AVS2010 Session BI-TuM: Cells on Surfaces

This work relates to a method of patterning human neural stem cells as a suitable platform for performing different studies of cell proliferation, migration, and differentiation. The patterning of cells has been achieved by using microcontact printing (MCP) to create micropatterns of Poly-L-Lysine (PLL) and Fibronectin (FN), on substrates coated with cell repellent poly ethylene glycol (PEG) deposited by plasma. This substrate is particularly interesting since it is anti adhesive in liquid, but protein adhesive in the dry state. More details of the preparation method are described in [1]. Briefly, microstructured polydimethilsiloxane stamps have been fabricated by casting silicon masters produced by photolithography. The PDMS stamps have been inked with a FN or PLL solution, then dried with a nitrogen stream and put in conformal contact with the PEG substrate. By this method we are able to create fouling (PLL, FN) / antifouling (PEG) contrast on the surfaces where the cells are incubated and by modification of the spot size and distance, the influence of the cell environment on stem cells maintenance and fate studied.

Human Umbilical Cord Blood - Neural Stem Cells [2] were grown on the platforms with PLL or fibronectin pattern. After incubation for 4 days on PLL patterns consisting of 105µm squares spaced 300 µm, the cells are predominantly localized within the square. Such behavior is conditioned by agents added to the incubating medium. After being exposed to dBcAMP, the cells extend neuronal projections outside the squares, but cell bodies are patterned within the active domain. Immunocytochemistry was applied to trace neuronal lineage specific markers and their redistribution on the pattern domain upon influence of differentiating agents. It is found that the maintenance and fate of stem cells can be controlled by a combination of the protein type layer deposited by MCP and culture medium composition: presence of serum, neuromorphogenes and growth factors. Proper arrangement of soluble factors and bioactive surface domains allowed to work out conditions for developmental stage-specific immobilization of neural stem cells to the surface.

1. A. Ruiz, L. Buzanska, D. Gilliland, T. Sobanski, L. Ceriotti, S. Coecke, P. Colpo, F. Rossi. Micro-stamped surfaces for the patterned growth of neural stem cells. Biomaterials 29 (2008) 4766-4774.

2. Buzanska, L., Jurga, M., Stachowiak, E.K., Stachowiak, M.K., Domanska-Janik, K. Stem cells Dev., 15, 391–406, 2006.

Human umbilical cord blood originated neural stem cells (HUCB-NSC) are promising candidates for developing human cell based in vitro neurotoxicity assays. The origin of the cells makes possible to avoid species specificity extrapolation problems and avoid ethical issues compared to the embryonic stem cell research. Besides the easy cell line like culturing procedure of these cells they were shown to be able to differentiate to neuronal, asctrocytic and oligodendroglial phenotypes. This feature makes them also excellent subjects for developmental neurotoxicity studies.

The extracellular matrix compounds (ECM) used for anchoring the cells on culture dish or device surfaces may induce various ECM-dependent stem cell developmental responses mediated by integrin receptors. Investigation of stem cell - extracellular matrix interactions on biomolecule microarrays provides fundamental knowledge on neural differentiation as well as key inputs for stem cell based assay development.

The most powerful tools for microarray fabrication are microspotting and microcontact printing, offering flexibility regarding composition, concentration, spotted volumes or printed pattern geometry. Both methods are proven to be gentle enough to handle biomolecules for cell interaction studies.

In this work, the piezoelectric spotting and microcontact printing technologies were tested for rapid fabrication of biomolecule arrays on cell repellent surfaces.

Cell anchoring molecules, like extracellular matrix proteins (fibronectin and vitronectin) and poly-L-lysine (a polycationic polypeptide) were printed on anti-adhesive poly(ethylene) oxide-like films deposited by plasma-enhanced chemical vapor deposition. The microspotted and printed patterns were characterized by ellipsometry and microscopic techniques. The morphology of seeded stem cells and expression of certain differentiation marker proteins (GFAP, beta-tubulin III) were visualized by immunostaining and fluorescent microscopy, including also characterization after treatment with various concentrations of a known neurotoxic compound (MeHgCl). Toxicity results obtained on adhesive molecule spots were compared with results of a conventional neutral red uptake assay performed in coated 96 well plates.

Effect of the nature of biomolecules on cell behavior (including cell adhesion, morphology and survival) were investigated and are discussed focusing on generation long term cell-pattern stability, as well as stem cell differentiation into particular cell types.

The ability to both understand and direct cellular adhesion is vital in a variety of fields including bioanalysis, medical diagnostics and implant materials design. When a cell approaches a material, its ability to adhere and proliferate will depend greatly on the surface properties of the underlying material. One such material, poly(dimethylsiloxane) (PDMS), has many ideal bulk characteristics but its surface is hydrophobic and does not inherently promote cell growth. To remedy this, many surface treatments of PDMS, particularly plasma treatments, have been used to increase the adhesion strength and bioactivity of PDMS towards cells.

We have recently reported a novel method of patterning the bioactivity of PDMS [1]. We have further exploited the patterning by spatially-selective modification with various biologically active functionalities that permit cell patterning. A different, photolithographic technique has also been developed that exploits the hydrophobic recovery of PDM S, and while we were successful at patterning hydrophobic and hydrophilic areas using this method, cell growth was unfortunately not limited to only the hydrophilic regions on the photolithographically produced patterns.

