This protocol combines electrospinning and microspheres to develop tissue engineered scaffolds to direct neurons. Nerve growth factor was encapsulated within PLGA microspheres and electrospun into Hyaluronic Acid (HA) fibrous scaffolds. The protein bioactivity was tested by seeding the scaffolds with primary chick Dorsal Root Ganglia and culturing for 4-6 days.
This procedure describes a method to fabricate a multifaceted substrate to direct nerve cell growth. This system incorporates mechanical, topographical, adhesive and chemical signals. Mechanical properties are controlled by the type of material used to fabricate the electrospun fibers. In this protocol we use 30% methacrylated Hyaluronic Acid (HA), which has a tensile modulus of ~500 Pa, to produce a soft fibrous scaffold. Electrospinning on to a rotating mandrel produces aligned fibers to create a topographical cue. Adhesion is achieved by coating the scaffold with fibronectin. The primary challenge addressed herein is providing a chemical signal throughout the depth of the scaffold for extended periods. This procedure describes fabricating poly(lactic-co-glycolic acid) (PLGA) microspheres that contain Nerve Growth Factor (NGF) and directly impregnating the scaffold with these microspheres during the electrospinning process. Due to the harsh production environment, including high sheer forces and electrical charges, protein viability is measured after production. The system provides protein release for over 60 days and has been shown to promote primary nerve cell growth.
One of the ongoing challenges in neural tissue engineering is creating a nerve conduit (NC) that mimics the extra cellular matrix, where nerves grow naturally. Research has shown that cells respond to several factors in their environment including mechanical, topographical, adhesive and chemical signals1-3. One of the primary challenges in this field is determining the appropriate combination of signals and fabricating a system that can maintain cues for an extended period to support cell growth4. Peripheral neurons are known to prefer a soft substrate, be directed by aligned fibers, and respond to nerve growth factor (NGF)5-7. NCs that can provide chemical cues for weeks have been shown to provide improved functional recovery closer to that of allografts, the current gold standard for nerve repair8,9.
Various materials and production methods can be used to produce mechanical and topographical cues10-13. Mechanical cues are inherent to the material chosen, making selection of the appropriate material for the application critical1,13. Production methods to control topographical cues include phase separation, self-assembly and electrospinning1,13. For microscale applications, microfluidics, photopatterning, etching, salt leaches, or gas foams can also be used14-17. Electrospinning has emerged as the most popular way to engineer fibrous substrates for tissue culture due to its flexibility and ease of production13,18-23. Electrospun nanofibers are fabricated by applying a high voltage to a polymer solution causing it to repel itself and stretch across a short gap to discharge24. An aligned scaffold can be created by collecting the fibers on a grounded rotating mandrel and nonaligned scaffolds are collected on a stationary plate25. Adhesion signaling can be achieved by coating the fibrous scaffold with fibronectin or conjugating an adhesion peptide, such as RGD, to the HA before electrospinning26.
Chemical signals, such as growth factors, are the most difficult to maintain over extended periods because they need a source for controlled release. Many systems have been attempted to add controlled release to electrospun fibrous networks with varying levels of success. These methods include blend electrospinning, emulsion electrospinning, core shell electrospinning and protein conjugation27. Additionally, electrospinning is traditionally done in a volatile solvent, which can affect the viability of the protein28, therefore maintaining bioactivity of the protein must be considered.
This approach specifically addresses combining mechanical, topographical, chemical and adhesive signals to create a tunable scaffold for peripheral nerve growth. Scaffold mechanics is precisely controlled by synthesizing methacrylated Hyaluronic Acid (HA). The methacrylation sites are used to attach photo reactive crosslinkers. The crosslinked material is no longer water soluble and is exclusively broken down by enzymes29. The amount of crosslinking changes the degradation rate, mechanics and other physical properties of the material. Using HA with 30% methacrylation, which has a tensile modulus of ~500 Pa, creates a soft substrate that is close to the native mechanics of neural tissue and is typically preferred by neurons26,29. Electrospinning on a rotating mandrel is used to create aligned fibers for a topographical cue. Using electrospinning along with microspheres provides chemical signals within the scaffold over extended periods. To support neurite growth microspheres containing NGF are used to create the chemical signal. Unlike most electrospun materials HA is soluble in water so the NGF does not encounter harsh solvents during production. To add an adhesive signal, the scaffold is coated with fibronectin. The completed system contains all four types of signals described above: soft (mechanical) aligned (topographical) fibers with NGF releasing microspheres (chemical) coated with fibronectin (adhesive). Production and testing of this system is described in this protocol.
