Fabrication of complex three-dimensional tissue architectures using a magnetic force-based cell patterning technique
- First Online:
- Cite this article as:
- Akiyama, H., Ito, A., Kawabe, Y. et al. Biomed Microdevices (2009) 11: 713. doi:10.1007/s10544-009-9284-x
- 269 Views
We describe the fabrication of three-dimensional tissue constructs using a magnetic force-based tissue engineering technique, in which cellular organization is controlled by magnetic force. Target cells were labeled with magnetite cationic liposomes (MCLs) so that the MCL-labeled cells could be manipulated by applying a magnetic field. Line patterning of human umbilical vein endothelial cells (HUVECs) labeled with MCLs was successfully created on monolayer cells or skin tissues using a magnetic concentrator device. Multilayered cell sheets were also inducible on a culture surface by accumulating MCL-labeled cells under a uniform magnetic force. Based on these results, we attempted to construct a complex multilayered myoblast C2C12 cell sheet. Here, patterned HUVECs were embedded by alternating the processes of magnetic accumulation of C2C12 cells for cell layer formation and magnetic patterning of HUVECs on the cell layers. This technique may be applicable for the fabrication of complex tissue architectures required in tissue engineering.
KeywordsCell patterningThree-dimensional tissue constructMagnetite nanoparticlesMagnetic forceTissue engineering
Cells in normal tissues and organs are well-known to be found in an orderly arrangement with surrounding homotypic and/or heterotypic cells and that their spatial organizations are often crucial in driving native functions. In the field of tissue engineering, although some tissues composed of multiple cell-types have been successfully fabricated (Levenberg et al., 2005; Koike et al., 2004), the arrangements of individual cells are rarely regulated at the micrometer scale. It is essential to mimic the natural microenvironment for the construction of functional tissue substitutes (Khademhosseini et al., 2006). In this regard, cell patterning techniques at the micro-scale have attracted great attention, and they have been applied not only to tissue engineering but also to cell biological studies including analysis of cell–cell interactions (Hui and Bhatia, 2007), measurement of mechanical stress generated within multicellular aggregates (Nelson et al., 2005) and drug screening (Khetani and Bhatia, 2008).
In tissue engineering, most efforts have focused on the scaffold design such as surface configuration (Hashi et al., 2007; Riboldi et al., 2008) and immobilization of cell adhesion proteins (Koh et al., 2008; Hu et al., 2003). As a scaffold-free method to construct tissue substitutes, the Okano’s group developed a cell sheet-based procedure. Here, a cultural substrate surface was grafted with a thermo-responsive polymer, poly (N-isopropylacrylamide) (PIPAAm), and the cell layers that formed on the polymer were easily harvested by a change in temperature (Ohashi et al., 2007; Tsuda et al., 2007). Conversely, cell patterning methods developed in the past decade have been dependent on the design of biomaterial surfaces to control the arrangement of living cells. Soft lithographic approaches such as the microcontact printing (Brock et al., 2003) and microfluidic channel flow patterning (Jiang et al., 2005) are regarded as exemplars. However, since the substrates used in the lithographic methods are non-biodegradable, applications of the methods have been very limited in the tissue engineering field. In contrast, cell patterning techniques such as three-dimensional (3D) photo-patterning (Liu Tsang et al., 2007) and ink-jet printing (Wilson Jr and Boland, 2003; Xu et al., 2006) do not require surface modification of the substrate. Although these techniques provide a potentially powerful method to construct 3D cell patterns, many problems currently exist in using such approaches. For example, the 3D photo-patterning method requires tumorigenic UV irradiation for encapsulation of the cells to the matrix, and the ink-jet printing approach causes loss in cell viability due to dehydration during the processing. Therefore, the development of a new cell patterning procedure applicable to tissue engineering is still required.
