Morphological changes in neurons by self-organized patterned films∗

In tissue engineering, micro/nanofabrication is important to modify substrate surfaces for regulating the attachment and growth of cells. In neuroscience, it is significant for neural regeneration; this involves guiding and extending dendrites and axons by a cell culture scaffold which acts as an extra cellular matrix. In this study, we prepared highly regular porous honeycomb-patterned films by a simple casting technique and cultured neurons to investigate their morphologies on the patterned films. The morphologies of neurons were examined by a scanning electron microscope and a confocal laser scanning microscope. The neurons were round and the neurites extended randomly on the flat film. The patterns influenced the morphologies of neurons. The morphologies of neurons were changed by varying the pore size of the honeycombpatterned films. The neurites spread along the rims of the honeycomb pattern. These results suggest that the self-organized honeycomb-patterned films are useful biomaterials for neural tissue engineering. [DOI: 10.1380/ejssnt.2005.159]


I. INTRODUCTION
Brain, spinal cord injuries, and neural-degenerative diseases are likely to require the transplantation of neural cells and tissues. The mammalian central nervous system has little capacity for self-repair. The replacement of lost and dysfunctional neurons by tissue transplantation or nerve graft has been investigated [1]. These approaches have been developed as useful tools for restoring function in the damaged central nervous system. In tissue engineering, these approaches are significant for the reconstruction of tissues and organs not only to carry out research on cells or liquid factor but also to develop scaffolds [2,3]. Scaffolds that have 3-D structures can induce adhesion, proliferation, and differentiation of cells and reorganized tissues and organs. The materials that are used for tissue engineering are made of biodegradable polymers. It has been reported that micro/nano-patterns influenced the morphologies, proliferation, and differentiation of cells [4][5][6]. It has been found that various micro-patterned substrates fabricated by lithographic techniques can reconstruct artificial neural networks by controlling the morphologies of adhered neurons and outgrowth of neurites [7][8][9][10][11][12][13]. These techniques are expected to be applied for neural regeneration in the future. However, these techniques require high energy and involve many processes. In addition, materials for substrates are limited. We have reported that the honeycomb-patterned films are prepared by self-organization [14][15][16]. The patterned films have regular pores in micro/nano-meter. They can be prepared with ease, low energy, and low cost. Moreover, we can control the pore size of these films. In this study, we cultured neurons on honeycomb-patterned films and investigated the morphological changes in neurons and neurite extension by varying the pore size of the films.

A. Preparation of self-organized honeycomb-patterned films
Poly ( -caprolactone) (PCL) and an amphiphilic polymer were mixed together and dissolved in chloroform in a weight ratio of 10:1 ( Fig. 1(a)). We cast the polymer mixture onto the glass substrates. Honeycomb-patterned films with regular pores were prepared by blowing humid air on the surface of the polymer solution ( Fig. 1(b)).

B. Preparation of the PCL flat film
The polymer solution was dropped onto a cover glass. The cover glass with the polymer layer was spun at 1000 rpm for 30 seconds by using a spin coater (MIKASA, 1H-D7).

C. Pretreatment of PCL films
The PCL flat film and PCL honeycomb-patterned films were soaked in 1-propanol solution for 5 minutes and then washed with ethanol. The PCL films were then attached to a cover glass, placed in culture dishes, and sterilized by exposure to UV rays. These films were then soaked in poly-L-lysine solution (50 mg / L, 0.1 M Boric acid, pH 8.3) for 1 hour to coat poly-L-lysine on the films.

D. Preparation of neural cells
Neural cells were prepared from the cerebral cortices of embryonic day-14 mice (CLEA Japan, Inc). In brief, the cerebral cortexes of embryonic day-14 mice were dissected and the meninges were carefully removed. The tissues were transferred into 15-ml tubes with culture medium containing 55 µM 2-mercaptoethanol and gently triturated with a fire-polished pasteur pipette until most of the tissues were dissociated into single cells. The cell number and viability were determined. Cells were seeded onto the PCL flat film to estimate the population of neural stem cells (NSCs). After incubating the cells for 6 hours, cells were immunocytochemically stained for nestin to identify and estimate their population.

E. Cell culture condition
The neural cells were seeded onto the PCL flat film and PCL honeycomb-patterned films at a density of 2.0 × 10 4 cells/cm 2 . They were cultured in serum medium (Opti-MEM (Invitrogen), 10% fetal bovine serum, 55 µM 2-mercaptoethanol (Invitrogen)) for the first day. After the second day, they were cultured in serumfree medium (Opti-MEM, B27 supplement, 55 µM 2mercaptoethanol). They were incubated at 37 • C under a humidified atmosphere of 5% CO 2 .

F. Scanning electron microscopic observation
The cultured cells were fixed with 2.5% glutaraldehyde in phosphate-buffered saline (PBS). They were washed with PBS and water. Subsequently, the samples were dehydrated by washing in increasing ethanol concentrations and then air-dried. The samples were sputtered with platinum and investigated with a scanning electron microscope (Hitachi, S-3500).