The process begins with the production of the microspheres with a Water-in-Oil-in-Water Double Emulsion. The emulsion is stabilized with a surfactant, Polyvinyl Alcohol (PVA). The inner water phase contains the protein. As it is added to the oil phase, containing the PLGA shell material dissolved in Dichloromethane (DCM), the surfactant creates a barrier between the phases protecting the protein from the DCM. This emulsion is than dispersed in another water phase containing PVA to create the outer surface of the microspheres. The stable emulsion is stirred to allow the DCM to evaporate. After rinsing and lyophilizing you are left with the dry microspheres containing the protein.
After the microspheres are completed they are ready to be electrospun into scaffolds. First you prepare the electrospinning solution. The viscosity of the solution is critical to proper fiber formation. Solutions of pure HA do not meet this requirement; PEO is added as a carrier polymer to allow for electrospinning. The microspheres are added to the solution and electrospun resulting in a fibrous scaffold with microspheres distributed throughout.
Once the production is complete, the protein should be tested to verify its viability. To do this, a primary cell that responds to NGF can be used. This protocol uses Dorsal Root Ganglia (DRG) from 8-10 day old chicken embryos. The cell bundles are seeded onto scaffolds containing microspheres filled with NGF or ones that are empty. If the NGF is still viable you should see enhanced neurite growth on the NGF containing scaffolds. If the NGF is no longer viable it will not promote neurites to extend and should appear similar to the control.
The exact procedure described herein is focused on neural support, however, with simple modifications to the material, electrospinning method, and proteins the system can be optimized for various cell and tissue types.
1. Water/Oil/Water Double Emulsion Microsphere Production
Table 1: Example Protein Solutions. The following protein solutions have been successfully encapsulated and electrospun using this protocol. Other hydrophilic protein solutions can be used as needed.
Note: To visualize the protein location in the microsphere add Rhodamine 2 µg/ml to the PLGA solution31 and encapsulate a FITC conjugated protein. Figure 1 shows an example.
2. Electrospinning with Microspheres
3. Protein Bioactivity Testing
Microspheres 50±14 µm in diameter with an over 85% protein encapsulation have been consistently produced and electrospun into scaffolds. Size was determined by imaging samples of microspheres from three separate production batches. The images where captured on an optical microscope and lengths where measured using commercial lab software. Figure 1 shows a histogram of the size distribution. Encapsulation rate was also tested from three separate microsphere batches, by quantifying the protein that escaped during the production process.
Figure 2 shows representative microspheres with Rhodamine (2 µg/ml) in the PLGA shell and bovine serum albumin (BSA)-FITC encapsulated within. The images were taken using a fluorescent microscope. The shell can be clearly seen (red) with the protein concentrated on the interior (green).
To evaluate encapsulation, the microspheres were filled with BSA and tested for protein release with a Bradford Protein Assay. PLGA breaks down by hydrolysis of ester bonds, creating larger gaps for protein escape. The microspheres continue releasing for over 60 days (Figure 3). The release begins with an initial burst then continues as microspheres break down.
After electrospinning the microspheres can be seen throughout the 3D fibrous structure. Figure 4 shows an SEM image of the complete scaffold where the spheres can be seen in multiple layers (indicated by arrows). Distribution of microspheres in the scaffolds was measured to be 15.3 ± 3.6 spheres/mm2. The scaffold holds the microspheres in place preventing migration away from the target site. As the microspheres break down they release NGF into the scaffold to encourage neurite growth33,34. This creates a continual delivery of protein throughout the scaffold to support the growth of cells and encourage cells to infiltrate into the scaffold.
