Abstract
Neural regeneration is extremely difficult to achieve. In traumatic brain injuries, the loss of brain parenchyma volume hinders neural regeneration. In this study, neuronal tissue engineering was performed by using electrically charged hydrogels composed of cationic and anionic monomers in a 1:1 ratio (C1A1 hydrogel), which served as an effective scaffold for the attachment of neural stem cells (NSCs). In the 3D environment of porous C1A1 hydrogels engineered by the cryogelation technique, NSCs differentiated into neuroglial cells. The C1A1 porous hydrogel was implanted into brain defects in a mouse traumatic damage model. The VEGF-immersed C1A1 porous hydrogel promoted host-derived vascular network formation together with the infiltration of macrophages/microglia and astrocytes into the gel. Furthermore, the stepwise transplantation of GFP-labeled NSCs supported differentiation towards glial and neuronal cells. Therefore, this two-step method for neural regeneration may become a new approach for therapeutic brain tissue reconstruction after brain damage in the future.
Introduction
The establishment of an efficient method for brain tissue regeneration is desired for the treatment of various diseases with brain parenchymal defects. Brain hemorrhage/infarction is a major cause of death and disability, with 13 million new cases per year1,2, and brain cancer, one of the most malignant human cancers, has 13.7 million new cases per year worldwide3. Both diseases are life-threatening and lead to enormous social and economic costs2,3; thus, brain tissue regeneration has been studied4. However, there is no established method to date, and the development of an effective therapeutic strategy is urgently needed.
The brain is known to be vulnerable to ischemic damage, in which loss of the blood supply leads to the loss of neuroglial cells, the tissue framework, and extracellular matrix components, resulting in the loss of brain parenchyma volume and the formation of a cavity1,5. This loss of tissue volume is irreversible, and spontaneous regeneration or wound healing processes do not occur, unlike in the skin, GI tract, and liver6,7. One of the reasons is that the edge of the damaged brain tissue is sealed with a glial scar that prevents proper inflammatory cell migration and vascular extension for regeneration8,9. In general, neural regeneration is much more difficult in adults10 than in infants because the number of NSCs that are present as endogenous sources for neuroglial cells is low in adult brain tissue11,12,13.
Neural tissue engineering can solve these problems by using biomaterials to create a scaffold to compensate for volume loss, enabling cells to migrate into the damaged area and reconstruct the entire tissue structure14,15,16,17,18,19. Among various biomaterials, hydrogels have been shown to regulate the cellular differentiation of mesenchymal stem cells (MSCs)20, iPS cells21, and various cancer cells22,23 and have recently been used for various biological applications24. Hydrogels are hydrophilic polymers commonly used in tissue engineering that can contain up to 90% water25 and have biological advantages, including a high oxygen content, nutrient permeability, and appropriate interfacial tension26. Natural substances have been used for hydrogels19, with various limitations, such as substantial variation27, the induction of immune reactions, and the release of components of tissue-derived pathogens28. To overcome these limitations, synthetic polymers have become widely used in hydrogels19. Neural tissue engineering using synthetic polymers and neurons may contribute to the regeneration of lost brain tissue29, but only a few reports have been published so far5.
In this study, to develop a new hydrogel material, we first focused on the electric charge of the substrate in cell adhesion30,31,32,33, and we found that an equal ratio of anionic and cationic monomers was most suitable for the attachment, growth, and differentiation of NSCs. Thereafter, the porous structure of the hydrogels was generated by cryogelation for 3D culture of neuroglial cells. In a traumatic brain injury model mimicking late event as cavity formation or surgical resection in mice, implantation of porous hydrogels with a similar stiffness to brain tissue induced host-derived neuroglial cell infiltration in the pores of the gels, along with vascularization. Furthermore, stepwise transplantation of NSCs into the hydrogels 2–3 weeks after implantation of the gels, where a vascular network formed and host glioneuronal cells had migrated, led to effective NSC survival, migration, and differentiation. These results indicate that porous hydrogels with a specific electric charge combined with stepwise NSC transplantation may represent a new therapeutic method for traumatic brain injuries with cavity formation.
Materials and methods
Preparation of hydrogels with various charges
Hydrogel sheets with a thickness of 1 mm were synthesized by using two parallel glass plates. To synthesize the differentially charged hydrogel, various combinations of a cationic monomer, 3-(acryloylaminopropyl)-trimethylammonium chloride (APTMA, Tokyo Chemical Industry Co., Ltd., Tokyo, Japan), and an anionic monomer, 2-acrylamido-2-methylpropane sulfonic acid, sodium salt (NaMPS, Sigma-Aldrich, St. Louis, MO, USA), were used. Dimethylacrylamide (Sigma-Aldrich) was employed as a neutral monomer. Five differentially charged hydrogels were constructed by using ratios of cationic (C) and anionic (A) monomers of 0:1, 1:5, 1:3, 1:1, and 3:1, which were designated C0A1, C1A5, C1A3, C1A1, and C3A1 gels, respectively. Polymerization of 1 M monomer was performed with 4 mol% N,N’-methylene-bis-acrylamide (Sigma-Aldrich) as a crosslinker with 0.1% (v/v) N,N,N’,N’-tetramethylethylenediamine (TEMED, Wako Pure Chemical Ind., Ltd., Japan) and 0.4% (v/v) ammonium persulfate (APS, Wako) at room temperature overnight. Alternatively, UV irradiation was used for polymerization. Polymerized hydrogels were immersed in a large amount of phosphate-buffered saline (PBS) to remove residual unpolymerized chemicals for 1 week, and the PBS was refreshed every day. Subsequently, gel disks were punched out of the gel sheet with a hole punch and sterilized in an autoclave at 120 °C for 20 min. The gel disks were then placed on polystyrene tissue culture dishes and equilibrated in cell culture medium overnight.