To understand why the photolithographic method was unsuccessful, we looked both at the ability of a cell to proliferate on the different surfaces, as well as the adsorption of serum proteins, the presence and conformation of which directly affect the adherence of a cell to the surface. To better understand the cell selectivity or lack thereof, fibronectin was adsorbed on the patterned surfaces. A monoclonal antibody HFN7.1 was used to identify exposed integrin binding sites. The antibody was immunofluorescently labeled using a secondary FITC conjugated antibody. The difference in relative availability of binding sites was visualized using confocal microscopy and compared to the relative adsorption of fibronectin, and cell proliferation on the patterned surfaces. The availability of binding sites was shown to correlate with cell attachment on the stencil-masked patterned surfaces. However an increased availability of binding sites was seen on the hydrophobic recovered regions of the photolithographic patterned surfaces. We plan to directly measure the adhesion forces of chosen cells on the areas of “good” and “bad” conformation.

[1] N. Patrito, C. McCague, P. R. Norton, N. O. Petersen, Langmuir 2007; 23, 2, 715 - 719

Mechanical properties of substrates have been shown to be crucial factors for cell behavior, which includes the differentiation of stem cells or the malignant transformation into cancerogeneous cells (Engler et al. (2006) Cell 126, 677 and Cross et al. (2007) Nature Nanotechnol 2, 780). There have been published several techniques to measure contractile forces of cells, exerted onto the underlying substrate, since 1980. These approaches, however, only rely on compliant substrates and not on rigid ones as well known from load-bearing implants and culture dishes.

We present an approach to quantify the contractile forces of an ensemble of cells growing on rigid substrates based on nanomechanical cantilever sensors. In particular, we measured the relaxation of micro-cantilevers as the result of trypsin-release of about 100 fibroblasts. The optically measured change in cantilever bending, detected by means of the Cantisens Research system (Concentris GmbH, Basel), together with the number of fibroblasts counted was converted to the contractile cell force using the STONEY formula.

For the measurement of the contractile cell forces the selected cells were cultured on silicon cantilever arrays. The arrays consist of 8 micro-cantilevers each 500 µm long, 100 µm wide and 1 μm thin that enables us to detect stresses as small as 0.01 mN/m. Following adhesion and contractile force generation over night, the cantilever arrays were introduced into the Cantisens Research system to monitor the cantilever relaxations upon trypsin-mediated release of the cells from the substrate.

When rat2 fibroblasts are seeded on the silicon cantilevers they adhere and develop the morphology indistinguishable from that on standard culture dishes. Upon trypsin-induced release of the cells from the silicon substrate bending, the free end of the cantilever changes its position according to the following function D = 0.5D₀ (1-tanh(t/τ)), where D₀ corresponds to the deflection amplitude and τ the time constant of the cell release. While τ is apparatus and process specific, D₀ directly relates to the contractile forces of an ensemble of cells. The contractile force of an individual rat2 fibroblast on silicon corresponds to (17±7) µN. This value is reasonable, but high compared to the contractile forces of fibroblasts exerted on compliant substrates, a behavior expected from the studies on differently stiff compliant substrates. The contractile cell forces are strongly dependent on the state of the cell that explains the rather large error bar.

The method will support the fundamental understanding of cell-materials interactions with implications for cell-based biosensing and implant design.

Rare-earth metal nanoparticles are among the most promising candidates to be used as probes for visualization and targeted drug delivery. Compared to the ion-based gadolinium containing complexes used clinically as contrast agents today, gadolinium oxide (Gd₂O₃) nanoparticles show a considerably improved relaxivity and thus enable an increased resolution and an increased contrast enhancement. However, surface modification of these nanoparticles is essential in order to improve the biocompatibility and diminish any potential toxic effects. In the present study we have evaluated the impact of Gd₂O₃ nanoparticles (as synthesized, dialyzed, and functionalized with polyethyleneglycol, PEG) on the production of reactive oxygen species (ROS) from human neutrophils.

Gd₂O₃ nanoparticles were synthesized (via the polyol route), functionalized with PEG and characterized as previously described in Ahren et al.¹. Neutrophil granulocytes were isolated from heparinized whole blood using density gradient centrifugation. Generation of ROS by neutrophils upon addition of IgG-opsonized yeast, in presence and/or absence of as synthesized, dialyzed and functionalized Gd₂O₃ nanoparticles was studied with luminol-dependent chemiluminescence. In addition, the morphology of neutrophils after interaction with Gd₂O₃ nanoparticles was evaluated by fluorescence microscopy.

The ROS production from neutrophils challenged with IgG-opsonized yeast after exposure to as synthesized Gd₂O₃ nanoparticles was significantly decreased compared to control without nanoparticles. This indicates that the as synthesized nanoparticles are not well suited to be directly used in a living system without further modification. However, after dialysis and functionalization with PEG, no inhibitory effects were observed, possibly indicating that the high concentration of diethylene glycol (DEG) present in the as synthesized nanoparticle preparation is responsible for the inhibitory effects. Indeed, we in the present study also show that even a low concentration of DEG (0.3%) inhibits neutrophil ROS production. Our results indicate that dialyzed and PEG-functionalized Gd₂O₃ particles may be suitable in an in vivo situation as they do not impair the neutrophils capacity to produce ROS in response to a pray.

¹ Ahren M, Selegård L, Klasson A, Söderlund F, Abrikossova N, Skoglund C, Bengtsson T, Engström M, Käll P.O, and Uvdal K, Synthesis and characterization of PEGylated Gd2O3 nanoparticles for MRI contrast enhancement. Langmuir 2010, 26 (8) 5753-5762