Finally, Dorsal Root Ganglia (DRG) were used to test the viability of NGF in the microspheres, within the scaffold. Figure 5 shows a DRG seeded on a scaffold containing microspheres loaded with NGF next to one containing microspheres with no protein in them. There are longer neurite extensions visible extending from the DRG on the NGF scaffold, indicating that the NGF remained viable and is stimulating growth.
Figure 1. Microsphere size distribution. Microspheres produced by this protocol are 50±14 µm in diameter. The graph shows a histogram of the microsphere diameter in 5 µm bins.
Figure 2. Fluorescent Image of Microspheres. 2 µg/ml Rhodamine was included in the PLGA solution to visualize the microsphere shell (red) and BSA-FITC was encapsulated to visualize the protein (green). Protein can clearly be seen within the sphere. Scale bar = 20 µm. Please click here to view a larger version of this figure.
Figure 3. Protein Release Curve. BSA release from the microspheres over 60 days was measured using a Bradford Protein Assay. The initial burst is likely due to BSA that was attached to the outer surface. As the PLGA breaks down, protein is released over time.
Figure 4. Scaffold containing Microspheres. The electrospinning process allows microspheres to be incorporated throughout the depth of the scaffold. (A) Arrows indicating the location of microspheres. Scale bar = 500 µm. (B) High magnification image of one microsphere with the nanofibers from the scaffold above it. Scale bar = 20 µm. Please click here to view a larger version of this figure.
Figure 5. NGF Bioactivity Test. Dorsal Root Ganglia primary cells were harvested from 8 day old chick eggs and seeded on scaffolds with NGF filled microspheres (A) or empty microspheres (B). DRGs were stained with a neurofilament antibody, FITC secondary antibody and DAPI to visualize the cell nuclei. Images show increased neurite outgrowth in the presence of NGF loaded microspheres as indicated by FITC stained neurites (green). This shows released NGF is viable and can promote growth. Images were taken with a confocal microscope. Scale bar = 200 µm. Please click here to view a larger version of this figure.
Many studies have shown that nerve cells can be directed by topographical cues (fiber alignment) and chemical cues (growth factors)1,2,10,11,35. Electrospinning is a facile method to create aligned fibers. Growth factors encourage nerve growth but in order to include them into nerve conduits (NC), a method for sustained release is required. To create a more robust system with both cues, these two signals should be combined. Several methods have been previously studied to provide extended release of protein within the electrospun fibrous scaffold for nerve regeneration but none to date have been capable of sustained release for over 60 days. Additionally, patterning of growth factor release has not been possible. Herein we describe a system to combine electrospinning with PLGA microsphere delivery to provide both signals.
Traditional methods to fabricate PLGA microspheres, such as the protocol outlined by de Boer et al., produced much larger microspheres33 (300±100 µm) that could not withstand the electrospinning process. Several attempts to incorporate them into the scaffolds were unsuccessful. This led to the use of a sonicator, in place of the vortex mixer, to better disperse the first emulsion and create smaller spheres. Once the microspheres were small enough, they easily passed through the needle and electrospun along with the rest of the solution. The primary electrospinning solution in this protocol is dissolved in water, rather than a harsh solvent, to prevent damage to the NGF and provide a hospitable substrate for cell growth28.
Many methods have been tested for providing topographical and chemical signals to nerve injury sites. Some of the early attempts at providing NGF to injuries involved putting solutions of free NGF or microspheres containing NGF into the center of a solid silicon nerve conduit (NC)36. These supported nerve growth; however they did not provide additional cues or allow the ability to pattern the location of the growth factors in the NC. In a more recent example Yan et. al. used poly[(lactic acid)-co-[(glycolic acid)-alt-(L-lysine)]] (PLGL) modified by grafting Gly-Arg-Gly-Asp-Gly (RGD peptide) and NGF to fabricate PLGL-RGD-NGF nerve conduits37. Similarly, these provided the chemical signal that promoted nerve growth, but did not provide other cues. Another study used NGF encapsulated in microspheres with electrospun scaffolds above and below them38. This system does provide both the topographical cues of a fibrous scaffold in combination with NGF as a chemical cue. However, the release only lasted a week, which is not long enough to support peripheral nerve repair, and the location of the microspheres is not well controlled.