Immunostaining for β-tubulin III
The cultured cells were fixed with 10% formalin in PBS for 30 minutes at room temperature. The samples were washed with PBS 3 times for 5 minutes. They were then incubated in blocking solution (5% goat serum, 0.2% Triton X-100 in PBS) for 1 hour. Then, the samples were incubated with mouse monoclonal anti-β-tubulin III (1:500) in PBS for 2 hours. After washing with PBS, the samples were incubated with fluorescein isothiocyanate (FITC)conjugated anti-mouse IgG (1:200) for 2 hours. After washing with PBS and water, the samples were air-dried e-Journal of Surface Science and Nanotechnology

Immunostaining for nestin
The cultured cells were fixed with 10% formalin in PBS for 30 minutes at room temperature. The samples were washed with PBS 3 times for 5 minutes. They were then incubated in blocking solution (5% goat serum, 0.2% Triton X-100 in PBS) for 1 hour. Then, they were incubated with mouse monoclonal anti-nestin (1:1000) in PBS for 2 hours. After washing with PBS, the cells were incubated with biotinylated anti-mouse IgG (1:1000) for 2 hours at 37 • C. Then, they were incubated with Alexa 488-conjugated avidin (1:2000) and Texas-red conjugated phalloidin (1:50) for 2 hours at 37 • C. After washing with PBS and water, the samples were air-dried and then mounted with mounting media for confocal microscopic observation.

A. Preparation of self-organized honeycomb-patterned films
We could prepare self-organized honeycomb-patterned films by casting a polymer solution. The pore size could be controlled in the range from 3 to 10 µm by changing the casting volume [16]. We prepared PCL flat films and PCL patterned films (3 µm, 5 µm, 8 µm, and 10 µm in diameter) for culturing neural cells.
The rims of the honeycomb-patterned films widened with increasing pore size of the patterned films. The porosity of each film was about 50% (Fig. 2).

B. Cell preparation from cerebral cortexes of embryonic mice
The viability of neural cells prepared from the cerebral cortices of 4 mice (embryonic day-14) was 90∼95%. We investigated the population of neural NSCs in the prepared cells by staining with nestin. We found that the cell mixture contained 90∼95% of NSCs (Fig. 3).

C. Morphologies of adhered neural cells and neural extension
After 5 days of culture, we investigated the morphologies of adhered neural cells and the extension of neurites on both PCL flat films and PCL honeycomb-patterned films. The neurons were stained for β-tubulin III, a neuron-specific tubulin for their identification and for the observation of neural networks on the films (Figs. 4 (a1)-(f1)).
The neurons possessed highly branched, multi-polar neurites and rounded cell bodies (Fig. 4(a)). On the other hand, several neurons aggregated near each other and adhered with lamella structure around the cell body on the patterned film (pattern pore size: φ3 µm) (Fig.  4(b)). Single neurons adhered with lamella structure on the patterned film (pattern pore size: φ5 µm) (Fig. 4(c)). The neurons were round with no lamella structure on the patterned film (pattern pore size: φ8 µm) (Fig. 4(d)). The neurites extended along the honeycomb-pattern rims and neurons formed network structures on the patterned e-Journal of Surface Science and Nanotechnology  films (Figs. 4(a)-(f)). Further, on the patterned film (pattern pore size: φ10 µm), each round neuron adhered to the pattern rims. The number and branching of the neurites were reduced compared with those on the flat films. The neurons formed simple network structures on the rims of the patterns. The neurons were round, similar to the morphology of neurons on the flat films ( Fig. 4(e)). Furthermore, we found another specific morphology of neurons on patterned films with a pore size of 3 µm. Neurons formed large spheroid-like round aggregates on these films. The neural aggregates formed large bundles of neurites which extended in the shape of radiation (Fig. 4(f)).
Morphological changes in neurons and formation of lamella structures are likely to depend on the change in pore and rim sizes (Table 1). This suggests that honeycomb patterns act as a guide as well as scaffold for neurons and neurites.

D. Investigation of the number of neurites on the honeycomb-patterned films with varying pore sizes
The honeycomb patterns influenced neurite extension and outgrowth. We investigated the number of neurites per neurons on the PCL flat film and patterned films (5,8, and 10 µm in diameter). We found that the number of neurites decreased with increasing pore size and rim size of the patterned films (Fig. 5).
Collagen gels and sheets are used as biomaterials for tissue engineering. However, the mechanical strength of these materials is weak, and other polymer substrates are less biodegradable and biocompatible. The advantage of self-organized patterned films is that they are prepared from various biodegradable and biocompatible polymers. The preparation of honeycomb-patterned films dose not require high energy or high cost.
In tissue engineering, it is important to culture cells in 3-D scaffolds [17]. We can culture cells with other cells on both sides of the honeycomb-patterned films. It is believed that the cells can interact with each other via the pores of the patterned films. Moreover, the patterned films can easily be formed in various shapes. With these patterned films, it will be possible to prepare 3-D structures, such as tube, roll, and various multi-layered structures. We can co-culture neurons with glial cells or endothelial cells in these 3-D scaffolds to reconstruct 3-D tissues. The patterned films are candidates for tissue engineering scaffolds.

IV. CONCLUSIONS
We could prepare biodegradable polymer patterned films and control the pore size of the patterned films. Neurons formed neural networks on the patterned films, and the morphologies of cells could be changed by varying pore size of the patterned films. The self-organized honeycomb patterned films are useful biomaterials for neural tissue engineering.