-based polymer and silk is very low28,41,42. In contrast, aligned electrospun nanofibers have been successfully implanted into NGCs to promote SC alignment, elongation and migration as well as neurite outgrowth43,44. It may also prove compelling to compare SC behavior on line-patterned substrates against those on substrates with aligned nanofibers, since lined patterns and aligned nanofibers are two of the most common guidance mechanisms incorporated into NGCs45.
The micropatterning protocol detailed utilizes PDMS coated coverslips as cell culture substrates, with a maximum surface Young’s modulus of 1119 kPa. Such stiffness mimics many tissue, yet it may not be possible to model osteogenesis of mesenchymal stem cells, which generally require the surface Young’s modulus to exceed 1 Gpa46. For such situations, glass is an alternative candidate, however the adsorption of Pluronic F-127 requires relatively high surface hydrophobicity which glass does not possess. To increase hydrophobicity, glass can be treated with dimethyl dichlorosilane in dichlorobenzene. Following this, UV-Ozone treatment can be used to increase hydrophilicity for microcontact printing47.
Ultimately, a cell culture platform was developed where ECM stimuli can be tuned individually with protein expression quantified. We have determined that SC regenerative capacity is promoted by certain mechanical and chemical ECM cues, which can subsequently provide inspiration in the future design of biomaterial applications such as NGCs and cell transplantation processes where ECM features may need to be optimized24. Nonetheless, tuning these ECM cues can be a challenging undertaking, particularly in vivo. Moving forward, this platform can be utilized to parse key mechanisms involved in the phenotypic transition of SCs as regulated by the ECM. By achieving this, manipulation of intracellular cues may be possible to promote SC regenerative capacity without the need for dedicated platforms in vitro48,49. This has the potential for groundbreaking work in the development of technologies for nerve repair.
The authors have nothing to disclose.
The authors gratefully acknowledge funding support from the University of Cincinnati. The authors also thank Ron Flenniken of the University of Cincinnati Advanced Materials Characterization laboratory for support.
Albumin from Bovine Serum (BSA), Texas Red conjugate | Thermo Fisher Scientific | A23017 | BSA staining to show micropatterns |
Anti-mouse IgG, HRP-linked Antibody | Cell Signaling Technology | 7076S | Antibody used for western blot analysis |
Anti-rabbit IgG, HRP-linked Antibody | Cell Signaling Technology | 7074S | Antibody used for western blot analysis |
BrdU | Thermo Fisher Scientific | B23151 | Reagent used to measure cell proliferation |
BrdU primary antibody conjugated with Alexa Fluor 488 | Thermo Fisher Scientific | B35130 | Used to visualize BrdU in cell proliferation assays |
Collagen I | Thermo Fisher Scientific | A10483-01 | Protein used to coat coverslips |
Compression force test machine | TestResources | Instrument to quantify mechanical properties of polymers | |
Dulbecco's Modified Eagle Medium | Thermo Fisher Scientific | 11965092 | Cell culture medium |
Fetal Bovine Serum | Thermo Fisher Scientific | 16000044 | Cell culture medium supplemental |
Fibronectin | Thermo Fisher Scientific | 33010-018 | Protein used to coat coverslips |
Fluorescence microscope | Nikon | Eclipse Ti2 | Fluorescence microscope |
Halt Protease and Phosphatase Inhibitor Cocktail (100X) | Thermo Fisher Scientific | 78440 | Protease and Phosphatase Inhibitor |
Laminin | Thermo Fisher Scientific | 23017015 | Protein used to coat coverslips |
Mounting medium with DAPI | Thermo Fisher Scientific | P36971 | Coverslip mountant and nuclei staining |
Mouse c-Jun primary antibody | Thermo Fisher Scientific | 711202 | Primary antibody to visualize c-Jun protein |
Mouse β-Actin primary antibody | Cell Signaling Technology | 3700S | Loading control for western blot experiments |
Penicillin-Streptomycin | Thermo Fisher Scientific | 15140122 | Cell culture medium supplemental |
Photoresist SU 2010 | KAYAKU | SU8-2010 | Photoresist |
Pluronic F-127 | Sigma Aldrich | P-2443 | Block non-specific protein binding |
Rabbit c-Jun primary antibody | Cell Signaling Technology | 9165S | Primary antibody for visualization of c-Jun protein |
Rabbit myelin basic protein primary antibody | Abcam | ab40390 | Primary antibody for visualization of MBP |
Rabbit p75NTR primary antibody | Cell Signaling Technology | 8238S | Primary antibody for visualization of p75NTR |
Rhodamine phalloidin | Thermo Fisher Scientific | R415 | Visualization of cell cytoskeleton |
RIPA buffer | Abcam | ab156034 | Cell lysis buffer |
RT4-D6P2T Schwann cell line | ATCC | CRL-2768 | Cell line used in experiments |
SYLGARD 184 PDMS base and curing agent | Sigma Aldrich | 761036 | Tunable polymer used to coat coverslips |
Trypsin | Thermo Fisher Scientific | 15090-046 | Cell dissociation reagent |
UV-Ozone cleaner | Novascan | Increase hydrophicility of PDMS | |
Versene (1x) | Thermo Fisher Scientific | 15040066 | Cell dissociation reagent |