Preparation of the C1A1 porous hydrogel via cryogelation
For the preparation of porous hydrogels, cryogelation was employed using C1A1 hydrogels. The aqueous solution containing monomers, crosslinker, 4 mg/ml TEMED, and 0.5 mg/ml APS was purged with argon for 20 min to remove dissolved oxygen, inhibiting radical polymerization. The solution was cooled in an ice water bath to slow the speed of polymerization. The glass molds precooled in a cooling bath (− 16 °C) and injected with reaction solution were inserted into a zipper plastic bag and immediately placed in an ethanol cooling bath (NCB-3300, TOKYO RIKAKIKAI Co, Ltd., Tokyo, Japan). Frozen samples were kept at − 16 °C overnight to allow crosslinking polymerization. Samples with a UV initiator (0.5 mg/ml alpha-keto, Wako) were irradiated. After polymerization, the molds were warmed with cold tap water to defrost the samples and opened under ethanol to prevent gel rapid swelling. Then, the prepared gels were immediately placed into deionized water. The water was changed frequently for several days until it reached the conductivity of pure water.
Scanning electron microscopy (SEM)
SEM measurements were obtained using a JSM-6010LA scanning electron microscope (JEOL Ltd., Tokyo, Japan). Samples were freeze-dried, sputter-coated with a gold/palladium mixture using an E-1010 ion sputter system (HITACHI, Hitachi, Japan) and analyzed at an acceleration voltage of 20 kV using a BSE detector.
Fluorescence observation of hydrogels
The porous hydrogel was equilibrated in 0.1 mM fluorescein isothiocyanate (FITC) solution to stain the gel matrix. Then, the FITC solution in the pores was squeezed out in a water bath for 1 min. 2D and 3D fluorescence imaging was carried out by using an all-in-one fluorescence microscope (BZ-X700, KEYENCE, Osaka, Japan).
Measurement of the mechanical and physical properties of the C1A1 porous hydrogel
(i) The stiffness of the C1A1 porous hydrogel was measured as the Young’s modulus under compression using Tensilon RCT 1310A (ORIENTEC CORPORATION, Tokyo, Japan), as described previously34. Briefly, a 100 N load cell was used to detect the stress response of the sample to the applied strain. Disc samples 10 mm in diameter were obtained from the gel sheets. The compression velocity for all samples was set to a constant 10% strain per minute (~ 0.220–0.365 mm/min depending on the sample thickness, which is determined by the gel composition and the degree of swelling). Young’s modulus was calculated from the slope of the beginning of the compressive curve.
(ii) The surface electric charge of the hydrogel was measured as the zeta potential by using a submicron particle size analyzer (Delsa Nano HC, BECKMAN COULTER, Brea, LA, USA) as described previously34. Briefly, for detection of the potential, a standard particle suspension (500–600 nm in diameter; Otsuka Electronics, Osaka, Japan) consisting of a polystyrene-latex core with a hydroxypropyl cellulose shell was diluted 300 times with a 10 mM sodium chloride (NaCl) aqueous solution. A specimen sheet presoaked in 10 mM NaCl aq was placed on a quartz block cell with a liquid-flow passage, and the standard particle-dispersed solution was used to fill the passage. The zeta potential was evaluated by laser-Doppler electrophoretic light scattering (ELS) (n = 4).
(iii) The surface electric charge of the pores was measured as the Donnan potential, as described previously35. Briefly, the C1A1 porous hydrogel was rinsed in a distilled water bath to remove unreacted reagents until the electroconductivity of the bath water at room temperature, which was monitored with a conductance meter (FiveEasy F30, METTLER-TOLEDO, Columbus, OH, USA), reached 0.5 µS/cm, comparable to that of pure water. Then, the gel was equilibrated in 10−5 M NaCl solution. The Donnan potential of the gel as a function of bulk depth was measured by a self-developed potential analyzer constructed from an oscilloscope (HDO6034, TELEDYNE LECROY, Chestnut Ridge, NY, USA), amplifier (8700 CELL EXPLORER, DAGAN, Minneapolis, MN, USA) and manipulator (DMA-1511, NARISHIGE, Tokyo, Japan) of a submicron electrode. A tiny electrode with a 150-nm tip edge diameter was inserted into the gel bulk stored in 10−5 M NaCl aq. at 0.8 µm/s velocity. A carbon electrode was used as the reference.
(iv) Measurement of the weight swelling ratio W of the C1A1 porous hydrogel. Gels were cut into smaller pieces (usually 10 mm cylindrical sample) and weighed to obtain their weight in equilibrium swollen state (msw���) and then freeze-dried and weighed to obtain their dry weight (md��). Freeze-dried samples were soaked in cyclohexane and quickly weighed to obtain the weight of samples in nonsolvent (msw,CH���,��). The equilibrium water regain (W) and dry samples were calculated according to the following equations36:…