In recent years, nanotechnology-based approaches have been applied to induce cellular organizations in tissue engineering (Mironov et al., 2008). Among them, one of the promising procedures is the magnetic manipulation technique using magnetic nanoparticles, because cells labeled with magnetic particles can be remotely manipulated by applying a magnetic field (Dobson, 2008). For magnetic labeling of target cells, we have used magnetite cationic liposomes (MCLs), which are cationic liposomes containing 10 nm magnetite nanoparticles designed to improve the accumulation of magnetite nanoparticles into target cells by using electrostatic interactions between cell membranes and MCLs (Shinkai et al., 1996). Multilayered cell sheets using various cells have been constructed by accumulating MCL-labeled cells on the culture surfaces by magnetic force (Ito et al., 2004; 2005). In addition, two dimensional cell arrangements were controlled using magnetic concentrator devices, in which magnetized micrometer-thick steel plates were embedded and the ranges and positions of the magnetic fields could be regulated at the micrometer scale (Ino et al., 2007a). In principle, using the magnetic cell manipulation technique, cell patterns can be created irrespective of surface conditions. In the present study, we applied the magnetic cell patterning technique to form cell patterns on a monolayer of cells, living skin and multilayered cell sheets. Based on these results, we also attempted to fabricate complex 3D tissue constructs by combining the magnetic force-based cell sheet formation and cell patterning procedures.
2 Materials and methods
2.1 Cells and cultures
Human umbilical vein endothelial cells (HUVECs) and normal human dermal fibroblasts (NHDFs) were obtained as frozen cells following primary culturing from the supplier (Kurabo, Osaka, Japan). These cells were grown in commercially available growth medium (HuMedia-EG2 for HUVECs, and Medium 106 S for NHDFs; Kurabo). Mouse fibroblast NIH3T3 cells and myoblast C2C12 cells were grown in Dulbecco’s modified Eagle medium (DMEM; Sigma-Aldrich, St. Louis, MO, USA) supplemented with 10% fetal bovine serum (FBS; Biowest, Nuaille, France), 0.1 mg/ml streptomycin sulfate and 100 U/ml potassium penicillin G (Wako Pure Chemical Industries, Osaka, Japan). Cells were cultured at 37°C under a humidified atmosphere of 5% CO2 and 95% air.
2.2 Preparation of magnetite cationic liposomes
The magnetite (Fe3O4; average particle size = 10 nm) used as the core of the MCLs was donated by Toda Kogyo (Hiroshima, Japan). Magnetic characteristics of the magnetite at 796 kA/m (room temperature) were 2.0 kA/m, 63.9 Am2/kg, and 2.6 Am2/kg for coercivity, saturation flux density, and remanent flux density, respectively. MCLs were prepared as described previously (Shinkai et al., 1996). Briefly, N-(α-trimethylammonioacetyl)-didodecyl-D-glutamate chloride (TMAG, a cationic lipid), dilauroylphosphatidyl-choline (DLPC), and dioleoylphosphatidyl-ethanolamine (DOPE) in a molecular ratio 1:2:2 were dissolved in chloroform, and the solvent was evaporated to form lipid films. The resultant films were hydrated by vortexing in colloidal magnetite nanoparticles to form liposomes, followed by sonicating for 30 min.
2.3 MCL uptake by cells
The uptake of MCLs by cells was examined as previously described (Shinkai et al., 1996). Briefly, C2C12 cells (5 × 105 cells/dish) were seeded into 60-mm cell culture dishes (Greiner bio-one, Frickenhausen, Germany) and cultured to be subconfluent for 24 h. The medium was changed using freshly prepared medium containing MCLs (net magnetite concentration, 100 pg/cell), and cell culturing was continued. The cells were sampled periodically (1, 4, 8 and 24 h) after MCL addition to measure iron concentration and cell number using potassium thiocyanate (Owen and Sykes, 1984) and trypan blue dye exclusion, respectively. To examine the effect of MCL uptake on cell viability, control dishes without the MCL addition were also prepared. The number of viable cells in the dishes was counted by the trypan blue dye exclusion after 48 h of the medium change. Cell viability in the presence of MCLs was determined by defining the viable cell number of the control dishes as 100% cell viability.