Other researchers have focused on incorporating the chemical signals into the fibers. One method creates and emulsion of the hydrophilic protein with a hydrophobic polymer. This method was able to provide topographical and chemical cues, however 60% of the protein was released in the first 12 days39. Another example used coaxial electrospinning to encapsulate NGF within poly(lactic acid-caprolactone) (P(LLA-CL)) fibers40,41. This method provided results similar to the autograph in a rat sciatic nerve model. However, special equipment is need for coaxial electrospinning and it is unclear at what rate the protein was released, and if this could be sustained. The combination of topographical cues in the form of aligned electrospun fibers and chemical cues through microspheres that release for over 2 months as described here, provides multiple synergistic cues and is highly tunable to provide a customized environment.
There are several issues that need to be considered when electrospinning; environmental factors can affect electrospinning of any solution. An unexpected change in humidity, for instance, can have a significant effect on the electrospinning results and cause inconsistent fibers. Having a separate room that can be environmentally controlled is very helpful. A temperature of 70 °F with 50% humidity was used for these experiments. Another potential problem with an electrospinning system is short circuits; they often create obvious sparks between the syringe and metal pieces on the pump. To prevent this, place plastic or another insulator between them; food storage bag weight plastic works well. Similarly, stray electrical charges in the environment can cause material to accumulate in unexpected locations. To avoid this you can create an enclosure with acrylic sheets or other insulating material. More information on adjusting electrospinning conditions can be found in the review by Sill24.
The next steps for this procedure will include encapsulating other growth factors to create more dynamic scaffolds containing multiple growth factors. For example, NGF has been shown to work synergistically with Glial Cell Derived Growth Factor to support sensory and motor neurons3. Using PLGA microspheres, we can control the release rate by altering the L:G ratio, which would allow us to deliver various growth factors at different rate. A gradient of the microspheres can also be generated to build a more complex support system that directs cell behavior. Additionally, adhesion factors will be conjugated directly to the scaffold material through a Michael’s addition reaction26. Finally, the system will be tested with an in vivo rat sciatic nerve model.
The authors have nothing to disclose.
This work was partially funded through the Richard Barber Foundation and a Thomas Rumble Fellowship (TJW).
DAPI | Invitrogen | D1306 | |
Irgacure 2959 | BASF | 24650-42-8 | Protect from light |
PEO 900 kDa | Sigma-Aldrich | 189456 | |
Methacryloxethyl thiocarbamoyl rhodamine B | Polysciences, Inc. | 23591-100 | Prepare stock solution in DMSO |
Syringe Pump | KD Scientific | KDS100 | |
Power Source | Gamma High Voltage | ES30P-5W | |
Motor | Triem Electric Motors, Inc | 0132022-15 | Must attach to a custom built mandrel |
Tachometer | Network Tool Warehouse | ESI-330 | Use to monitor mandrel speed |
Omnicure UV Spot Cure System with collimating adapter | EXFO | S1000 | |
Needles | Fisher Scientific | 14-825-16H | |
Coverslips | Fisher Scientific | 12-545-81 | |
Polyvinyl Alcohol | Sigma-Aldrich | P8136-250G | |
Isoporopyl Alcohol | Sigma-Aldrich | I9030-500mL | |
Bovine Serum Albumin (BSA) | Fisher Scientific | BP9703-100 | |
BSA-FITC | Sigma-Aldrich | 080M7400 | |
β-Nerve Growth Factor (NGF) | R&D Systems | 1156-NG | |
65:35 Poly-Lactic-Glycolic-Acid (PLGA) | Sigma-Aldrich | 1001554270 | |
Dichloromethane | Sigma-Aldrich | 34856-2L | |
Coomassie (Bradford) Protein Assay | Thermo Scientific | 1856209 | |
3-(Trimethoxysilyl)propyl methacrylate | Sigma-Aldrich | 1001558456 | |
Fibronectin | Sigma-Aldrich | F2006 | |
DMEM | Lonza | 12-604F | |
FBS | Atlanta Biologicals | S11150 | |
PBS | Hyclone | SH30256.01 | |
Glutamine | Fisher Scientific | G7513 | |
Pen-Strep | Sigma-Aldrich | P4333 | |
Paraformaldehyde | Alfa Aesar | A11313 |