Three kinds of Nylon micro meshs (NMMs) which had 20, 10, 8 µm pore size were purchased from Spectrum Laboratories (Spectrum, CA, USA). And NMMs with 40 µm pores were taken from a cell strainer (BD Falcon, NJ, USA).

NMMs were sonicated with acetone, isopropylalcohol (IPA) and DI water for 5min per each and dried by air stream. NMMs were coated with Poly-D-Lysine (PDL, Sigma, MO, USA) by soaking the NMMs in PDL solution (0.1 mg/ml in borate buffer, pH 8.5) at 37℃. After 6 hour, PDL solution was aspirated and NMMs were immersed in 70% ethanol for sterilization. After the sterilization, NMMs were rinsed with DI water. The NMMs were finally dried in clean-bench or incubator.

Polyelectrolyte multilayer (PEL) were coated using layer-by-layer method. The cleaned NMMs were immerged in the solution of Poly allylamine hydrochloride (PAH, 20 mM, pH 9.0, Sigma, MO, USA) for 1 min, followed by five times washing with DI water. The PAH coated NMMs were immerged in the solution of Poly sodium 4-stlrensulfonate (PSS, 60 mM, pH 9.0, Sigma, MO, USA) for 1 min, followed by five times washing with DI water. The process of PEL coating were repeated ten times and finished with the PAH coating. NMMs were sterilized by UV for 5 min and dried in clean-bench or incubator.

Agarose gel (gelling range: 36~39℃, Amresco Inc., Ohio, USA) was dissolved in 100℃ DI water and 0.7% (w/v) agarose hydrogel was formed. After cooling the agarose gel to 36℃, cell suspension was mixed with the agarose and plated on PDL coated NMMs.

Hippocampus tissues were dissected from E18 Sprague/Dawley rat embryo. The tissues were washed three times with Hank's Balanced Salt Solution (HBSS) and mechanically dissociated using 1 ml pipette tips. Cell suspensions were centrifuged for 3 min at 1,000 rpm and the cell pellets were suspended again in plating medium which was composed of serum free Neurobasal/B27 (Invitrogen, CA), 12.5 µM L glutamate (sigma, MO), 2 mM L glutamine (Gibco) and 1% penicillin streptomysin (invtrogen, CA). Finally, the cell suspend plated onto the coated NMMs at the density of 200 to 300 cells/mm². Cultures were maintained in an incubator (5% CO₂, 37℃). After three days from cell plating, half of the medium was replaced with fresh culture medium without L glutamine every 3 days.

Neurons were stained by Live-Dead viability/cytotoxicity kit (Invitrogen, CA, USA). 10 µl of ethidium homodimer and 2.5 µl calcein acetocymethyl ester were mixed in 5 ml of phosphate buffered saline (PBS). 200 µl of the dye solution was added in 1 ml sample and covered with aluminum foil to protect the dye from photobleaching. After the incubation (5% CO₂, 37℃) of the sample for 20 min, the solution in the sample was removed and NMMs washed with PBS.

Appropriate mechanical properties of 3D scaffold are important in 3D tissue engineering (Lee et al., 2008). First, flexibility is required to maintain the shape of scaffold and handling. Second, micro and nano-scale structural architecture is desirable to mimic in vivo microenvironments (Curtis and Wilkinson, 1998; Bayliss et al., 1999). Third, some porosity is required for the diffusion of nutrients, metabolic wastes and interconnectivity of cells.

Fig. 1A shows the compliance of the NMM to the mechanical bending. It was also easy to handle by hand for forceps. Fig. 1 also shows the ordered structures of NMM in micrometer scales. The thickness of the individual nylon fiber was 30 or 35 µm and the thickness of NMM was 75 µm, 45 µm, 55 µm, 60 µm, respectively (Fig. 1B~E). NMMs have many regular pores with the pore diameters of 8 µm, 10 µm, 20 µm, and 40 µm (Fig. 1F~I). These properties made NMMs suitable for 3D cell culture scaffolds.

Biocompatibility of materials is a crucial parameter, since it can be decisive criteria as a culture scaffold. Most of the reported scaffolds were composed of ECM molecules (Badylak, 2002), or had bioactive molecules to modulate cellular activities (Zhu, 2010), or they had controllability in chemicals (Yildirim et al., 2010).

Fig. 2 shows the surface modification of NMMs using positively charged polymers. We coated NMMs with either poly-D-lysine (PDL) or polyelectrolyte multilayer (PEL). PDL is one of the most common polymer used as a cell adhesive coating material. PEL was used to modify the NMM surface with much denser positive charge layers through layer-by-layer coating technique, which turned out to be more efficient in terms of reproducibility and reliability.

In order to investigate the effect of chemical controllability and viability of cells on the NMM, we prepared different pore size of NMMs coated by PDL and PEL and cell plate. Fig. 3 shows excellent cell viability at 9 DIV. Neurons tend to grow on and in-between the micro-fibers and pores were be filled with neuronal growth.

Entrapment of cells in three dimensional space has a special meaning that it provide distinct differences in cell morphogenesis, behavior, gene expression, differentiation from 2D environment (Holmes et al., 2000; Zaman et al., 2006). In other word, the recovered 3D environment cause different viability and functionality of cell.

Many of the studies used natural hydrogels or synthetic gel such as alginate (Pangas et al., 2003), collagen (Batorsky et al., 2005) and poly ethylene glycol (PEG) (Elisseeff et al., 1999) to entrap cells. We used agarose hydrogel which is a common material in biological laboratories for the cell entrapment (Fig. 2). Fig. 4 shows the neuronal growth in agarose hydrogels. Neurons were well dispersed in the gel, and neurites were formed and extended (Fig. 4B, E). They tend to get clustered in the gel after forming thick fiber bundles. As the gel was not functionalized to be cell adhesive, the clustering process was likely to occur in the matured neuronal cultures. However, neurons were still viable and maintained in 3D NMM scaffolds.

As NMMs had ordered structures in micrometer scales (Fig. 1), each nylon fiber can provide the topographical guidance cues to the growing neurites. Fig. 5A shows an example of the guided neurite outgrowth along the 3D mesh fibers. The fiber thickness was comparable to the size of soma and neurites followed the edge of the fibers. This is also shown in scanning electron micrograph in Fig. 5B. Fig. 5C shows more 'patterned' neurites on the mesh. Although the fiber itself was too thick to control thinner neurites, its edges clearly had an effect on neurite guidance.