2.4 Devices for cell patterning
Steel plates (thickness, 10, 30, 100 and 200 μm; Toyo Koban, Tokyo, Japan) were used as magnetic field gradient concentrators. To fabricate linear cell patterns, these steel plates were sandwiched between acrylic resin plates (height, 10 mm). For arbitrary cell patterns, acrylic resin plates were cut using a laser beam by computer-aided design (CAD), and the letter “M”, “A” or “G” was engraved on the acrylic resin plates (each character size, 10 × 10 mm), and then steel plates with a thickness of 200 μm were embedded into the grooves (Ino et al., 2007a). To magnetize steel plates for magnetic cell patterning, these devices were placed on neodymium cylindrical magnets (diameter, 30 mm; height, 15 mm; magnetic induction, 0.4 T). Magnetic flux density measured at the surface of the magnetic field gradient concentrators using a gauss meter (F. W. Bell, Orlando, FL, USA) was 0.18 T.
2.5 Cell patterning onto a monolayer of cells
2.6 Cell patterning onto skins
To fabricate cell patterns on living tissues, skins derived from neonatal Wistar rats (2- or 3-days old; Kyudo, Tosu, Japan) and Balb/c mice (3-weeks old; Kyudo) were surgically obtained. The skin was directly laid on the cell patterning device with a 200 μm-thick steel plate placed on the inner side of the skin (hypodermal side) to be topside, and a cylindrical silicon rubber tube (inner diameter, 10 mm; length, 9 mm) was put on the skin to create a cell culture space. Magnetically labeled HUVECs (2.5 × 103 cells in 0.3 ml medium) were then seeded into the inner-side of the silicon rubber tubes and cultured for 45 min. The skins were washed with phosphate-buffered saline (PBS), and fixed in 4% paraformaldehyde (PFA) in PBS. Rabbit anti-human CD31 antibodies (Dako, Glostrup, Denmark) were used for immunostaining of HUVECs on the skin surface. Skins were incubated with 1% bovine serum albumin in PBS at 37°C for 60 min for blocking, followed by incubation overnight at 4°C with anti-CD31 antibodies and 37°C for 60 min with the secondary antibodies (biotinylated goat anti-rabbit immunoglobulin G (IgG); Dako). Thereafter, the skins were incubated at 37°C for 60 min with peroxidase-conjugated streptavidin (Dako). Each step was followed by washing three times with PBS at room temperature. Peroxidase activity was visualized after soaking in 0.02% diaminobenzidine tetrahydrochloride containing 0.005% hydrogen peroxide (brown staining indicates peroxidase activity) at room temperature for 10 min.
The animal experiment was reviewed by the Ethics Committee on Animal Experiments, Faculty of Engineering, Kyushu University.
2.7 Cell patterning onto cell sheets
The patterning procedure onto multilayered cell sheets is illustrated in Fig. 3(a). To control the size of cell sheets, cylindrical silicon rubber tubes (inner diameter, 17 mm; length, 5 mm) were placed at the center of the 35-mm culture dishes. Magnetic tissue-engineered cell sheets were fabricated as described previously (Ito et al., 2004). Briefly, NHDFs (1.5 × 106 cells/dish), NIH3T3 cells (2 × 106 cells/dish), or C2C12 cells (2 × 106 cells/dish) labeled with MCLs were seeded into the inner-side of the silicon rubber tubes in the culture dishes, and the magnets were placed on the reverse side of the culture dishes to provide a vertical magnetic force. After culturing for 2 h, the magnets were removed, and the dishes were placed on the magnetic field gradient concentrators. Subsequently, HUVECs (1 × 104 cells/dish) labeled with MCLs were seeded onto the cell sheets and cultured for 4 h.
For histological evaluation, C2C12 cell sheets were washed three times with PBS, fixed in 4% PFA in PBS and embedded in paraffin. Thin slices (4 μm) were stained with hematoxylin-eosin (H&E). To visualize HUVECs on cell sheets, the cells were stained using mouse anti-human CD31, goat anti-mouse IgG alkaline phosphatase conjugate, nitro-blue tetrazolium chloride (NBT), and 5-bromo-4-chloro-3’-indolyl phosphatase p-toluidine salt (BCIP) (Tubule Staining Kit for CD31; Cat. no. KZ-1225; Kurabo), according to the manufacturer’s protocol.
2.8 Incorporation of patterned cells into 3D tissue constructs
The fabrication procedure of 3D tissue constructs, in which HUVECs patterns were embedded in a C2C12 cell layer, is illustrated in Fig. 4(a). Firstly, a C2C12 cell layer (1 × 106 cells/dish) was constructed in a 35-mm dish using a similar method for cell sheet fabrication as mentioned above. After culturing for 1 h, the magnet was replaced under the cell patterning device with a 200 μm-thick steel plate, and then magnetically labeled HUVECs (4.3 × 103 cells/dish) were seeded onto the C2C12 cell layer. After culturing for a further 2 h, magnetically labeled C2C12 cells (1 × 106 cells/dish) were seeded onto the C2C12 cell layer patterned with HUVECs using the cylindrical magnet. After culturing for 1 h, the dish was rotated perpendicularly and magnetically labeled HUVECs (4.3 × 103 cells/dish) were seeded again to fabricate a crossing pattern of HUVECs. Finally, after culturing for 2 h, magnetically labeled C2C12 cells (1 × 106 cells/dish) were layered on the top of the second pattern of HUVECs and cultured for a further 2 h.
2.9 Fluorescence microscopy
Fluorescence microscopy of the prepared cell sheets was performed using Cell-Tracker (Molecular Probes, Eugene, OR, USA). HUVECs patterned onto monolayer cells, cell sheets, and skins were pre-stained with CMTMR (5-(and-6)-(((4-chloromethyl)benzoyl)amino) tetramethylrhodamine; orange fluorescent probe). To observe each pattern of the HUVECs within the 3D constructs, the HUVECs at the first and second patterning were pre-stained with CMTMR and CMFDA (5-chloromethylfluorescein diacetate; green fluorescent probe), respectively. These cells were observed using a fluorescence microscope (Olympus, Tokyo, Japan). For 3D analysis, the HUVECs and C2C12 cells were pre-stained with CMTMR and CMFDA, respectively. Omnifocal and 3D images of the multilayered tissue constructs were observed using a fluorescence microscope (BZ-9000, Keyence, Osaka, Japan) and processed with a BZ-Analyzer (Keyence).
3 Results and discussion
3.1 Magnetic labeling of cells with MCLs
MCL uptake amount by cells used in this study
The cytotoxicity of MCLs may be an important issue for their use in tissue engineering fields. As shown in Table 1, the magnetic labeling of cells with MCLs at 100 pg/cell did not affect the cell viability. In subsequent experiments, the cells were incubated with MCLs at 100 pg/cell for 24 h (HUVECs and NIHDFs) or 4 h (NIH3T3 cells and C2C12 cells) for magnetic labeling of the cells.
3.2 Cell patterning onto a monolayer of cells
The patterning procedure of HUVECs onto a monolayer of C2C12 cells is illustrated in Fig. 1(a). Magnetic field gradient concentrators with various thicknesses of magnetized steel plates sandwiched between acrylic resin plates were used to attract magnetically labeled target cells to the desired position for cell patterning. After C2C12 cells reached a confluent state, the culture dishes were placed on the cell patterning devices (magnetic field gradient concentrators with a magnet), and then HUVECs labeled with MCLs were seeded to the dishes. When a steel plate with a thickness of 30 μm was used, an arrangement involving the formation of a line of single cells was observed (Fig. 1(e)). This observation suggests that this method facilitates single cell manipulation to construct tissue architectures. A single cell patterning event was also achieved using a steel plate thickness of 10 μm (data not shown). The use of steel plates with thicknesses of 100 μm and 200 μm gave rise to linewidths of cell patterns that were almost consistent with the thickness of the steel plates (Fig. 1(f) and (g)). These results indicated that linewidths of cell patterns could be controlled by the thickness of the magnetized steel plates.
3.3 Cell patterning onto skins
3.4 Cell patterning onto cell sheets
To fabricate arbitrary cell patterns on the cell sheets, magnetic field concentrators, in which acrylic resin plates had the letters “M”, “A” or “G” engraved and steel plates with a thickness of 200 μm were embedded into the grooves of the acrylic resin plates, were used as cell patterning devices (Fig. 3(c)). Cell sheets formed when the dishes were placed on the cell patterning devices, and magnetically labeled HUVECs were seeded onto the cell sheets formed inside the silicon rubber tubes. The CD31 staining revealed that HUVECs were successfully patterned in the letter of “M”, “A” or “G” according to the cell patterning device on NHDF, NIH3T3 or C2C12 cell sheets (Fig. 3(d)–(f)). Moreover, the line widths of the cell patterns on the cell sheets were almost consistent with the thickness of the steel plates (200 μm, Fig. 3(g)), and well-shaped patterns were successfully fabricated (Fig. 3(h)).
3.5 Incorporation of patterned HUVECs into 3D tissue constructs
As shown in Fig. 4(b)–(f), although the patterns of HUVECs were maintained during the fabrication, the linewidth of the cell patterns was somewhat narrower compared with the thickness of the patterning device (200 μm). We believe that this was not caused by insufficient magnetic force because the C2C12 multilayered sheet was not thick (approximately 150 μm) and it was therefore possible to apply an appropriate magnetic field. We observed that C2C12 cell sheets gradually shrank and a reduction of 60% in size was observed when they were cultured for 2 d as floating sheets (data not shown). Therefore, the reduction in the pattern width was due to shrinking of the C2C12 multilayered sheet. Shrinking of tissue constructs may present a problem in 3D tissue fabrication without the use of scaffolds. To minimize the shrinking of tissue constructs, the introduction of an extracellular matrix such as collagen and gelatin may be necessary.
In tissue engineering field, cell patterning technologies may have potential applicability to fabricate tissue constructs with complex cell arrangements such as capillary blood vessels and neural networks. For the application of magnetic force-driven cell patterning, however, there are some limitations such as difficulty in long-term maintenance of cell patterns. This is because the magnetic force-based cell patterning technique described here could not control the direction of cell migration in contrast to the conventional methods involving lithography-based approaches. On the other hand, Wilhelm et al. (2007) demonstrated that the mobilization of magnetically labeled endothelial progenitor cells adhered to Matrigel could be induced by applying external magnetic field in vitro and in vivo, indicating that magnetic force can potentially control the position of magnetically labeled cells even after cells attached to the substrate. Consequently, we believe that the application of strong magnetic field may facilitate not only to pattern the cells on thick substrates, as discussed above, but also to maintain the stable cell patterns.
In this study, we have presented a simple and rapid cell patterning method to allocate target cells at the desired position on various surfaces using MCLs and a magnetic force. Moreover, when the magnetic force-based cell patterning was used in combination with the cell layering technique, a 3D tissue construct in which patterned endothelial cells were embedded was fabricated. These results suggest that the magnetic force-based cell patterning using MCLs is a promising approach to construct complex 3D tissue substitutes required for tissue engineering.
We thank Toda Kogyo Co. for supplying the magnetite. This work was supported in part by Grants-in-Aid for Scientific Research (nos. 19686049 and 20034043) from the Japan Society for the Promotion of Science (JSPS).